Small and Quiet PCs

Motherboard Duct Experiment

2007-02-09T04:32:00.001+08:00

Will a motherboard duct better direct the airflow inside a small form factor PC to cool it's components more efficiently and quietly? A quick experiment to answer this question.

Cut from two pieces of thin cardboard.

Roughly what the duct looks like.

The duct height runs just above the ram and is wedged between the CPU heatsink and it's 92mm fan. The duct extends from the rear of the case (where it is sealed just above the 120mm case fan) towards the front of the motherboard where it is left open.

Top view.

Here the case side panels are installed, the duct is taped to the panel to form a seal.

Airflow is blown down into the duct by the CPU fan as well as drawn in from the opening at the front of the motherboard (left side in the photo) and drawn out through the rear case fan. The case side panels also has intake vents that coincedentally fall inside the duct height and will provide additional fresh air intake.

The duct fits closely to the 120mm case fan, but hopefully should not be overly restrictive for airflow.

Results to follow...

HP e-vectra : makeover

2006-11-08T00:23:00.005+08:00

The e-Vectra refinished with a fresh coat of paint.

This 6 year old PC was certainly showing signs of wear, including some scratches, scruffs, sticker stains, and areas where the color was beginning to fade.

I decided to repaint it's exterior and selected two cans of model spray paint. I selected glossy paint as I wanted a high gloss finish. First step was taking apart all the panels and washing them in detergent, then using masking tape cut-to-shape to seal off any areas I didn't want painted over (like the HP logo).

I painted the original cream colored panels silver, and for the dark blue panels I selected a slightly brigher royal blue.

The final finish was not quite as glossy as I hoped for, despite layering on the paint and waiting to dry between layers. Probably the rough surface of the case didn't help. However the case now looks close to new and is still shiny under bright lights.

Here is the finished case.

And the "before" look.

These photos were taken more than a year ago, but I just haven't had the chance to post them (it's now May 2008). The machine has since been given to a friend who needed a computer and has served faithfully as a web/email machine.

Back to HP e-Vectra : a closer look

Linux on the Wall

2006-08-21T02:37:00.000+08:00

My experiences building a small, low cost, low power consumption, diskless, linux data collection terminal left me with many thoughts to try and build a small silent linux PC. Certainly there are plenty of mini-ITX systems out there, but most tend to be somewhat pricey compared to say a regular desktop system. With the PC Chips VIA C3 800 mhz budget board, being only 5cm wider than a mini-ITX board, let's see what can be done.

This time around I found a Matsonic version of the PC Chips board. Besides the different color scheme (I actually prefer this traditional green to the red of the PC Chips board), and the Matsonic branding, everything else on the board is identical to the PC Chips M789CG. Even the contents of the manual is the same.

The Matsonic model number is MSCLE266.

CPU HEATSINK
The first step is to remove the noisy 40cm CPU fan and see if we can find a sutiable fanless heatsink replacement. I was able to remove the stock CPU heatsink fan without too much trouble. Beside the two plastic push pins, the heatsink is stuck to the CPU by a piece of double-sided thermal adhesive tape; wiggling/twisting and pulling up with my fingers did the job. I suspect removing the heatsink before turning on the system helps (the heat probably makes the tape stick even more), however please remember to take care as the CPU/board can be damaged and you may void your warranty.

The heatsink mounting holes are placed 60mm diagonally across from each other similar to northbridge heatsinks. The row of capacitors just next to the CPU also must be taken into account. Searching what was available in the local stores, I came across this Zalman, fan shaped northrbidge cooler, let's see how well it works.

The Zalman has two adjustable flanges that jut out from opposite corners of the heatsink, and had no problems fitting the mounting holes. Also the despite the wide fan-type fins, it is actually quite narrow along the other axis so there was no problem with the caps.

The base of the Zalman however was not as large as the CPU die and did not cover it completely, but there should be sufficient contact for proper heat transfer. Arctic Silver was used for the thermal compound.

With the motherboard placed in the open on top of a desk, an ambient of 29C, the temperatures were as follows:

STATE = CPU - MB - HDD
IDLE = 44C - 35C - 34C
PRIME95 = 73C - 37C - 36C

74C is a little bit on the hot side, however it should still be within the C3 CPU's operating range, and the system remained stable. However temperatures would get hotter enclosed in a small case, let's see if we can find a larger heatsink that will fit this board and provide better cooling.

I had an old Zalman fanless VGA heatsink with similar adjustable mounting flanges which also fit on the board. It has two large aluminum heatsinks that sandwich the VGA card and are connected by one heat pipe. The mounting block is shown on the CPU, the large heatsink plate then attaches to this with the heatpipe held in between.

Here is the heatsink plate, it has quite a large surface area and is more similar to the stock heatsink seen on fanless VIA mini-ITX boards. Unfortunately it extends over the RAM slots, and about 2 cm would need to be cut off the plate to make it fit. For testing though, I did manage to shift the plate a little to the left to clear the slots, using only one out of the two screws to affix the plate to the mounting block. The side view below probably is easier to understand. The proper mounting position for the plate would actually slide the plate left just covering both RAM slots. The way it is mounted here is not ideal as the contact surface between the mounting block and heatplate is reduced (due to the indentation for heatpipe and other ridges).

However when the CPU was running, the plate got hot enough to suggest there was sufficient area for heat transfer. Unfortunately the CPU temperatures were no better than the Zalman fan-type NB heatsink, and the motherboard temperature actually ran 1-2C hotter probably because the heatsink covered much of the board. Despite it's larger size this VGA heatsink did a worse job, maybe if the heatpipe was installed and the second plate also used it would run cooler. Also the heatpipe would allow the second plate to be connected to an alumnimum case for even better heat dissipation, but for now we'll go back to the fan-type NB heatsink.
Adding a system fan would probably be safer so I tried several different fans with the open board, all placed next to the CPU blowing horizontally across the board. The temperatures under Prime95 where all quite acceptable.

Papst 80mm @ 7v = 46C - 37C - 36C
AMD stock 70mm @ 5v = 45C - 35C - 35C
Noname 50mm @ 5v = 54C - 36C - 36C

COMPONENTS
I spent a few trips looking through the grocery store, stationery stores, as well as the local dime stores for containers of suitable size. I considered cookie tins, food gift boxes, tupperware type containers. Eventually I decided to try this plastic A4-sized paper tray first. It's a bit larger than the motherboard, and a little tall, but it's tranlucent blue material doesn't look too bad (plus matches the blue Zalman heatsink), it has built in vent holes, and the shape looked quite suitable to fit the system without much modification. If the system ran well, I could always try other smaller containers.
For mounting the motherboard, I cut a piece of black cardboard to the shape of the plastic tray. As the tray is both wider and deeper than the MB, I decided to push the MB all the way to the front so as much of the connectors/peripherals connected to the rear I/O port would remain within the case for a tidier look.
I used two nuts to space the motherboard up from the base board. An additional nut on the other side to secure the screw.

I just happened to have a stick of GEIL RAM with a blue heat sink, which nicely matches the Zalman. It even has a built in temperature gauge, now it might actually be useful with a transparent case.
I attached the 70mm stock AMD fan to the base board with double-sided tape.

Of course the Pico PSU is perfect for this system. It's a little out of focus, but it's the tiny yellow board in the lower left of the photo.

And the last component, a Toshiba 512MB flash drive for loading the OS and saving any files/settings.

The slightly lower front side of the plastic tray makes for a nice opening that allows access to the I/O panel and a path for the cables to exit. The tray is held onto the base board with twist ties at the four corners.

Here is the system with the cover on. I temporarily attached a hard drive so I could run Prime95 and monitor temps under Windows XP (unfortunately not yet familiar enough with Linux to setup a temperature monitoring utilitly like LM_Sensors). Under Prime95, the CPU temp was the same as without the cover, the motherboard temp was higher by 6C (41C). After being run for a while, the RAM, PicoPSU, NB heatsink all felt quite warm. Even the top of the case and bottom also felt warm. However once it was mounted on the wall, it seemed to run cooler. I was unable to measure the temperatures under Prime95 (as now running Puppy Linux), but no part of the case felt warm. With the vent holes now on top, this probably makes for much better airflow/heat exhaust compared with being placed flat (top totally enclosed).
Because of it's light weight, the system is easily mounted on the wall; I used two photo frame hooks.

BOOTING PUPPY LINUX

Installing Puppy Linux onto the USB flash drive was quite simple (this was described in Small System for a Small Budget ). The bootup time from the flash drive took about 1 minute and 20 seconds from the time the power switch was pressed.

Here it is driving my Dell 24" at 1600x1200 on the left (the right screen is connected to my main rig). Despite it's tiny 60MB size,Puppy Linux is a surprisingly usable OS (with the basic functions of browsing, email, chat, word processing, etc. and Open Office, Skype, among other programs can be added). Although certainly not fast, the app loading, screen refreshing times are reasonable.

Power consumption running Puppy Linux, 17-39 VA total power measured from the AC outlet. Assuming the AC/DC is working at a PFC of around 1, that would mean 17-39Watts.

One problem though with Puppy Linux is that it has no option to suspend the system, meaning you will need to power off the system instead. Also after shutdown, it does not automatically power off the PC either, you have to manually push the power switch which is a minor hassle. I suspect there should be some utililities/tools in Linux that would remedy this. Leaving this PC on all the time is not too bad with only 17W consumption, but still being able to suspend and wake from keyboard would be better.

And how about noise. Well without any hard drives, the system is certainly very quiet. The stock AMD 70mm fan is the only source of noise, and although very quiet when run at 5v, a very fain hum is still audible in a very quiet environment. As the system runs relatively cool, finding a quieter fan should not be too difficult to make this system practically inaudible.

SUMMARY

Total Cost:
MB+CPU = $55
512MB RAM = $50
PicoPSU + 60W AC/DC adaptor = $60
Toshiba 512MB USB Flash drive = $15 Zalman NB Heatsink = $10
Plastic Tray + Cardboard= $5
70mm fan = $5
Puppy Linux = $0
TOTAL $200

This makes a nice small and quiet budget system for running/learning linux and performing light computer work in near silence. Mounting it on the wall means no extra floor or desk space is needed, and depending on your tastes it can also make for a nice piece of techno-wall-art or simply an interesting conversation piece. In the current setup, the PicoPSU has an always on green LED, an additional red one when powered up, and the flash drive a blue LED when accessed; RAM with LEDs (like Ballistix Tracer) could also be added to show the true working of the PC, or otherwise cosmetic case lighting added if that's your fancy.

What's next, well actually the cables inside are still a little messy (the unused molex/floppy cables from the PicoPSU). Also there are quite a few cables running down from the system (VGA, lan, keyboard, mouse, power, sound, fan controller) and although it is hidden from view behind the LCD, it would be nice to have a cleaner look. Changing to USB wifi, wireless keyboard mouse (with small dongles), and connecting the fan directly to 5v will help. Possibly replacing the base board with a more permanent alumnimum base. I've never been much of a fan for LED lighting, but for this case I am somewhat tempted...

Also how about a linux box in a cookie tin, this one just fits the motherboard with a little room to spare...

Go to Small System for a Small Budget
Go to Tiny Motherboard with 5W Processor for $60

Small System for a Small Budget

2006-08-20T04:00:00.000+08:00

A budget system for under $260, monitor and operating system included!

In attempt to put together a small low cost data collection terminal, this little $260 system was born. This system makes use of the PC Chips M789CG budget motherboard we looked at earlier, with an onboard VIA C3 800 mhz processor and also a 8.4" LCD monitor to keep cost down. Certainly the low processing power and small screen means it won't fit most user's needs, but besides being a data collection terminal, it could work well for light computer use (email, chat, skype, some browsing) as well as fit easily in tight spaces.

SETUP
Installing it inside a cardboard box is actually not part of the cost saving solution (a proper PC case is included in the budget). It was meant to test if a small sized case would allow the system to run with sufficient cooling. Although it does make for an interesting case and even has a convenient carrying handle!

Here are the components setup inside the box. The motherboard is fixed to the box with twist ties, while the PSU and HDD are attached with screws.

The power supply is a generic mATX sized. It's one fan is the only exhaust vent in this cardboard box case.

A simply flap opening allows access to the motherboard I/O panel.

LCD DISPLAY
The screen is a tiny 8.4" LCD (the brandname is Super), the lowest cost display I could find with a native resolution of 800 x 600. Although not recommended for extensive use, for simple data entry or short periods of use it is acceptable. It also has a built in TV tuner as well as AV inputs, built in speaker, and remote control. The high gloss surface shows it is mainly for viewing video, but once it is turned on, it's not bad for PC use. The screen is bright and sharp at 800 x 600 (it supports higher resolutions by interpolation, but those resolutions are not very usable). Contrast is a little strong, but can be adjusted, and colors are decent. The screen is powered by a low wattage 12v AC/DC adaptor. Instead of using the adaptor, the LCD ran fine powered off the 12v from the PC mATX power supply by simply connecting the LCD power cable to a molex connector.

OPERATING SYSTEM
Windows XP Pro was only used for initial testing, certainly this OS alone would blow the budget. The system booted up rather quickly, despite it's low processor, memory, bus speeds. XP was only run briefly, but for simple file navigation and launching of monitoring utilities, there was no noticable difference compared with a modern day processor.

Here is a shot of the system running CPUZ and Speedfan.

Here is a shot of the BIOS hardware monitor screen. The CPU runs at 1.65v, unfortunately there are no undervolting options. The temperature and fan speed is also monitored, but again no PWM support to control the fans.

As for boot devices, the BIOS does allow for a plethora of options including USB removable media. Hmm... how about booting and running the OS off a USB flashdrive!? I searched the internet and came upon BartPE , which is a stripped down version of Windows XP that can be boot from a CD (and also USB flash drive) used mainly for system maintenance purposes. This is probably better explained in Wiki. I managed to get BartPE to boot off a USB flash drive on my Asus Pundit system, as the BIOS allows the USB drive to emulate a regular hard drive,

but on the Via system, my Toshiba USB flash drive was only recognized as a floppy which does not work with BartPE. So I had to abandon this option and search further for a USB flash drive bootable OS. Eventually I came across a string of linux distros including DSL (Damn Small Linux), Feather Linux, and Puppy Linux. Being unfamiliar with Linux, I have always been shy to try it. However I ended up deciding to try Puppy Linux as there were several reviews mentioning successful USB boots using the easy install wizard that came with the distro, perfect for someone unfamiliar with Linux. Also there seems to be quite an active Puppy community with a lively forum and continuing upgrades, much more so than the other small distros.

Additional benefits include that Puppy loads completely into RAM and runs off RAM without needing to read/write to the flash drive during the session; only upon exiting will Puppy save/write changes to the USB flash drive. This is advantageous as USB flash drives have limited write cycles - even though the number of cycles is quite large, constant writing will likely wear out the USB flash drive sooner. On the other hand this does mean if there is a sudden power cut, any work will be lost, but if the data is saved to a server then this is not an issue. In addition with the entire OS loaded into RAM, the system should run faster. Well some of you may think that means the system needs a lot of memory, but actually the entire OS is only 60MB and will run on a system with as little as 128MB RAM.

The installation steps are quite simple.
1) downloaded the .ISO file
2) burned it to a CD using Nero (other CD burner utitilities will also work)
3) booted the live-CD
4) followed the setup questions/prompts to get into the OS
5) ran the USB install wizard (USB flash drive was plugged in before bootup)
6) booted Puppy Linux from the USB flash drive!

I initially tried Chubby Puppy, which includes Open Office. This booted without problem from the CD, USB install script ran succesfully, but when booting from the USB flash drive, a list of errors appeared and Puppy did not start up. Instead I tried the official version 2.02 (without Open Office) and reran the newer Install wizard and this time everything worked fine. Booting from USB was really a piece-of-cake with this wizard, just simply answering a few questions and following a few prompts!!

ENCLOSURE
The system ran fine and temperatures were acceptable under load in the smaller cardboard box test setup. I tried to find a similar sized metallic budget case, but I could only find a larger book-type PC case.

Even this case is too large and there is a lot of extra unused space. Basically the front half of the case (right side in the photo) could be cut off as the system has no optical drive, and the hard drive could be squeezed above the PCI slots, but we won't attempt that here.

The power supply is a custom size, slightly smaller but longer than the mATX power supply. In addition to the PSU fan, the case also has a small 60cm fan next to where the HDD is installed. Don't be misled by the tiny keyboard, it's actually only 9" (or 23 cm) wide, and of course that's only a 8.4" LCD (1/4 the surface size of a 17"). The tiny keyboard was just handy for the install, a barcode reader will actually be used for most of the data input.

TEMPERATURE AND POWER CONSUMPTION
Using Speedfan to read the temperatures from the motherboard sensors, and a clamp meter to measure the VA power draw at the AC outlet:
This is inside the steel book-type case running Windows XP. These readings were taken inside an air-conditioned office, so I am guessing the ambient was around 24-25C.

STATE = POWER - CPU - MB - HDD
IDLE = 56VA - 38C - 30C - 26C
PRIME95 = 74VA - 57C - 35C - 27C

As the generic case power supply has no PFC the actual AC power consumption in watts should be significantly lower, maybe only 60% of the stated VA figures. In anycase the power consumption figures are very low, and remember this already INCLUDES the power consumed by the LCD as it is powered directly off the PC power supply. With the HDD directly next to the fan, it runs very cool.

FINAL CONFIGURATION
$55 = PC Chips M789CG Motherboard with VIA C3 800mhz CPU
$25 = 256MB DDR400 RAM
$55 = 40GB 2.5" ATA Notebook HDD
$100 = 8.4" LCD Monitor (800x600 native)
$35 = Generic Book-type PC case with PSU
$0 = Puppy Linux OS
--------------------------
$270 TOTAL

Ok, the title does read under $260, well a notebook HDD can easily be replaced by a regular 3.5" 80GB HDD which only costs $45 resulting in a $10 savings.

Alternatively the HDD can be eliminated altogether and replaced with a USB Flashdrive. The one I used is a Toshiba 512MB which cost $15, giving a total system cost of only $230.

How about noise, well although it is not a loud system, with the generic PSU fan, tiny 60mm case fan, and tiny heatsink fan, certainly it does not qualify as a quiet system either. However there is a lot of silencing, potential, especially with such a low power consumption system and diskless OS!

Back to Tiny Motherboard with 5W Processor for $60
Goto Linux on the Wall

HP e-Vectra : a closer look

2006-08-12T04:04:00.001+08:00

According to this CNet article, the e-Vectra began shipping around April 2000 and was targeted as a low-cost, easy to maintain corporate PC solution starting at $549. It seems one of the main reasons for the external power supply was for "easy support", making it easy to identify power supply failure and easy to replace.

The AC brick that came with my e-Vectra has a 19v 3.16A output, or approximately 60 watts.

Here is a view of the back including a standard collection of ports/connectors. The circular power jack is located on the very left. The key is for locking the hard disk door at the top of the case.

Sliding back the hard disk door reveals the 3.5" HDD inside. From HP's manual, I believe there should be a cage that installs around the HDD before being inserted, however this was missing in my machine.

There are two rubber grommets at the bottom of the hard disk bay, it seems that HP was paying attention to vibration reduction/silencing when they designed this machine.

The case cover slides forward revealing the single 80mm fan on the right.

Directly below the fan is the CPU heatsink, which is grounded to the chassis by a wire and two screws. The slim optical drive on the left slides out easily with the locking tab pushed down.

Removing the heatsink reveals the Mendocino Core Celeron 533 underneath. So this what thermal compound looks like after 6 years!

The single SDRAM slot is accesible from the front.

Here is a look at the motherboard after being removed from the chassis. It measures roughly 21 x 21cm, slightly smaller than mATX and a little bigger than mini-ITX. All caps and other components still look to be in good condition.

A sticker inside the chassis reveals the real age of this tiny machine, 6 years and 4 months.

A nice family portrait, all the components of the e-Vectra stripped down. The chassis is made from steel and actually makes the machine quite heavy for it's size, but this should also help with sound/vibration dampening. The plastic panels have been detached from the chassis for cleaning and repainting.

You can see here the original color of the plastic panel where I removed a sticker. Unfortunately the yellowing does not scrub off with detergent, so looks like a new coat of paint is in order.

Some closeups of the various components inclulding the CPU, RAM, fan, fan mount, hard drive and optical drive.

Overall the tiny PC is not very loud, but not quiet enough to qualify as a silent PC. However with it's heavy steel chassis, single fan, and relatively low power consumption, it does have good quieting potential. The old 10GB Quantum Fireball HDD does have a whine that can be heard even with the drive door closed, so that probably will need to be replaced. I am trying to get hold of some metallic paint to redo the panels before putting back together this tiny machine.

Go to HP e-vectra : makeover

Back to Tiny 6 Year Old External PSU PC

Tiny 6 Year Old External PSU PC

2006-08-06T14:20:00.002+08:00

While shopping for a small PC case, I stumbled upon this tiny little PC in the used-PC shop. It's a Hewlett-Packard e-Vectra machine, measuring 3.5" x 9.4" x 10.7" (9 x 24 x 27 cm). That's roughly the size of a telephone directory! Interesting thing is that it uses an external AC/DC power supply (like the Pundit) except this is a very old system. It's a Celeron 533Mhz with 128MB SD RAM, 10GB HDD, slim CD-ROM, and a Win98 license; so I'm guessing circa 2000, making it 6 years old**. I didn't realize they had PC's with external power supplies so long ago.

This photo probably gives a better idea of it's size. The LCD is actually a small 8" screen so that's somewhat misleading, but that pen should give a better idea of scale.
It seems that these machines were mainly sold to the corporate environment but now can be found floating around ebay and second hand swaps. I picked up this system for $75 including the external AC/DC brick. Normally I wouldn't even think of buying such an old processor, but it's tiny size, external PSU, and my recent tinkering with Linux (which runs so happily on such low powered machines) made this a perfect fixer-upper project.

The system booted up fine and surprisingly quick for such a dated machine. It had Win Me installed on it. There is one system fan directly under the vent holes, and the HDD is in the compartment on the left side. The main noise is the HDD whine (it's an old 3.5" 10GB) and the fan noise is audible but not obtrusive. We'll see what can be done to silence this tiny machine... The case has some stickers on it and is discolored, so it looks like a paint job will also be needed... Metallic blue/silver? Maybe something brighter?

Apparently people have been recycling these little machines for a while now; discussed here in the whirlpool forum. Technical datasheet for the E-Vectra (pdf).

** As for being old, when one thinks of a Celeron533 or even a Pentium III system, well to me that's pretty old and in my mind a long long time ago. But when I realized it was only 6 years, I was a little surprised. Certainly six years is no short time, but to think of the amazing progress in processing power over just these few years. What other items do you use that are six years or older? Maybe a refridgerator, toaster, jacket, desk lamp, home telephone, washing machine, DVD player...

Goto HP e-Vectra : a closer look

Tiny Motherboard with 5W Processor for $60

2006-08-06T14:07:00.001+08:00

This PCChips M789CG (v3.0A)motherboard comes with a low power onboard VIA C3 processor, and priced under $60 makes it perfect for a low cost DIY small and quiet PC. Measuring only 17 x 23cm, it's smaller than mATX and slightly larger than Mini-ITX (17 x 17cm). VIA has been putting their processors on Mini-ITX boards for a while now (and now even Nano-ITX at 12 x 12cm), but these are priced much higher, unlike this PCChips version that seems targeted to the budget user.

Although labelled as a "2000+" processor, it is actually a 866Mhz VIA Nemiah Samuel2 processor with a 133Mhz FSB. Unfortunately it's low power consumption (5W typical) also comes at a price. It's a slow performer rated below a Pentium III 500 Mhz making it unsuitable for any processor intensive tasks (ie. 3D, video encoding/decoding, number crunching, etc.). Some benchmark comparisons at bluesmoke.net, cpuid.com, tecchannel.de, and dansdata.com.
However for simple day-to-day tasks like browsing, email, chat, data entry, mp3 playback, or even a file server it's more than sufficient.

The board comes with a decent selection of I/O intefaces, including 2 IDE channels, 4+2 x USB 2.0 ports, parellel/serial ports, PS/2 mouse/keyboard ports, onboard VGA, onboard 10/100M LAN, and onboard 6 channel audio. It utilizes DDR266 RAM, however plugging in a higher speed RAM will also work fine (like DDR333 or DDR400 which is more readily available); the board will automatically set it to run at 266mhz. The BIOS is relatively modern and will support large disk sizes, boot from USB devices, boot from LAN, etc. It even includes two onboard temperature sensors, however no BIOS fan control nor via Speedfan. With CrystalCPUID, the multiplier can be lowered and the CPU speed changes accordingly, however the system locked up after a short while; voltage could not be changed.

There is one small fan on the CPU. Although it's not loud, it's not quiet enough for a "silent" PC. The fan is easily removed, but it's small size means a quiet replacement will probably be hard to come by. The heatsink is very tiny and not suitable for fanless operation. Unfortunately it seems to be glued to the CPU so removing it may require some solvent and extra care.

Here from the side, you can get a better idea of how thin the CPU heatsink is. From other VIA boards, it seems a pretty large heatsink would be necessary in order to run completely fanless, however a slightly larger heatsink might allow a larger and quieter system fan to provide sufficent coolilng.

With it's small size and low power consumption (ie low heat output) this makes installing a system into a large variety of custom enclosures possible (if you search the net, bread box, toaster, humidor, or maybe even a jewelry box, toolbox, cookie tin, etc.).

Specifications of this board at PCChips and customer reviews at newegg.

Goto Small System for a Small Budget
Goto Linux on the Wall

Touchscreens Update1

2006-07-03T23:52:00.001+08:00

For now, the 10.4" touchscreen is being tested on my Intel Mac Mini in my bedroom, and the 7" touchscreen on my HTPC as a second display/control interface out in the living room. I am still experimenting with both setups and they are very much WIP, but have made some interesting discoveries in the process.

10.4" TOUCHSCREEN
Previously I had my Mini attached to a 27" TV set in my bedroom which was fine for watching DVDs, Applestore videos, viewing photos, and even some light internet browsing. However anything that involved more substantial reading was a little hard on the eyes (text is never too sharp using composite video although OS X's browser does allows web page font text to be enlarged very easily).
Switching to the touchscreen did mean a trade-off as viewing video on a 10.4" LCD certainly cannot compare with a 27" CRT TV, however there were quite a few advantages to speak of. With the touchscreen on my lap, it became much easier to navigate using my finger than having to drag a mouse on my bedsheets. In fact using a touchscreen was so easy that even an 11 month-old child had no problem interacting with a children's program designed for a significantly older age group (with all the saliva being applied in the process, that also meant the screen needed a good cleaning afterwards).
Even typing on an onscreen keyboard worked quite well and could be done quickly. With the keyboard set wide enough, there is enough room to place both hands on it similar to touch typing on a real keyboard (the Apple remote is shown for scale). However because of it's smaller size and without being able to feel the keys, you do have to keep your eyes on the screen.

INPUT DEVICE OF THE FUTURE?
The keyboard shown above is Touchstrokes for OS X from Assistiveware.com is a little pricey at $59; actually Assistiveware has an even more expensive version for $299 called Keystrokes which offers additional features like predictive typing (typing just the first few letters will cause a row of common words/phrases to be displayed which can be selected simply by touching the desired word) and audio feedback of what is being typed (reading out the letter or word). This software is targeted at people with disabilities or injuries requiring alternate input methods, however such functionality might one day lead to a tool superior to the keyboard/mouse combo for normal PC use?

Let's try to imagine, instead of the keyboard and mouse you most likely have in front of you now, you have a flat lcd-like panel slightly longer than your current keyboard and twice as deep, angled up slightly towards you Let's call this the "Input Panel" (this is just an input device, your primary display is still the same screen you are using).
Using your finger you draw a circle on the dark Input Panel which causes your computer to turn on and the Input Panel to light up displaying an array of customized shortcut buttons for one-touch access to your common apps (think Optimus mini three) . Everything can be customized to your prefence, the layout and appearance of the background and the buttons depending on your fancy. The Windows task bar parks itself at the very top (or bottom) of the Input Panel.
Launch your favorite app by touching the appropriate button, or alternatively by using your finger to draw a symbol, maybe a $ to launch your accounting package, the outline of an envelope to launch Outlook, B for browser, whatever your preference (think Strokeit mouse gestures utility). As soon as your application launches, the Input panel image dynamically changes tailoring itself to the application.

For wordprocessing, the image of a keyboard appears on the bottom half of the panel. Again the look and layout customized to your taste, whether skinned, animated, colored, fonted, etc. (think Optimus keyboard). As you begin typing the very first characters, the most common words and phrases starting with these characters (based on your typing habits) will be displayed on the Input Panel just above the keyboard. Simply moving your finger up to touch the appropriate word will select it, or you may choose to ignore them for the shorter words (think Keystrokes). A Display Window on the upper half of the Input Panel will display the most recent lines of text as you type, basically mirroring the selected region on your main display. Need to delete the last word, simply draw a back arrow over the keyboard with your finger, or select certain text simply by dragging your finger over where it's displayed in the Display Window. Need to move the focus of the display window to a different area of the document, simply put your finger on the virtual navigation button and drag it in whichever direction you wish to scroll.

Worried about tactile/sensory feedback? There are a small row of tiny bumps where the ASDF JKL; keys appear, too small to obscure the display, but just big enough to feel so the fingers know where to rest for touch-typists. The surface of the Input Panel could also be produced with the right amount of resistance/play that pushing down would provide the necessary feedback.

For games, no need to remember all the different keys/commands or worry about hitting the wrong buttons at the worst times. Only gaming keys will be displayed, with the symbols, words, color-coding that allow them to be quickly and easily located (think Zboard gaming keyboard). Pertinent gaming information like score, ammo count, communication windows, etc would be displayed on the upper half of the Input Panel allowing more screen space for displaying your virtual environment (think Logitech gaming keyboard).

For image editing apps like Photoshop, all the toolbars are displayed on the lower half of the Input Panel. The top half would be a dedicated display window/touch tablet focusing on the area being edited (think Wacom Cintiq sans pen). Alternatively switch to album mode where a collection of photos appear scattered across the Input Panel. They can be arranged, rotated, resized, labeled simply by using the fingers (think Multi-touch Interaction Research demo).

I'm sure many people have already seen this demo by Jeff Han from NYU that has been out for about four months, but I only happened upon this video during my recent search for touchscreen interfaces. I was extremely impressed by this demo (don't forget to push play in the Quicktime window or alternatively view it on Youtube.com), it brought back images of Keanu Reeves in Johnny Mnemonic or more recently the UI in Minority Report.

Digging a little deeper, it seems many of the user interface ideas are actually not new and there have been researched/projects into this before. Edward Tse at University of Calgary studied applying a similar interface to Warfcraft as well as interacting with Google World as demonstrated in this Google video.
Andrew Wilson from Microsoft's Adaptive Systems and Interactions group demonstrates another earlier multi-touch technology TouchLight demo (wmv) as well as a very cool PlayAnywhere demo (wmv).
And last but not least, some novel application of touch technology currently available for the entertainment sector, www.fogscreen.com - demo (wmv) and www.i-bar.ch - demo. . . . ok getting back to my own touchscreens. . .

MINOR ANNOYANCES
The vga, usb, and power wires dangling from the screen are somewhat messy, but they never got in the way during use. Fingerprints and streak marks leaves the screen aesthetically unpleasant, but this is not noticable when the display is on. The flash slightly exaggerates the marks in the photo.

OS X vs WINDOWS
The touchscreen drivers supplied with my Lilliput screens do not support the Intel version of OS X and initially I was unable to get it to work. However thanks to www.mp3car.com forums I learned that a Beta version was available at www.egalax.com.tw, which I assume is the actual manufacturer of the touchscreen controller. The configuration options are significantly less than the Windows version, however it allowed the touchscreen to function without issue.

OS X normally has an easy-to-use user interface, but I will have to say it's not as touchscreen-friendly as Windows or maybe I'm just not familar enough with it to know where to make the necessary tweaks. Scrollbars, menu bars, and buttons tend to be quite small in OS X, and although still accessible with a finger, requires more precision and retries.
Windows on the other hand allows you to customize the appearance of the interface making things like scrollbars as wide as you like so even people with large fingers would have no problem.

Windows XP also comes with an onscreen keyboard, however it is a little small if you plan to type more than a few words. Here is the Jitbit Virtual Keyboard which can be purchased for $12 and can also be made transparent.

Screen-resolution-wise, the native 800x600 looked best both in OSX and XP, however 1024x768 was also acceptable; at this resolution text appeared a little clearer in OSX, possibly better font smoothing, but there were a few rogue lines in the lower right of the display not to be found under XP.

7" WIDE-TOUCHSCREEN

Using the second onboard VGA connector from my AOpen EZ18 SFF (nforce2 chipset), I had no problems adding a custom resolution of 800 x 480 (screen's native resolution). Using Nvidia's Nview, the system happily drove both displays, keeping my main plasma display at 1280x768. My thinking is to have all icons, menus, control bars, taskbars, etc. displayed on the 7" touchscreen, while the video/images would be displayed on the main plasma screen. Any adjustments, manipulations, settings could be performed on the touchscreen without disturbing the video on the main display.
One problem I discovered was that Windows XP was not made to run on a screen with 480 vertical resolution. Many of the configuration windows, popup dialog boxes, etc. were simply too tall that they extended off the top/bottom of the screen; unfortunately critical buttons like NEXT, APPLY, CANCEL, etc tend to be located at the bottom of such dialog boxes.
Setting up another custom resolution of 1000x600 solved this problem (this photo). It's not the native resolution, but all text/images looked just as nice even above the screen's native 800x480 resolution (unlike the 10.4"). Possibly because of it's smaller pixel-size or better anti-aliasing in the 7" model.
The application control panel does shrink respectively, and some of the smaller buttons can be more difficult to touch using a finger. Possibly getting some skins with larger sized buttons would help. Alternatively there are a lot of nice car front ends with big easy to access interface like Streetdeck or RoadRunner that could possibly be tailored for HTPC use.
I did have a lot of trouble getting the video window/control interface to automatically open up on the correct display upon launch (video window maximized on primary display and control interface on touchscreen); this was with PowerDVD and ComproTV. Further for PowerDVD, even with the video window maximized in the main display, it would tend to launch the video at a reduced size (assume it defaults to the resolution of the touchscreen where the application was launched). Certainly there must be HTPC applications with better multi-display support? MCE?

Otherwise if I don't find enough apps which support dual-display then there's the option to setup the touchscreen to clone the main display, but that wouldn't be as ideal. Any comments from users with dual-display/touchscreen experience is welcome.

Back to Small Touch Screen LCD

Small Touch Screen LCD

2006-06-28T00:50:00.000+08:00

How about a small touch-screen for the Mac Mini?
These touch-screens have been around for a while, traditionally used in car PCs with some creative modders adding them to their SFF PCs, and even being offered on some high-end HTPC cases like the Silverstone LC18. In the past these units were quite pricey, but recently they can be found on eBay and various online stores for around $200. The touchscreen interface makes it suitable for doing away with the mouse and its small size makes it an ideal display where space is tight; either acting as a second screen for a HTPC connected to a plasma/projector or as a screen for a second PC (media server, kitchen PC, etc).

There are many different brands and models on the market, I selected the Lilliput 7" widescreen model 629GL-70NP/C/T. It offered the highest native resolution 800 x 480 I could find for its size and also has an extensive set of touchscreen drivers including one for Mac OSX. As the manufacturer did not have any distributors in my country, they offered to ship to me directly for $209 plus a very reasonable $9 shipping cost. Actually the shipping cost was for two screens, I also purchased its larger 10.4" brother.

A front and back view of the 7" widescreen. The unit has a d-sub VGA input, 2 composite video inputs, and a stereo audio input. The touchscreen has a USB interface.

Its supplied accessories include connecting cables/dongles, stand with adhesive base, remote control, driver disk, and two power adaptors; one wall brick rated at input of 110-240v with an output of 12v 1500mA, and a cigarette lighter type car 12v adaptor with a small box which contains circuitry I believe to regulate the fluctuating car voltages.

And here is a look at the 10.4" model FA1042-NP/C/T. It's very similar to it's 7" brother, even uses the same wall power adaptor, however besides it's larger size, it has a 4:3 aspect ratio, higher native resolution of 800x600, a wall mount stand, and costs $40 more.

Hooking it up to a PC was straightforward and touch screen drivers installed easily under Windows XP. With the Nvidia drivers, a range of resolutions were offered (using a 6800GT graphics card) however for the 7" I was unable to add a custom resolution of 800x480. The closest was 840x480 or 720x480, but maybe Powerstrip would fare better. In anycase the standard resolutions 800x600 and 1024x768 worked fine, but of course the image was stretched on the widescreen; in order to keep the correct aspect ratio of 15:9, 1280x768 also worked but everything appeared very small.

Here you can see the 10.4" (4:3) set at 800x600 on the left, the 7" (15:9) set at 720x480 on the right, and a Dell 24" (15:9) set at 1920x1200 in the background.

TOUCHSCREEN INTERFACE
The touchscreen interface worked better than I expected. Navigating through Windows was no trouble whether using the supplied stylus or just my finger (although not recommended for people with large fingers). Even for writing it's not so bad (photo), however sufficient pressure must be applied otherwise the screen will falsely detect the pen/finger has been lifted up.
In addition to the 4 point and more detailed 25 point calibration, there are several settings in the driver that can be optimized depending on how you plan to use the screen including multi-monitor support. To access the right click, it can be set so that when the cursor is held stationary at one point (around 3-4 seconds) it will act as a right click. This works fine, except of course when you're scrolling through a large window and have the cursor held down on the scroll button for 3-4 seconds.

One problem that I found only on the 10.4" model was there is an approx 1.5mm deadzone on the left of the screen. There's a clear transition where the touch screen ends, it's as though it wasn't aligned properly during installation or the touchscreen is just too small for the LCD (click to enlarge the photo). This gap is not large and the driver can be set so the cursor automatically extends out to reach this area, but the problem is it won't be able to track your stylus if you run it along this edge.

IMAGE QUALITY
I already expected that touchscreens would have reduced brightness/contrast because the touchscreen element is not completely transparent, but was very disappointed with the 7" screen. The screen is very speckled, like there are two layers of low quality screen savers stuck on top of the LCD. Or kind of like someone forgot to peel of the layer of protective plastic that sometimes ship with screens. The 10.4" (and other LCDs to a lesser degree) also have some speckle, but it's barely noticable unlike the coarse and dense grain on the 7". This makes text blurry and difficult to read, and smooth tones look blotchy. Trying to capture the look of a display in a digital photo is always difficult, but these images give some rough ideas what I'm talking about (click photo to enlarge).

After searching on the internet, I found this forum posting on MP3car.com as well as on computeractive.co.uk, both mentioning this problem. It seems the manufacturer has added an anti-glare coating/layer? (these touchscreens are quite reflective which creates problems for car use under bright sunlight). It seems the older 619 model which has a more reflective, but sharper, clearer and brighter screen more suitable for indoor use.
That's really unfortunate as otherwise the screen image looks quite decent. Even set at 1280x768 characters appear smooth and legible on the 7" screen if you look past the speckle. The colors are pleasing to the eye and color balance can be tweaked by the monitors OSD.
For the viewing angle sweet spot, this is quite narrow; the color/contrast will change with a slight head movement even though the image remains visible through a decently wider range.

Here is the 7" next to a Phillips 7" LCD digital photo frame (no touchscreen) and provides comparable image quality. On images where there is no smooth tone, the speckling of the touchscreen is not that noticable.

And here are all three monitors. Certainly the Dell 24" still looks the best, but both touchscreens are certainly acceptable.
I actually spent have a day running office apps on the 10.4" and it was quite usable albeit the small size of 800x600. Unlike the 7" however, it seems there is no/poor anti-aliasing or smoothing as any resolution above it's native 800x600 looks bad.

CONTROLS
The LCD controls are located in a row across the bottom of the frame. On the 7", once the power is plugged in, the POWER switch lights up red. After the unit is turned on, this changes to green. The other buttons only light up when they are pressed. The OSD is relatively clear and usable as far as screen OSD goes. The left arrow provides a nice shortcut for AUTO ADJUST and the right arrow to step through various brightness settings (changing from day to night).

NOISE AND HEAT
Well the 10.4" does seem to have an internal fan, however it is very quiet and from the front can barely be heard. It runs warm, but not hot. The 7" however seems to run quite hot, both the front and back of the panel. In a 29C ambient room, the front of the panel got hot enough that it was uncomfortable keeping my finger on it for too long. It seems the 7" does not have a fan, so no noise can be heard when it's operating. However when the PC is shut down and the monitor goes into standby, a soft high pitched hum can be heard. This only goes away when a VGA signal is supplied again or the screen is turned off.
Both units do have built in speakers. Certainly not hi-fidelity, but not completely horrible either. The 7" speaker sounds similar to the Mac Mini's internal speaker and the 10.4" a little better. The 7" screen however will only pass through the sound when a video connection is selected and not when the PC/VGA connection is selected. The 10.4" passes through sound even in PC mode.

MAC OS X
The Mac Mini's DVI-Dsub adaptor had no problem working with these screens, however I was unable to get it to detect the touchscreen even after installing the OS X drivers. However when I boot up my Mini under Windows XP (exact same connections) the touchscreen works fine. I suspect the drivers were written for the G4 version of OSX and not compatible with the newer Intel version. I contacted the manufacturer and will see if I get any reply, in the meantime I'll try to see if it'll work on my G4 Mini.
For the 7" screen, my Intel Mini only gave three resolution options, 800x600, 1280x768 and 1280x1024.

FIRST THOUGHTS
Although the 7" is not ideal because of it's very speckled/grainy display, it could still be a decent interface for my HTPC. I originally thought of affixing this to the front of an Antec NSK2400 case, connected as a second display, and powering it off the PSUs 12v rail. This would eliminate the need for a mouse and allow for other operations to be performed without disturbing the video playing on the main screen. The high heat level has me a little concerned, instead maybe I will try mounting it on top. The high pitched sound in standby is also unfortunate, but as a second screen making use of the power button may not be so bad.
The 10.4" will probably be paired with the Mac Mini which is currently connected to my TV in the bedroom. Unfortunately the Mini doesn't support dual displays, although with the DVI-Video adaptor you can get both S-Video and Composite video out simulatenously (one to the TV and one to the 10" screen). Unfortunately Composite video quality is much worse than VGA/Dsub but I'll have to see if it's acceptable on the 10.4". If it does work out with my Mini, I'll probably look into painting the frame silver or white (although the glossy black and white combo on my new keyboard doesn't look too bad).

For those interested, there are more discussions regarding these touchscreens and some very nice modding/installation work on www.mp3car.com forums including a poster who installed a Mac Mini and a 7" touchscreen into his Mini Cooper.

Goto Touchscreens Update1

PSP : Small and Silent Surfing

2006-05-25T01:20:00.000+08:00

Sony added an official internet browser to their Play Station Portable (PSP) back in August last year. I've had mine since the very early days of it's first release, but never got around to updating my firmware until only recently. So how well does Sony's "Portable Entertainment Revolution" (aka PSP) perform for surfing the internet, well let's find out...

BEAUTIFUL SCREEN
The PSP actually has a very beautiful screen that is wonderful for viewing digital photos. The back light is bright, tones are natural, and color saturation is just right.

The 4.3" wide aspect 16:9 screen is actually a decent size and packs a high resolution of 480 x 272 pixels with 16.77 million colors. Those who have actually used a PSP know what I mean. Unfortunately my screen shots hardly do the PSP justice; you might notice the moire pattern appearing in many of the shots (like looking through a mesh/screen door), but this is not visible on the actual screen. It is very difficult to photograph lcd screens using a digital camera without moire (you can read more about this on Wiki).

ULTRA PORTABLE AND DECENT BATTERY LIFE
The unit is very portable and equipped with Wifi, allowing surfing from practically anywhere where you have wireless coverage. It's much easier to bring this to the local Starbucks than a laptop (by the way in a recent trip to London I was appalled at the lack of free wireless internet access - whether as a paying customer at Starbucks, or at the London Library, a four star hotel, or Heathrow airport, only PAID wireless connections were available!).
Although it does require two hands to hold/operate, it weighs only 280 g or less than 2/3 pound. The battery life is acceptable, lasting over three hours surfing the internet and downloading podcasts with screen brightness set at maximum. Fully recharging the 1800mAh Li-Ion battery took less than 2 hours.

SILENT
As there are no hard drives or fans, the only source of noise is the UMD (optical disc drive). Occassionaly the PSP will perform a seek if you have a UMD inside, but simply take it out and you can surf in absolute silence. Well ok, operating the keys/controls on the PSP makes a little bit of noise, but they don't click and can be pressed lightly.

USER INTERFACE
How the web page is fitted to the screen can be set in the Display Mode option. I found Just-Fit mode worked the best where "Page width is adjusted to match screen width."

Only needing to scroll vertically makes reading pages easier.
In Smart-Fit mode, "Page content is reordered so page width matches screen width." which also means a page can be streched out very long as it tries to stack up all the frames.

Navigating around a web page was actually quite comfortable and easy to get used to. The analog controller on the left (similar to the "eraser-head" pointers on IBM notebooks) moves the cursor around the screen. Simultaneously pressing down the (square) button on the right allows scrolling around the whole web page. Alternatively a four way cursor key on the left allows jumping around buttons/links on the page. Pressing the (circle) button on the right is the same as a left click on the mouse. One nice feature is if you hold down the button over a link, it will open the link in a new page (hey Internet Explorer should have such a handy shortcut as well, instead of the slower right-click "Open in New Window").
How about entering text? Well this is really where Sony drops the ball. Obviously the PSP is not meant as a text entry device, but certainly they could've come up with a much better interface. Instead they opted to copy the numeric keypad tap method used in old generation cell phones. So for example, GHI4 is grouped on one square, and JKL5 on the next. Pressing the GHI4 square once gives G, twice H, and so on. For those accustomed to sending a lot of SMS, this entry method could be usable, except there is no numeric keypad and you must use the cursor keys to scroll to the button you want to press (so if I just tapped D, I have to push DOWN-DOWN-LEFT-LEFT to reach PQRS7 and to select S click four times). There was a lot of talk of an add-on USB flip-down keyboard being developed by the company Logic3, unfortunately that was dropped with claims of lack of support from Sony. In anycase Sony could have included predictive text input (found on recent cell phones) or made better use of the analog controller and 10 plus buttons available on the PSP.
Certainly this rules out anything that requires significant amount of typing (writing/replying emails, posting to forums/blogs, chat/messenger, etc.) but it is still usable for entering addresses once (and adding them to the bookmark), account/passwords once (and using cookies to memorize them), or a quick search. Lucky there is the bookmark feature for easy calling up of a web page.

SPEED AND COMPATIBILITY
Unfortunately there are more limitations here. Web pages do take longer to load, especially pages that require a lot of bandwidth. On the most part a little patience means browsing news sites, blogs, forums, email, and even photo sites is quite acceptable.

If you've ever tried surfing the internet on a mobile phone, using GPRS or even 3G, that is excruciating slow. The page loading speed using the PSP is somewhere between that of 3G and a regular PC.

Although the PSP browser now supports Macromedia Flash content, there are still pages/content that it is not compatible with. I could not get MSN to load at all (but no problems with Hotmail), and on some sites the PSP came back complaining "There is not enough memory".

I guess PSPs 32MB system RAM is not sufficient, and it's not making good use of the memory stick. Of course any sites that require plugins or other software to be installed on your system won't work either. Hopefully with the constantly increasing size of the memory sticks (now more than 2GB available), future updates to the browser will help overcome some of these limitations.

RSS
The PSP does support RSS Channel. RSS or Real Simple Syndication allows for easy access to a summary of the content on a website (with RSS support) such as new sites, podcasts, blogs, etc. However it seems that the PSP is only supporting audio content at this time. Subscribing to an RSS feed is as simple as clicking on the website's RSS link and confirming PSP's prompt. Some sites that have audio feeds that worked with my PSP include CNN, Engadget, Tech Nation, Business Week, Nvidia Podcast, Insomnia Radio... More sites listed here on psprss.net and radio.xmlstoragesystem.com .

Here the PSP is playing back a podcast from CNN. After a while the screen will automatically turn off (to save batteries) while the audio will continue. An iPod might be a better solution for catching up with podcasts, especially if there are regular podcasts you follow (download by iTunes on a PC, dock your iPod for transferring, and listen at your leisure without worrying about broken/poor connection), but with PSPs wifi you have the option to browse and listen to new podcasts on the fly. The PSP built-in speaker is also quite decent if you don't want to use headphones.

CONCLUSION
Well the PSP isn't about to replace the PC as the main browsing tool. However being completely silent, extremely portable, and having acceptable battery life plus a very nice screen of a usable size, certainly makes it a nice complementary browser for reading up on the latest news, blog posts, forums, email, even perusing photo sites. It also works nicely for quick checking of information like the weather or market data, and keeping up with podcasts. All this can be conveniently done from practically anywhere (with wireless coverage); while having breakfast in the kitchen, lying down in bed (even without disturbing a sleeping spouse or significant other), sitting on the window sill, even someplace private (fill in your private place here) where one normally wouldn't have access to the internet. It's also easy to slip into a big pocket, small backpack, or purse if you want to take it outside where you'll have access to Wifi.

Unfortunately the poor text entry interface makes it impractical for any use where you need to type more than a handful of words (such as posting to a forum, replying email, filling in online forms, etc). Also the incompatibility with some web pages and limited memory, could prove annoying. The slower loading times on the most part is not so bad unless you visit sites requiring high bandwidth or need to frequently load new pages. Compared to anyone who's tried surfing on a cell phone or PDA, this certainly would be a much better experience (although some PDAs with larger screens, wifi, and a touch screen might not be so bad either). The PSP can't match a laptop's speed, compatibility, functionality, or keyboard, but then again the PSP is priced at a tiny fraction of the cost (under $200) and is more portable with similar battery life.

There are other small and portable "multimedia" machines that offer internet browsing capabilities like Nintendo's DualScreen or Gizmondo. Certainly would be interested to hear about other's experiences with these devices as small and silent surfers.

P2-AE2 : Optical Drive

2006-05-24T23:29:00.000+08:00

I've had my Pundit for about four months, and it's been happily running on my desktop all this time (and silently). Because of how I placed it, when the optical drive tray is open, the bottom faces me and I have to reach around to put in a CD/DVD. Flipping over the drive (mounting it upside down) would solve this problem. As I rarely use the optical drive, I didn't look into this possibility until today.

The optical drive cage and the bottom of the HDD cage are on one frame, which is held down to the bottom of the Pundit's case at three points. After removing the top of the HDD cage, this frame can be removed simply by sliding/pushing it towards the back of the case and lifting it up.

I had to first remove my larger 80mm CPU fan in order to slide the frame back, however with the stock 70mm fan it probably is not necessary.

Here is a closeup of the three points where the frame is held to the base. One on the right side (near the optical drive) and two on the left side (near the HDD).

The optical drive slides into the frame and is held in place with two screws on each side. The screws on the left side are actually recessed from the edge (this can be seen from the bottom).

In the back, there is also an interface board/plate attached to the drive with two screws. This converts the regular IDE plug and a FDD sized power plug into what I assume is the standard interface plug for slim optical drives.

Here you can see both sides of this interface board/plate.

And here is the Toshiba SD6112 slim DVD-RW that came with my Pundit.

As you may have guessed, the asymmetrical mounting holes (one side recessed) means the drive can only be mounted one way. So unfortunately mounting the optical drive upside down is not an option (and any drilling or frame deformation is not in my plans).

However I hope this take-apart helps poster Coconut who asked about the possibility to install an optical drive in a Pundit that comes without one. I believe the mounting holes, size, and interface should be pretty standard and therefore any standard slim optical drive should be fine. From the images of the NEC-6750A on Newegg.com, it looks nearly identical to my optical drive and I do not see any reason why it wouldn't fit.
However you do need to be sure the Pundit barebone you are getting does include the interface card/panel. This is attached directly to the optical drive, so if your Pundit comes without an optical drive this would have to be taped to the frame or packaged seperately in the box with the screws(assuming it is included).

One thing to note, the opening on the front panel does make for a snug fit for the optical drive tray. As the drive frame is only slid into place (and also there is some play in the front panel), you do need to carefully line up the frame. This is not difficult, but just may take some trial and error and may need some adjustment over time.

Pundit Bared
With the drive frame removed, I did take a few shots that show more parts of the motherboard I didn't see before.

Here is a better look at the mini-PCI slot at the left. I suppose it is possible to find a mini-PCI wireless adapter and plug it into the Pundit for an internal wifi solution. Also the BIOS chip can be seen in the lower left corner.

On the right we see the power switch connector (green and white wires) and also an AUX and CD AUDIO connectors. I do not believe slim optical drives have the CD AUDIO connectors so not sure what this could be used for. Maybe some extra audio-in connections if you decide to use the Pundit as a recording station?

And since I have the Pundit open, a quick check to see how the foam/cool pack/notebook HDD/pencil case enclosure is holding up. There have been one or two users who reported finding condensation in such a setup, but fortunately I found no such signs. All felt dry and everything pretty much looked the same as four months ago.

Only the foam showed an impression of where it was pressed against the HDD and also seemed slightly whiter at the same spot. Possibly due to the extended exposure to the hot surface of the HDD, but it didn't seem to be anything of alarm.

Back to Overview

Back to Stealthed and Silenced

AOpen XC Cube hushed

2006-05-05T00:43:00.000+08:00

After having more time to tweak the cooling of my AOpen XC Cube with the PicoPSU, I am glad to report back I now have a much quieter SFF than before. I would like to give some background on the original airflow setup of this SFF which will give readers a better understanding of the changes I made.
On the left side panel towards the front of the case is a large intake vent. The HDD is located behind this vent and in my original setup I added a Zalman 80x15mm fan to increase the incoming airflow.

On the right side panel towards the rear of the case is a smaller exhaust vent. The CPU heatsink is located directly behind this vent and on the other side of the heatsink is a fan pushing air out in this direction.

Here is a close-up of the CoolerMate Ice Cube CPU heatsink I used to replace the stock cooler. The dense packing of fins is not ideal for exhausting hot air, however it did a better job than the stock heatsink at keeping the CPU cool.

The CPU fan I had installed on this heatsink is also a Zalman 80x15 fan.

And here is a closeup of the nice Athlon XP Mobile chip. The days of AMD's fragile and exposed cores (now the majority of newer Athlon 64 chips are covered with a stronger heat spreader) which actually allowed these hotter chips to transfer heat directly to the heatsink more effectively. This is actually an XPM 2200+ and not 2400+ which I mistakenly stated in my last post (it's been a long while since I worked inside this system).

The original PSU also pulled air through it's restrictive vents and exhausted air out the back of the case using an 80mm fan. So in summary, we had one large intake vent towards the front left, and two somewhat restricted exhaust vents, one through the CPU cooler at the right back, and the second through the PSU at the rear of the case.

CHANGING THE AIRFLOW
With the removal of the stock PSU and installation of the PicoPSU, a large opening in the rear panel was freed up for exhaust. Reusing the same 80mm fan, it now had a direct airflow path and the volume of air it could exhaust quietly increased. Therefore my thinking was the CPU vent could now be used instead as an intake instead which should help keep the CPU run cooler.
As the CPU fan was not located directly next to the vent, but had to draw air through the heatsink, I took some electricians tape to seal up the gaps around the fan mounting frame and the heatsink in order to insure most of the air would be sucked in from the vent instead of the sides of the cooler.

The intention is for the fan at the left of this photo to pull in air through the heatsink (from the right side of the photo where the vent would be located) allowing it to better cool the CPU. The hot air would then be exhausted by the rear case fan.

Oddly there seemed to be no temperature difference compared with the before when the CPU vent was used as an exhaust. However fiddling with the fan settings I found that completely turning off the intake fan at the left side of the case actually improved temperatures. It occured to me that the using both fans as intake was actually causing the airflow to work against each other. Even with only two fans, I was even able to run the rear fan at a slower speeds while keeping the system equally well cooled. Needless to say this allowed the system to be quieter than before.
I decided to add a duct around the CPU cooler to allow it to make full use of the intake vent. I fabricated this with 2mm thick paper board. Although flammable material is probably not the best choice for putting inside a PC, I don't expect the temperatures in this system to get high enough where it will pose a danger... but we'll see. The paper board is actually very easy to work with and holds it's shape nicely. A ruler and exacto knife allows it to be easily cut to the exact size, and a light etching with the knife along the seams allows it be bent. The end result actually looks very nice for something completed in such a short time.
I decided to use the same material to cover up the opening in the back panel. Cutting a circular whole for the fan was a little more difficult, here you see a misshapen hexagon. But from the outside it looks fine and it is rigid enough to serve it's purpose.
The fan is still mounted on the exterior of the case, but now with the extra room inside the case I could always move it back inside. In anycase to keep for consistent testing I'll leave it on the outside for now. The vent didn't do wonders, but did the allow the CPU to run slightly cooler by 1C.
I decided to add a small 40mm Sunon Mag-lev fan to see if it would help. Normally small fans tend to be screamers and do not move a lot of air, however when volted down this Sunon is very quiet and moves just enough air to help cool down hot components like a NB heatsink.

Unfortunately the short wire of this fan meant I had to plug it into the CPU fan header and move the CPU fan to the chassis fan header. On this motherboard these fan headers are run by different chips so their PWM settings are different (ie 15% CPU header does not equal 15% CHASSIS header). So I had to readjust fan settings for the CPU fan as well and cannot compare exact same settings. However this fan did help with system cooling and running at 9% setting brought down system temperatures by 1-2C compared with it being turned off.

FINISHING TOUCHES
For the power input jack of the PicoPSU, I decided to mount it where the Coaxial SPDIF out connector was located. The circular coax connector was slightly larger, but let the power jack fit securely once the exterior nut is screwed on. You can see it to the right of the parellel port (just below the fan) in the photo. I do need SPDIF audio for my HTPC, but I figured I could always use the optical SPDIF connector instead.
Then I remembered oddly the optical connector at the back of the AOpen is in fact audio in, the audio out optical jack is located at the front of the case, too ugly to have a wire dangling from there. Fortunately I have no use for the parellel port so off that came and the coax connector happily relocated there. In addition I sealed up the tiny vent holes with blue electrical tape. Sealing these holes actually made a significant improvement to the overall cooling, both CPU and system temperatures dropped by 1C and HDD temps dropped by 3C. I suspect these holes served as fresh air intake for the original PSU, but with the modified air flow these holes become unnecessary as there are no components it will cool on the way to the exhaust fan. In fact they reduce the amount of air being drawn in through the other two intake side vents. The final settings and temperatures:

Fan settings using Speedfan
LOW FAN - CPU 3% - Chipset (40mm) 5% - Exhaust (80mm) 9%
HIGH FAN - CPU 5% - Chipset (40mm) 9% - Exhaust (80mm) 15%

Temperatures based on an ambient of 28C
IDLE - LOW - CPU 42 - SYS 44 - HDD 34
PRIME95 - LOW - CPU 50 - SYS 50 - HDD 35
PRIME95 - HIGH - CPU 45 - SYS 45 - HDD 34

Noise-wise the system is quiet and it produces a soft smooth low frequency whoosh/hum that is quite pleasant to my ears. It's not inaudible like the Pundit, and even louder than the Mac Mini, but the noise level is still very low and noticably quieter than before. Installed back on the compenents rack above my TV screen, in the very quiet ambient in the wee hours of the night, it becomes inaudible beyond 6 feet. During the day or when my TV screen is on, the system cannot be heard even within two feet.

Probably a more open CPU heatsink that allows for better airflow as well as running a slower 92mm fan for exhaust will allow the system to cool even better and become even more quiet, however for it's purpose as an HTPC it is more than quiet enough already. One problem though that I did find after I had it all setup again as my HTPC is that the system has trouble powering up from suspend mode. I experienced this once while testing the system on my workbench, but subsequent trials proved no problems. I suspect that the PicoPSU is the cause here as I did not have this problem before. Although noise and temperature wise it is perfectly acceptable if I leave it on permanently, I am not happy with the idle power draw of around 50W as it would be like leaving a lightbulb on all the time that you weren't using (unfortunately Athlon XPs have poor power management at idle). I will have to see what can be done about this.

Back to PicoPSU Installed

Power Factor

2006-05-04T23:16:00.000+08:00

How much power a PC system draws from the AC outlet is often a good indication of how much heat the system dissipates and consequently how much work is needed to keep the system cool or for silent PC enthusisasts, gives us a rough idea how quietly it can be run. In all my previous posts, I have been measuring the amperage of the AC current being drawn by the system and multiplying it by the AC voltage to calculate the power consumed in watts, however this method is actually not entirely correct. For a DC circuit, Power=Voltage x Current, however for AC circuits with a reactive component there are other factors that must be considered.

In fact what I have been measuring is "Apparent Power" which has the units VA (volt-amperes). Apparent Power is made up of two components, "True Power" and "Reactive Power". "True Power" also called "Real Power" is the actual amount of power being dissipated or consumed by the system and is measured in Watts. This is the figure that I thought I was measuring. "Reactive Power" is the additional power being drawn by the reactive components of the system but not actually dissipated or consumed by the system. The "Reactive Power" affects the phase of the AC current but is returned and absorbed back to the AC power line. A more detailed and exact explaination can be found in this chapter of Lessons in Electric Circuits as well as in this SPCR article.

Power Factor defines the ratio between True Power and Apparent Power (PF = True Power / Apparent Power). For a purely resistive circuit (one without any reactance) the PF = 1, and consequently Apparent Power = True Power. However most generic PC power supplies do not have any power factor correction (pfc), and their PF is probably below 0.60 meaning if the system is dissipating a True Power of 60 Watts, the Apparent Power draw would be measured at 100VA or more.

More and more high-end PC power supplies are now being offered with pfc to improve PF, and also there are regulations in the EU requiring power factor correction. There are two types of pfc, Active and Passive. Active pfc often yields a PF very close to 1.0 whereas Passive pfc offers PF in the 0.6-0.8 range.

For all the external brick AC/DC adaptors I used, these are universal input voltage, switching power supplies which do have active pfc and therefore their PF is probably in the range of 0.95-0.99. So for all measurements given for the ASUS Pundit, PW200M, PicoPSU, and Mac Mini systems the True Power being consumed is slightly less than the Apparent Power figures I posted but should be within 5%. The stock Antec Aria PSU is also similarly equipped with active PFC. The AOpen XC Cube stock PSU is marked "with pfc" but I assume passive. This helps explain for the surprisingly large difference in Apparent Power draw compared with the PicoPSU. The True Power consumed with the stock PSU should be 80% or less of the measured Apparent Power, which brings it closer to the power draw of the PicoPSU. For the Tagan PSU however I am not clear whether or not it has any pfc. It does have universal input voltage, a common characteristic of active pfc, however it's specs are not marked with any indication of pfc and the measured Apparent Power is significantly higher than the Pico/PW200M psus suggesting that it does not have pfc.

Thanks to Linus and Devon at SPCR for setting me straight.

PicoPSU Installed

2006-04-29T03:00:00.000+08:00

It's been a while, but finally I had the chance to install my PicoPSU into the AOpen EZ18 XC Cube SFF. The EZ18 is an old model, but the case has a nice clean and elegant glossy white finish (also comes in black) which I like very much. AOpen continues to use the same chassis in their latest XC Cube models even after several years. This was my very first SFF and also my first SFF silencing project. I bought this machine nearly two years ago, and also about the same time I joined SPCR where I got a lot of helpful tips. The original thread discussing silencing options can still be found here as well as SPCR's review of a very similar model EZ65. Although after modding this system was quiet enough to be an HTPC, it wasn't quiet enough for me to put on my desk. I always felt the restrictive airflow to be a main culprit and if I could somehow remove/relocate the PSU it could be improved. Well now was the chance to see...

SYSTEM COMPONENTS
The EZ18 comes with an Nforce2 mothboard that not only had decent onboard graphics (for it's time), but also the Soundstorm chip that encodes any PC audio (regular stereo) into digital audio (SPDIF 5.1 signal) in real time. This was perfect as an HTPC as I had the audio connected to my home digital receiver with a 5.1 speaker setup, and allowed DVD audio streams as well as all PC audio (TV signal, MP3s, Multimedia presentations, etc) to be piped through the same digital connection.

The system components include:
- Athlon XP-Mobile 2400+ CPU running at 10x166 1.2v
- CoolerMate Ice Cube CPU Cooler w/ 80mm fan
- 2x512MB RAM
- Samsung 40GB Notebook HDD
- Samsung Combo Optical Drive
- Compro PCI TV Tuner
- 220W AOpen PSU (w/ PFC) modded with Adda 80mm fan
- Zalman 80x15mm intake fan

Here's a look at the system before putting in the PicoPSU.
I previously modded the original PSU, putting a quieter fan onto the outside.

With the design of this chassis, hot air is drawn out through the CPU cooler to the vent on the right side of the case (facing you in this photo). The stock cooler was a flimsly aluminum heatsink, and trying to find a suitable sized replacement heatsink that blew sideways in the right direction was very difficult. I was fortunate to find the CoolerMate Ice Cube that was copper based and even had heat pipes (not so common back then). I actually had to ask someone to help me buy this from Germany. It was a little large (the stock cooler only used a 70mm fan) and I had to cut part of the drive cage to get it to fit. Also the tight spacing of fins are not ideal for exhausting air via this channel, but fortunately the copper and heatpipes do a decent job of keeping the CPU cool.

The already quiet 2.5" notebook HDD is suspended inside the drive cage.

A PCI TV tuner card is installed. The fan on the right is a Zalman 80x15mm fan used to draw cool air into the case from the intake vent on the left side of the chassis.

In this top view, you can see the stock PSU mounted at the back of the chassis.

PSU SWAP

Here you see the tiny PicoPSU on the right next to the stock PSU. Although the stock PSU is already quite small (even smaller than the Aria's), the Pico still makes it look like a giant.

There are only intake grills around the four edges and not the back of the stock PSU; this restricts the airflow plath for exhausting hot air out of the case via the PSU's 80mm fan. Again we hope the Pico will give the case better airflow, less heat to remove, and better power efficiency.

There are three tall capacitors located right next to the motherboards' ATX power connector preventing the PicoPSU to be plugged directly onto the board.

However a short extension cable works fine. The EZ18 board also requires a P4 motherboard connector which is not included on the PicoPSU. I used a molex Y-splitter coupled with a molex-to-P4 adaptor for this. I will probably solder a P4 connector directly to the PicoPSU later to reduce the amount of cables/adaptors.

Here from the back we see the hole where the old PSU used to be. Yep that's the tiny Pico suspended inside. It's actually tied in place to the back of the chassis with a twist-tie. The Pico's 12V input plug unfortunately did not fit onto the PCI bracket that came with it, but we'll figure out a more permanent mounting solution later.

I decided to put back the same 80mm Adda fan from the original PSU.

And just some temporary tape to cover up the gaps so that we get proper airflow from the original intake vents.

POWER AND HEAT
So how did the Pico perform? Well in terms of power efficiency I was amazed. With the stock PSU the AC power draw measured from the power outlet was 70W idle and 85W under Prime95. With the PicoPSU, idle was reduced to 52W and under Prime95 65W. That's a 25% difference or a reduction of 18-20W; a much bigger difference than we saw with the Antec Aria PSU and also the Tagan ATX PSU. I guess the efficiency of the stock AOpen PSU is pretty poor, at least at these low wattages. If we assume 75% efficiency for the Pico, then the stock PSU would be at a horrible 57%.

Note: I later discovered in my measurements I was actually calculating Apparent Power and did not take into consideration Power Factor.

However for temperatures the difference was not as big as what I expected. The three fans in the system are controlled via PWM using Speedfan and settings kept the same.
- CPU 80mm fan = 15% (approx 1700rpm)
- Zalman 80x15mm intake fan = 3% (no rpm reading)
- PSU 80mm fan = 20% (no rpm reading)

Temperatures with the stock PSU under IDLE/PRIME95 and room ambient of 27C.
- CPU = 43 / 48
- SYS = 45 / 48
- HDD = 35 / 34

Temperatures with PicoPSU.
- CPU = 41 / 46
- SYS = 43 / 47
- HDD = 34 / 34

The CPU/SYS temps were lowered by only roughly 2C, and the HDD temp remained pretty much the same. I would have thought with the 20W reduction, plus the additional heat that has been moved out of the case to the external AC/DC adaptor, that the system would run much cooler.
However the airflow setup is probably still optimized for the original configuation with the stock PSU, and not taking advantage of the now excellent exhaust via the rear fan. I can probably replace the CPU cooler and use the CPU vent as an intake instead. With the Pico now there are many new options to rework the airflow in this SFF to perform quietly. I will have to take some more time to experiment and will post back...

Goto AOpen XC Cube hushed

Back to Pico Psu and PW200m DC-to-DC Power Supplies

A8NVM-CSM Aria Update

2006-04-29T01:46:00.000+08:00

MULTIMEDIA BRACKETS

Finally received the SPDIF audio and video connector brackets for my Asus A8NVM-CSM motherboard. It's a real mystery why Asus didn't bother including these brackets with their "multimedia" motherboard, as without these the digital audio and video out functions of this motherboard become unusable. The local Asus dealer does not carry these accessories either, go figure. Fortunately I had some friends travelling to Taiwan and was able to pick these up. Yes having friends travelling to many countries is always helpful to source hard-to-find PC components, especially uncommon silencing parts, but I believe most people have trouble understanding the silent PC enthusisat. Everytime someone helps bring back a component they ask "What is this for? You mean it makes your computer run more quietly?!?" and they have this funny look of disbelief as if you might be pulling their leg.

BIOS UPDATE = LOWER TEMPS
Well I originally purchased my Asus A8NVM-CSM with the Aria as a replacement for my dated HTPC, and now with these brackets I could finally get started. I took out the 7800GT high-end graphics card as I wouldn't be playing any games on my HTPC. Now with only one 92mm CPU fan and one 120mm case fan (both running at lowered speeds with Speedfan), the system was pretty quiet and perfectly suitable for HTPC use. Regarding CPU temperatures, thanks to Steerpike and Ersa's post I discovered that the older BIOS for this board was reporting a significantly higher CPU temperature. I updated my motherboard BIOS from version 0601 to 0702 and discovered this was indeed the case. The CPU and "NB?" readings immediately dropped 8C under the exact same conditions. I compared this both at idle and under load with the same results. The MB temperature was unchanged. Although these readings are only approximate temperatures to begin with, the fact Asus adjusted this in their newer BIOS leads me to believe the newer readings should be more accurate. In the back of my mind I was always wondering why my Opteron was idling so hot even when undervolted to 0.8v 1Ghz.

A NEW DVD BURNER
I recently picked up a LG GSA-H10N DVD burner for under $40. Prices of DVD burners have come down significantly, I guess in anticipation of Blu-ray and HD-DVD. It's a fast 16x DVD burner (10x DL+R, 8X RW+R) and also has the ability to write DVD-RAM at 12x. Originally I thought of putting this in my main rig to backup data, as my thinking is DVD-RAM format will be more reliable for data. However this optical drive is perfect for the Aria, and much better than the ASUS DVD-ROM I had previously installed.
In addition to being shorter in length (allowing more space in the tight SFF case), the eject button matches the Aria's front panel button perfectly (for the Asus I had to tape on some extra padding to get good contact). The eject mechanism is also much more polite than the Asus, there is a very short pause after eject is pushed before the tray comes out smoothly, unlike the Asus tray which will shoot out immediately and roughly at high-speed; I am always startled by this and have to make sure I get my hand out of the way quickly. The LG also works nicely with Nero Drive Speed which allows the spin speed to be slowed down (ie made quiet) when playing back DVDs and CDs.

MORE PHOTOS OF THE CPU COOLER
No further changes were made here, just a few close-up shots of the Thermaltake Silent Pipe CPU heatsink and Silverstone 92mm fan attached with twist-ties.

Here you can see the fan is actually elevated from the heatsink by a tiny white foam washer (taken from a pack of CDRs).

HARD DRIVE SEEK NOISE
The fan noise level is quite acceptable as a HTPC, however the hard drive seek noise is very loud. I am using a Hitachi T7K250 250GB 7200rpm/8MB cache, two-platter SATA2 HDD. A notebook drive would probably make a better choice, but I was hoping the Aria case would allow me to take advantage of larger capacity of a regular 3.5" drive. There should be just enough room to suspend the HDD which should reduce the seek noise, also turning on AAM (Automatic Acoustic Management) should help with some minor trade-off in performance. However why not give my recently acquired SmartDrive 2002C enclosure a try.
The drive cage of the Aria has three places for mounting a 3.5" HDD. The main spot is oriented horizontally, directly under the optical drive. The second and third are mounted vertically at the sides of the of optical drive. The side mounts are too narrow to fit the Smartdrive 5.25" enclosure, however it could potentially fit in the mount under the optical drive. One side of the cage's two mounting plate needs to be moved, but that can be easily bent down and out of the way.

Unfortunately, the vertical clearance is not enough to fit the Smartdrive unless the CPU fan is removed. Bluefront known at SPCR for his extensive and elaborate silent PC modding actually had a similar idea of using a PW200M inside an Aria case which he details in this post. He works off positive airflow (using the rear 12omm fan to draw in air instead of exhaust) and ducts the air to a fanless XP120 CPU heatsink. His duct idea is very nice, and even manages to mount a notebook hard drive a copper plates as part of the duct wall. I originally thought of using my XP120 heatsink as well, unfortunately the CPU socket on the A8NVM-CSM is too close to the edge and would not fit such a wide heatsink. With some clever ducting, it might still be possible to run a fanless CPU heatsink, but I think I will give such a project a miss for now as the Smartdrive enclosure was intended for my larger main rig.

AN INITIAL TEST OF THE SMARTDRIVE
But let's see how well the Smartdrive performs in anycase. As it will not fit inside the Aria case, I will have to test it outside.
After removing the bottom plate of the enclosure, the HDD slides in between two copper plates that run along the sides of the enclosure. The foam padding pushes against these plates so that it is a very tight fit for the HDD; I used a piece of cardboard to help wedge/slide in the HDD (seen in the photo here). The SATA and power connectors were easily attached and run out the rubber lined back opening.

Oops, I forgot I was using the PW200M which doesn't have a SATA power cable, and I did not want to add a molex-to-sata power adaptor to keep cable mess to a minimum. Hitachi HDD conveniently come with both SATA power as well as molex power connectors allowing the drive to be powered by either connector. Fortunately removing and reinstalling the bottom plate and it's six screws was no big deal.

Oops again. How silly; for those observant readers you'll notice I disconnected the SATA data connector instead of the SATA power connector! When the large plug didn't match up to the SATA cable from the motherboard I actually had to pause and think why... it's getting late.
Ok opening and reclosing the bottom plate one more time, finally third time is a charm.

Here it is sitting on top of the Aria. First temperature-wise, the drive idled around 40C with an ambient room temperature of 27C. This is very similar to the 39C temperature of the drive when installed inside the Aria case. We should keep in mind that although the air is warmer inside the case, there is also some airflow whereas sitting outside the case there is basically none. Running Sisoft's HDD bench continuously maxed out the drive temperature at respectable 43C and as soon as the bench was stopped, drive temps dropped immediately; this suggests that the Smartdrive was effectively transferring heat away from the HDD. The Smartdrive was also warm to the touch.

How about noise? Well the Smartdrive certainly did it's job of dampening the seek noise, but unfortunately it was still clearly audible. Even if mounted inside the case, I believe the seek noises would still be audible although the sound should be quite faint.
I put the Smartdrive on a folded piece of soft packing foam to make sure vibrations from the enclosure weren't being resonated onto the Aria. The seek noise did sound very slightly fainter, but not a significant difference. I could clearly feel the vibrations of the HDD, but with the foam and inertia from the heavy Smartdrive, this must have been dampened (the Hitachi drive does tend to vibrate quite a bit despite being only two platters).

In terms of the spinning noise, this was audible but a smooth tone that was not very loud. At around four feet, I wasn't able to make out the sound anymore (all fans turned off in a very quiet room). I am pretty sure inside a computer case this level of noise would be practically inaudible from more than two feet away. However I did notice if I placed my hand over the enclosure (very near to it but without touching it) the spinning noise became significatnly louder (a whooshing sound). I guess the noise was somehow echoing off my hand and the enclosure, so not sure how this interaction of noise would play up inside a PC case?
In anycase this is just a very initial test, I will post a better evaluation when I have a chance to install it in my main rig. From the initial testing however, I would summarize that the Smartdrive 2002C seems to do a very good job of keeping the drive cool, will probably quiet the spin noise of a relatively quiet drive to near inaudible levels, but will only be able to dampen and not silent loud seek noises.

Goto PW200M and a 7800GT

SmartDrive 2002C

2006-04-20T03:28:00.000+08:00

Many silent PC enthusiasts have heard of the SmartDrive enclosure for silencing hard disks, but because of it's expensive cost and limited availability not many have actual experiences with one. The SmartDrive has been produced by the Japanese company www.gup.co.jp for many years now, who claims over 60,000 units shipped outside of Japan. Recently stumbling onto their Japanese website, I discovered their retail/online store was selling their 2002C (copper model) for 5480 yen, or approximately US$50. It's still not cheap, considering you can get a 80GB HDD for the same price, but much less than the $70 or higher price seen at US/European online retailers. It was only after I made my purchase that I discovered apparently there was some kind of special promotion (as I don't read Japanese) and the normal price seems to be around 8000 yen closer to the US/Europe price. Needless to say I was happy that I just by chance lucked out to get the special price (maybe I should have gotten two...)

I do not live in Japan, but happened to have a friend visiting Tokyo who helped me with the purchase. The package came nicely wrapped and delivered by UPS Japan.

Here the SmartDrive is taken out and placed next to it's box. It has a nice anodized black aluminum finish and is actually a hefty weight. I purchased the SATA cable extender set for an extra $6.

Although it is possible attach a system's cables directly to the HDD, using the extension will make it possible to detach the HDD without having to open up the enclosure. However I discovered the extra $6 was not necessary as a set of both SATA and ATA extension cables is already included.

The bottom-side of the enclosure with the six screws used to hold the bottom plate in place.

Opening up we see there are two large copper plates inside the black alumnim enclosure. Also a heavy foam lining along the edges makes for a tight seal.

The enclosure has the form factor of a 5.25" drive, so mounting it inside a SFF will probably be quite difficult unless one is willing to sacrifice the optical drive. But we'll see if there is any chance to fit it inside the Aria case. I plan to test this enclosure and will post how effective it is in both dampening the noise as well as cooling the HDD.

Individual Processor TDP / TcaseMax

2006-04-15T05:08:00.000+08:00

Starting with it’s Revision E cores, AMD has begun programming individual chip’s TDP rating into their processors. In the past AMD used a conservative maximum TDP rating that was often the highest theoretical TDP across several models and all production batches. TDP was often used to compare how power hungry certain brands/lines of processors were against others, but now with this chip specific TDP, there is a way to quantify the sample-to-sample production variance and determine how a particular chip will perform in terms of power consumption. This becomes much more interesting to overclocking and silent PC enthusisasts. A low TDP chip will consume less power leading to less heat that needs to be dissipated (from both the CPU as well as the PSU) which means it can be cooled more easily or quietly. Conversely a higher TDP or a chip's ability to consume more power, is also speculated to indicate a better overclocking chip.

The majority of the information is compiled from AMD Opteron Processor Power and Thermal Data Sheet (pdf) and AMD Athlon 64 Processor Power and Thermal Data Sheet (pdf). Any clarifications, corrections, and new information is welcome.

TDP – WHAT IS IT?
TDP stands for Thermal Design Power, or also called Thermal Dissipation Power.

In AMD support forums we find this definition:

This is the maximum theoretical amount of power (in Watts) that a processor may consume and therefore dissipate as heat. TDP-values are crucial when it comes to designing cooling solutions as they specify the maximum amount of power (=heat)
that a cooling solution must be able to dissipate.Note that
AMD's TDP-values are absolute maximum, i.e. 'worst case' ratings.During normal operation, an AMD Processor will typically not reach its specified TDP. Other manufacturers may have a different definition for their TDP-values, e.g. they might give typical instead of absolute maximum ratings, i.e. they might specify the amount of power dissipated during normal operation (='typical' conditions).

In AMD’s Processor Power and Thermal Data Sheets (APPTDS) we find:

Thermal Design Power (TDP) is measured under the conditions of Tcase Max, IDD Max, and VDD=VID_VDD, and include all power dissipated on-die from VDD, VDDIO, VLDT, VTT, and VDDA.

TDP – HOW TO FIND IT?
Unfortunately the individual chip’s TDP is not indicated on the exterior or packaging of the CPU. It is however programmed into the chip as a TcaseMax value. From APPTDS

Tcase max is the maximum case temperature specification which is a physical value in degrees Celsius. This value is programmed into Rev D and later processors..
Tcase max is programmed during device manufacturing with part-specific values for Rev E and later processors with 'Variable'
indicated by the Case Temperature OPN character, and can be any valid Tcase max value in the range specified for the corresponding OPN

How is TcaseMax related to TDP? In APPTDS we find a series of thermal profile tables that translates Tcase Max to TDP based on the specific thermal profile of the processor.

The thermal profile is used to define the relationship between Tcase max and devicespecific Thermal Design Power for Rev. E and later processors with “Variable” indicated by the Case Temperature OPN character.

From the above definitions, it seems that during the manufacturing process, AMD will load each chip at the rated voltage/speed and measure the maximum case temperature and/or power consumption to obtain TcaseMax/TDP.

So how to read the TcaseMax from a Revision E or later CPU? A nice little program called AMD64 TcaseMax will read this value off your chip and automatically translate the value based on APPTDS tables. This handy utility can be downloaded here.

TDP – CLARIFICATIONS
Will the TDP rating be affected by the actual clock or voltage the CPU is being run at? No, TDP is measured at the rated voltage, max P-state (or rated/stock clock speed), and assume under maximum load so regardless of your clock speed or voltage, the TcaseMax utility should give the same TDP. The actual power consumption of the chip however will vary with changes in voltage and clock.

So does this mean that a dual-core Opteron 180 with a chip rating of 1.35v, 49C TcaseMax, and 35.0W TDP consumes only 35.0W when running at stock speed of 2.4Ghz under CPU load? Less than half the power of a slower dual-core Athlon X2 3800+ with a smaller L2 cache rated at 1.35v, 71C TcaseMax, and 89.0W TDP running at 2.0Ghz?! Even nearly half the power of a single-core Athlon 3500+ with a TcaseMax of 65C, 67.0W TDP running at 2.2Ghz? Well if our interpretation of TDP is correct, then yes that is exactly what it means.

Borrowing numbers from SPCR’s Desktop CPU Power Survey we see various A64 CPU’s with their TDP read by TcaseMax and their actual CPU power consumption. Although the measured CPU power draw does not exactly match up to the rated TDP, it is within -2.3 to 13.2W. The relative ranking in terms of measured power consumption matches the TDP rating except for the 3500+ Venice which has a slightly higher TDP rating by 1.4W than the X2 3800+ but measured in at 4.2W less. Considering that the motherboard used may not have been giving the exact rated voltage, the temperature/cooling differences of the CPU, as well as power measurement error, overall it seems that rated TDP does give a very good idea how the chip will perform in terms of power consumption.

In addition, even with all CPUs idling at the same clock and voltage (1Ghz 1.1v), we see the ranking of measured power consumption also matching that of the TDP. Again the only exception being the 3500+ Venice, which this time is drawing slightly more power than the higher TDP X2 4800+, but again this could easily be due to the factors mentioned above.

How does TDP affect the ability of a particular chip to be overclocked or undervolted? Well that is a question that unfortunately APPTDS does not seem to address. Any input that can help answer this question is welcome.

TCASE MAX IN DETAIL
Studying the APPTDS we find several figures that seem to determine the processor’s TDP. Each model will have a Thermal Profile that gives a specific Thermal Resistance (case to ambient) in C/W as well as a Tcase Max range. In addition there is a Local Ambient Temperature or Tambient. Plotting the various the Processor Thermal Profile tables, it seems that the relationship between Tcase Max and TDP is linear and can be roughly calculated using the following formula:

TDP = ( TcaseMax – Tambient ) / Thermal Resistance

Each series in the graph represents a different specific Thermal Resistance (C/W). We can see AMD slightly rounded down the highest TDP for some profiles. Most processor profiles are based on a Tambient of 42C, except the FX series which has a Tambient of 40C.

As we can see that different models have different Thermal Profiles and in general lower Thermal Resistance will give a higher TDP for the equivalent Tcase Max (and vice-versa). Will this give us any clues which models will have lower or higher TDP? Let’s look at a summary of APPTDS Thermal/Power Specifications tables. I’ve only included the Rev E chips, and this is per the last update of November 2005 for the Opterons, and March 2006 for all other models. In addition to giving the Thermal profile/Thermal resistance, the voltage range (VID_VDD) and TcaseMax range are also given for each CPU model/revision.

** Data points corrected as these seemed to be typos on the APPTDS.

Before we start salivating at the 20.6 TDP (for silent PC enthusisasts) or the 110.0 TDP (for overclockers), please keep in mind this is just a general range given in AMD’s specs and does not mean that there are even any chips that fall on these extremes. However we do see that in general the dual-core chips have a lower Thermal Resistance of 0.20, notably the higher speed X2s and Opterons. This would suggest a higher TDP, but from user’s posting on the TcaseMax utility forum and on SPCR, it seems most dual-core Opterons have an exceptionally low 35.0W TDPs although one Opteron 165 also had a rating that went up to 105.0W. It seems that it would be too soon to draw any generalizations with the limited data on actual chip TDPs.

However it is interesting to note we do find an Opteron Thermal Profile with a very high Thermal Resistance of 0.95C/W and an extremely low 7.4W-30.0W TDP range, but alas none of the listed models have this profile. The lowest TDP range can be found for the s940 dual-core 260/860 series OSAxxxFAA6CB/CC with 0.51C/W Thermal Resistance, 1.15/1.20v rating, and 13.7-55.0W TDP range; as well as the s940 single-core 240/840 series OSAxxxFAA5BL/BM with 0.53C/W Thermal Resistance, 1.35/1.40v rating, and 13.2-54.7W TDP range.

AND SOCKET 754 PROCESSORS
None of the s754 processors seem to have the individual chip TDP ratings despite having also moved to Rev E. However we can look at their maximum TDP rating just for comparison purposes.

The Rev E Athlon 64 3000 (ADA3000AIK4BX) as well as the Sempron 64s both have a rating of 1.4v, 42C Tambient, and 0.45C/W Thermal Resistance. The TcaseMax is 65C giving a 51W TDP for the Athlon, and TcaseMax of 69-70C giving a 59/62W TDP for the Semprons. Again this is the overall maximum TDP and not the TDP of the individual chip which will likely to be lower. The high Thermal Resistance, however does suggest that these chips in general should have lower TDPs.

The Turion (MT only, no ML) are also included in the APPTDS. The Thermal Resistance of these chips are amazingly high at 2.00-2.08 C/W. However instead of TcaseMax we find a very high TdieMax of 95C, and also the Tambinet is a lower 36C. Possibly this is because Turions come without the heat spreader found on other AMD64 processors. The maximum TDP is listed at 24-25W. Using the formula from above we get slightly higher figures of (28.4-29.5W).

No Batteries 2

2006-04-13T02:49:00.000+08:00

The other day, I stumbled across an improved version of the battery-less flashlight. This yellow version has quite a few improvements over the smaller blue version I found before, and costs only very slightly more priced at US$4.50.

First the beam of light is very focused and bright. It also remains bright without much noticable dropping off even after a few minutes. The second picture was actually taken on a brightly illuminated part of the floor.

The brighter light is achieved with one large LED in place of the three smaller LED's used in the past model. This probably also improves it's energy efficiency.

Here compared with the older version, you can see the more focused and brighter beam of the new version (although both lights were angled slightly differently, the perceived difference is quite similar to the photo).

Inside we find a larger rechargable battery rated at 3.6V 80mAh (8mA-14h) and indicated as a nickel metal-hydride type. This is twice the capacity of the older version.

This version now uses a hand-crank mechanism to turn the dynamo. Less effort is required to turn this crank compared with the pressure trigger. One slight design fault, the crank can be turned in either direction, but power is only generated when turned clock-wise. However the arrow indicating the correct direction is actually on the back-side of the grey lever and hidden from view.

Inside we see a set of gears translating each crank into multiple turns of the smaller dynamo.

An alarm or siren is one of the new features. The packaging actually claims 650dB, but I'm pretty sure it's an exageration even though it is quite loud.

A little compass is also included incase you get lost.

And a little jack for plugging in an optional (not included) cable to "charge your mobile phone". Once a plug is inserted into this socket, it will automatically turn off the LED and create a direct link between the dyanamo and the socket. Testing it with a voltmeter, the output is variable depending on how fast you turn the crank, but limited to a maximum of approximately 5.5-6.0v (it's a little hard to hold the voltmeter probes and crank at the same time).

The packaging, with it's interesting wording. "Hand Shake Torch" even though the mechanism is a hand cranked/turned dyanmo.

I was very happy with how significant the improvement was in brightness and duration of the beam. For only an extra $1 over the previous model, plus a list of extra features (although usefulness of these functions maybe limited to hikers/campers) make it a very good value.

Back to No Batteries

Notebook Silencing with a CF-IDE Adapter

2006-04-10T23:32:00.000+08:00

Here is my Sharp notebook that I've had for many years now, in fact it's been so long I don't remember when I got it. It was however one of the very first "sub-notebooks" introduced to the market long before this category of notebooks became popular. It was one of the first of Sharp's Mebius line (known as Actius Ultralite A100 in the US), weighing in just under 3 pounds and less than 1" thick, which was quite an accomplishment at that time. It came with Windows 98 (non SE) so I'm thinking 7-8 years old. That black block sticking out on the left is an Orinocco PCMCIA wifi adaptor (yes notebooks actually came without Wifi for those of you young enough to only be familiar with Centrino).

It comes with an Intel Mobile Pentium MMX 233mhz CPU. If I'm not mistaken, this is the Tillamook core rated at 2-5W, also the next to last chip from the original Pentium line before Intel moved to Pentium II. My history lesson from www.cpu-info.com. This is a 0.25 micron technology chip, compared with todays Core Duo which is at 65nm or 0.065 microns, that's nearly four times wider!!

Paired with the limited 32MB RAM and 3GB HDD, the system isn't suitable for most of todays apps other than extremely light work (aka surfing), it's hard to imagine this was my workhorse PC which I used daily for so many years (including a lot of database development work). However the screen is still quite nice even by today's standards (albeit 800x600 resolution), and it's handy when I need to do the occassional browsing from the rare part of my home where I don't have access to a PC or when turning on a large bright monitor is inappropriate.

Noise-wise the system is completely fanless, but unfortunately the very old 3GB HDD has an extremely loud whine. I'd like to think this has developed over time, as it's hard to imagine I could have put up with such noise all those years.
Replacing it with a quiet notebook HDD like the Samsung was a thought, but the $60 cost didn't seem like a good investment considering how infrequently I use this notebook. Another thought of using a Compact Flash (CF-IDE) adaptor crossed my mind many times. I have several extra CF cards laying around and such adaptors weren't too expensive. Although such a solution has it's limitations, for my particular use it seemed ideal. I managed to pick up such a CFDisk.2 IDE/CF adapter for $18 which was shipped together with my PW200M/PicoPSU purchase.

Compact Flash memory tends to be much slower than today's hard disks with much lower transfer rates. However this particular 3GB HDD doesn't quite come close to today's drive speeds, Sisoft Sandra measured a max transfer rate at a paltry 6MB/s. My San Disk Ultra II 1GB Compact Flash card is 50-60% faster rated at 9/10MB/s, not to mention it should also have faster seek times. The latency going through the CF-IDE adaptor will probably bring speeds down somewhat as I suspect it was not designed with high performance in mind, but still should be fast enough for my use.

Another problem is CF has limited write cycles, but again we are talking about more than 100,000 times and I figured it will be a long time before reaching that stage, the CF card comes with a 5 year warranty (lifetime in the USA), and I wouldn't have any critical data on the card.

I tested the CF-IDE adaptor using a USB external 2.5" notebook HDD enclosure and it worked without any problems. So let's begin the replacement surgery...
First, off come the 14 screws on the bottom of the Sharp. This actually was quite quick and easy to disassemble. Then I disconnected the keyboard and slowly lifted open the casing.

Here is a look on the inside. The CPU is located under the metal plate frame on the top half of the PC. At 3-5W not much of a heatsink is needed (sorry didn't take any photos this time as HDD was my main concern). The black rectangle on the left is the PCMCIA socket, the battery is the black block on the lower right, and the HDD in the lower left corner.

It's a Hitachi DK237A-32 HDD and interesting enough with a Nidec motor (although unlikely to be FDB - fluid dynamic bearing, simply based on the loud whining). The CF-IDE adaptor is marketed as

Intended for use with a cable, may not be mechanically compatible with existing drive trays of notebook computers.

However I found the ATA pins and bottom four screw-mounting holes to be positioned exactly identical to the 2.5" notebook HDD. Even the overall form factor suggested it would fit properly in any space formerly occupied by a 2.5" HDD, however if the notebook uses side mounting holes then some creativity would be needed for locking down the adaptor (aka double-sided tape or recycled cable ties) .

I had Windows Me installed on my notebook, and this took up a little less than 900MB, enough to fit on the 1GB CF card. I connected the old IBM HDD to my USB 2.5" HDD enclosure and copied the files to my desktop PC. Then using a regular USB card reader I copied the files onto the CF card. This did take longer than I expected; I didn't time it but certainly more than 15 minutes. I suspect writing so many files shows the CF card's weakness, but hopefully this will not affect normal use of loading an OS and just internet browsing.

Unfortunately when I plugged the CF-IDE adaptor into my Sharp notebook, the BIOS was unable to recognize it as a HDD. I subsequently tried plugging the adaptor into an ASUS A7V333 (Socket A) motherboard as well as an ASUS A8NVM-CSM motherboard, both had no problems identifying it as a 1G HDD. I tried copying the manual settings (cylinders, sectors, heads, etc) into the Sharp's BIOS settings, but still the adaptor/CF was not recognized.

I can attest that my Sharp was completely silent with the CF adaptor, but unfortunately without the BIOS recognizing it I was unable to boot up the machine. It was a good idea, but I guess the Sharp's BIOS was too out dated. I probably won't be using this notebook much anymore, but being my very first notebook I might just keep it for sentimental reasons.

NOTE : The CF-IDE adaptor is manufactured by PC Engine gmbh and carries a nice RoHS stamp. The RoHS directive was passed by the EU and stands for "the restriction of the use of certain hazardous substances in electrical and electronic equipment"; it "aims to protect human health and the environment through the restricted use of certain hazardous substances." With the cost of electronics coming down every year, certainly we can afford to pay a little more to help keep our environment clean. With everything becoming more and more disposable, most electronics eventually end up in landfill, plus the way "old" models are being replaced within shorter and shorter cycles, poses a real problem for our future. This coming from someone who lives in a city of 7 million where even the most basic recycling is not practiced, unfortunately most countries and people really aren't doing enough to take care of our environment and it's the next generations that will suffer.

PW200M and a 7800GT

2006-04-05T23:58:00.000+08:00

I was quite curious whether the PW200M could handle a high end graphics card like the Nvidia 7800GT. Although the 7800GT is not a top of the line model, it is a very decent performer (paired with the Opteron 144 = 21,500 3DMark2001: 13,900 3DMark2003 : 6,700 3DMark2005) and at the same time relatively energy efficient compared with similar performing cards. The PW200M is rated at 200W which should be more than sufficient to run my system with the 7800GT. But my AC/DC adaptor was only rated at 150W (I did not purchased the 220W adaptor used in the Shuttle SD11G5 barebone as it's too expensive). From my calculations though, 150W should still be enough. Using the Asus A8NVM-CSM's onboard graphics, maximum system AC draw with the PW200M was under 85W. Based on previous readings taken with the 7800GT using the less efficient Aria PSU, the graphics card needed an additional 80W under load. Total 165W AC draw at an assumed efficiency of 85% would give 140W DC. This is just under the AC/DC adaptor's rating and seems a little too tight for comfort, so just to be on the safe side, I used CrystalCPUID to undervolt the CPU down to 1.0v (which ran Prime95 stable for more than 48 hours). This should save around 15W, keeping in mind this undervolt will only come into effect after Windows boots up.

The 7800GT requires a PCI-E power connector and comes with an adaptor that requires two molex plugs. As the PW200M only comes with 2 molex plugs and both are being used (HDD/CD), this 1-to-3 molex splitter was needed.

I am happy to report that the system booted up with no problems. The maximum AC draw came in at just under 150W, factoring in 85% efficiency we are looking at around 130W DC, very close to our original estimate. Certainly this disproves all the marketing hype and general push for higher wattage PSUs(bigger is better), although there are some systems that can get very power hungry.

QUIET PLEASE
So the PW200M and the AC/DC brick did their job of running the system, but unfortunately with the stock AMD heatsink fan, the reference 7800GT cooler fan, and the 120mm exhaust fan, the system was nowhere near quiet. Unlike the quiet starting point and easy modding of the Pundit, this higher powered system posed more of a challenge.
First thing was the 120mm fan. Actually much of the noise was caused by vibrations as the rear panel by itself is now very thin and flimsy once the original PSU was removed. Soft rubber fan grommets (seen in blue) was an easy mod and significantly quieted down the vibrations, leaving only the whooshing noise of the fan.

For the 7800GT, despite consuming relatively low power and having fan speed control, was still too loud for a quiet system. There are quite a few quiet VGA cooler replacement options, most of which are "two-slot" solutions (meaning the heatsink will also take up the adjacent PCI slot).

The Arctic Cooling NV Silencer 5 is a nice option which has the added advantage of exhausting hot air out of the system through the PCI slot. However I happened to have a Zalman VF700Cu cooler on hand, which from various reviews seems to cool only slightly less effectively than the AC, but has a quieter fan.

The Zalman comes with nice blue RAM heatsinks that are stuck on with adhesive thermal tape. The VGA heatsink replacement was also relatively easy, involving removing/fixing a few screws and cleaning/applying some thermal paste.

Zalman provides a molex adaptor that allows you to set the fan at 5 or 12v, however I preferred to keep to the cards own dynamic fan speed control. It has a low speed setting for 2D use, and a high speed setting for 3D mode. The fan header on the card is a 2 pronged, whereas the Zalman fan is a regular 3 pronged. The plastic frame of the card's plug however is easily removed with some prying, leaving two exposed pins where the 3 pronged fan header can be plugged in (red matched to red, black to black, third wire only for rpm monitoring is not supported/needed by the 7800GT).
Here is a before and after look.

Next up, the CPU cooler. Although the stock AMD fan is relatively smooth, it can get quite loud. The low profile of the Aria case does not leave a lot of space between the CPU and HDD cage, so choice of a replacement quiet CPU cooler is limited. I had a Thermaltake Pipe101 heatsink lying around that I bought for an AMD Socket A system I never got around to installing. It's a copy of the once popular Socket A cooling champ, Thermalright's SLK900A/SP97; it's 500g of copper with 4 heatpipes, comes with K8 mounts, takes 80 or 90mm fan, and is now sold for under $20.
The only 92mm fan I had was a Silverstone FN82, which is a little peculiar as despite it's 92mm blades, comes with a frame with 80mm mounting holes. It is rated at 26dBA, 37CFM, 2100rpm, 0.22A Max, and fitted with Maglev bearing.

The Pipe101 comes with an aluminum frame which wraps around the heat sink for mounting the fan. But despite it being designed with 92mm mounting, the edge of the frame actually blocks part of the fan; around 1cm on each side (left and right). Worried this may block airflow or even cause backflow, I decided not to use this frame. Instead I tied down the fan using (you guessed it) some recycled cable-ties.

Considering how hot the CPU heatsink can get, this probably was not the best choice, but I can replace it after my initial testing. Between the fan and the heatsink, I sandwiched a small foam disk to insulate the vibrations (this was recycled from the packing of some blank CDRs). No vibration or chatter could be heard, and again only the whooshing of the fan. The height of the HSF is just right, leaving enough clearance to the bottom of the HDD. Testing th
The last step was not noise related, but related to the PW200M. The 12V DC input jack was attached to a PCI bracket that was included with the PSU. Here is a view from the inside. I wasn't quite comfortable with the exposed contacts so I wrapped some electricians tape to cover it up.

For the plug from the AC/DC adaptor, this was still the temporary splice from before. I managed to pick up a plug so that it could be properly soldered.
Unfortunately with the very thick cable, plus having to solder 2-3 wires to each contact on the very small plug, made it more difficult than it should have been.
However I did manage, although you can see the plug cannot be closed properly due to the thick wires. Wrapping some electricians tape however would make do for now. I am thinking to replace the wire completely or even using different set of plugs entirely.

HOW QUIET?

Well certainly this cannot be compared to my inaudible Pundit, and it's too loud for me to keep on top of my desk at ear-level to be considered quiet. However on the floor under the table it is very well behaved, and although still audible it is quiet and unobtrusive. During light use, the noise blends into the background. Even under load running Prime95 and 3DMark2003, although the noise level becomes louder, it is very smooth (consisting mainly of whooshing noise) and not what I would consider loud. Certainly it is a significant improvement over the original Aria PSU configuration.
Temperature wise though is where you pay for this quietness, although on the most part they remained acceptable considering the small size of the Aria and low noise of the system.

CPU fan (Silverstone 92mm 82NF) was set by Speedfan, SYS fan (CoolerMaster 120mm fan) using Fanmate controller (all the way down, and all the way up), VGA fan (Zalman VF700Cu) set using RivaTuner. Ambient temp was 27-28C. IDLE was in Windows, LOAD was with Prime95 and 3DMark2003 running simultaneously.

IDLE (CPU 5x 0.8v) LOW
- CPUfan=1500rpm, SYSfan=550rpm, VGAfan=20%
- CPU=50C, NB?=47C, SYS=45C, HDD=38C, GPU=49C

LOAD* (CPU 9x 1.0v) HIGH
- CPUfan=1600rpm, SYSfan=1100rpm, VGAfan=38%
- CPU=53C, NB?=50C, SYS=45C, HDD=38C, GPUmax=65C

** All temperatures above were recorded with motherboard BIOS version 0601 which seems to overstate temperatures. Upgrading to a newer 0702 BIOS drops the CPU and NB temperature readings by 8C; SYS and HDD remain unchanged.

NOTES
One thing I did notice was the external AC/DC adaptor got very hot after more than half an hour of running. It was too hot to hold for more than 2-3 seconds. This was the same even if the system was only idling. I suspect these bricks are designed to run hot, and also keeping in mind nearly 10-20% of the AC draw is dissipated by the brick, but still the thought of such a hot exposed component on the floor makes me uncomfortable.
In addition, when under 3D load, the 12v rails read from the motherboard sensors dropped from 12.2v all the way down to 11.4v. Although this did not cause any problems, there were no signs of instability, and still within 5% of 12.0v, it would seem that the AC/DC brick is being worked hard.
If I decide to keep the 7800GT in this system I will probably purchase the 220W Delta brick from Shuttle even though at $90 it is quite expensive. However I originally bought the A8NVM-CSM/Aria for an HTPC and the onboard graphics is sufficient for this purpose; I may consider upgrading my main rig and putting the 7800GT in there instead.

Next A8NVM-CSM Aria Update

Back to PW200M installed

PW200M installed

2006-03-27T21:51:00.000+08:00

The PW200M DC/DC adaptor found it's place inside my Antec Aria Small Form Factor case. The Aria was designed to be a silent case with multi-layered noise-dampening side panels. This case was reviewed in detail a while back at SilentPCReview. Unfortunately the Aria's PSU fan gets too loud preventing it from being a very quiet system, even under moderate load. This makes it a perfect candidate to see what we can do by swapping the noisy PSU for a small and silent PW200M.

The nice large 120mm fan seen at the back of the case seems to suggest good airflow, but unfortunately the air must be drawn through the PSU from the cramped interior making it less effective.

Here you can see the PSU from the inside (right side of the picture) with a small row of intake grills where the air must pass through before leaving the case. There are no intake grills directly opposite the fan (which would make for a more direct airflow), but only on the four sides (top, bottom, left, and right).
Here's a top view with the optical and hard drive cage removed.

Let's compare the Aria's PSU to the PW200M. You can see a significant size reduction.

The layout of the ASUS A8NVM-CSM motherboard, had the RAM slots right next to ATX power plug, so an extension cable was used to install the PW200M up away from the board. This was actually an unused 24-20 pin convertor that came with a Seasonic power supply.

It installed easily and without issue. Here the PW200M is suspended only by the wires, but just for testing. I made sure it was not in contact with anything before powering on. I also added a 120mm fan where the old PSU was to provide some case airflow.

No problems, everything powered up on the first try and booted into Windows. I checked the PW200M expecting it to be hot like when I first tested in an Athlon XP system, but I was surprised it was only slightly warm. The PW200M was not even in the path of the 120mm fan.

I decided to fix the PW200M towards the rear of the case in the spot of the old PSU. There was the most space here, plus I figured the extra airflow from the 120mm fan wouldn't hurt. This required a longer ATX extension cable, which I had previously purchased at a computer accessories store. I made sure that the PW200M was secure and did not contact any other areas of the case.

This was done with some handy (recycled) twist-ties. The tie on the right is securing the ATX connector of the PW200M to the chassis, and the tie on the left just to keep the molex wires out of the way from the 120mm fan.

Here you can see the difference in space compare with the original Aria PSU on the left/top photo.

TEMPERATURES AND NOISE
I took several readings at different CPU multiplier and Vcore settings (set using CrystalCPUID). All temperatures were measured by Speedfan in Windows, and the AC draw was measured as total system power consumption from the AC outlet (V x A) using a clamp meter. I also included some readings taken with an A64 3500+ CPU which I originally had in the Aria. Other components used were:
- Stock retail AMD CPU heatsink and fan (controlled by QFan)
- ASUS A8NVM-CSM motherboard (Nforce 6150 onboard VGA)
- 2 x 512MB Crucial Ballistix Tracers DDR400 RAM
- 250GB Hitachi T7K250 HDD
- ASUS DVD-ROM

The Opteron 144 CPU was run at 9x multiplier and 1.0v Vcore and loaded under Prime95 for over 24 hours with each PSU. The temperatures (highlighted in blue) show that with the PW200M setup, all temperatures ran 2-4C cooler. The 120mm fan used is a CoolerMaster fan turned (nearly all the way) down to less than 700rpm. Although this was not a quiet fan, at such low revolutions it was significantly quieter than the Aria PSU; the noise of the HDD spin and Qfan controlled AMD heatsink now became apparent.

** All temperatures above were recorded with motherboard BIOS version 0601 which seems to overstate temperatures. Upgrading to a newer 0702 BIOS drops the CPU and NB temperature readings by 8C; SYS and HDD remain unchanged.

* A small side note, a comparsion can be seen between the A64 3500+ and Opteron 144 being run at the same 9x multiplier and 1.2v Vcore (highlighted in yellow).Despite the Opteron's larger cache, it draws around 5W less power than the A64. The CPU temps are similar, but other system temperatures are mixed (one higher, one lower), possibly the rearrangement of wiring when switching the CPUs had this effect.

POWER AND HEAT
The overall system power consumption was also reduced using the PW200M. Here is a comparison across various states and settings.

Under CPU load, the difference is 9-11W or 11-12%. Under idle, 11-12W or 17-21%. Even when powered off, there is a 6W reduction.

Even after 24+ hours of Prime95, the PW200M did not feel hot, but only slightly warm to the touch. I am not sure why it was significantly hotter when I first tested it in an Athlon XP system. Maybe because in that system it was flush on the motherboard, or possibly the different current draw on the various voltage rails had some effect?
As for the external AC/DC adaptor, this did get warm, warmer than the lower powered AC/DC brick of my ASUS Pundit, but not hot. This however is a good sign showing that much of the heat is being dissipated externally through the AC/DC brick and not inside the Aria.

NEXT
Unfortunately my inaudible Pundit has really spoiled my ears, and although the Aria might now be considered a quiet system, it's still too loud in a very quiet room.
What the PW200M has done however, is remove the significant problem of a loud PSU, reduced the amount of heat in the case, improved case airflow, and made room for a larger CPU heatsink. This now allows further steps to be taken to make the Aria even more quiet.

Also I still plan to add a high powered Geforce 7800GT graphics card to the system. With an additional AC draw of less than 80W, this should still be well within the PW200M's capacity, however whether the improved airflow will allow such a card to be run quietly in the Aria remains a question.

Next PW200M and a 7800GT

Back to Pico PSU and PW200M DC-to-DC power supplies

Index

2006-03-23T23:53:00.001+08:00

Here is my attempt to organize and index my posts so they can be easily accessed from one page. This page will be updated as I put up more posts.

Small and Quiet PCs
===============

ASUS Pundit P2-AE2 slim barebone

APPLE Mac Mini

Silent DC-to-DC Power Supply Solutions

- Inside an Antec Aria SFF Case

- Inside an AOpen XC Cube EZ18 SFF Barebone

Small and Quiet Prebuilt Systems

Touch Screens

Budget Systems

Miscellaneous

===========================================

Booting WinXP on the Intel Mac Mini

2006-03-21T01:05:00.000+08:00

Booting Windows XP on the new Intel Macs, it was only a matter of time. Thanks to Blanka and Narf whose hard work has earned them the cash prize from Colin Nederkoorn' contest to be the first to provide such a working and replicable solution. The step-by-step guide has been posted at www.onmac.net as well as on their wiki page. The instructions are clear and straightforward, however not all issues are resolved and certainly there is a lot more to be tested; also your success may depend on what version of the XP install disc you're using as not everyone has been able to get XP running. Here are my experiences trying to install and boot Windows XP on my Core Solo Mini...

Thanks to an iMac discussion on SPCR, forum poster Erssa kindly pointed out the success story and provided a link to onmac.net. I skimmed through the instructions briefly last night, and it seemed clear and complete enough to follow. At first I was a little hesitant to attempt this on my new Mini; I was enjoying the comfort of FrontRow too much and the thought of something going wrong plus this was still a beta solution told me I should wait. Well curiousity got the cat, and to me the thought of being able to run Win XP was too interesting to not have a go.

I won't go through any details as again the step-by-step instructions at the above links are very clear, however I did have to do some things differently so I will mention them here. I will also try to give an idea of how much time it took for those wanting to attempt the same.

PREPARATION
The first step involved creating the installation CD for the Mac by patching your original Windows XP SP2 installation disk. I used a Volume License English version. It seems that your success may depend on what version (Retail/OEM, English/Non-English) as the instruction and patch files may need some adjustments accordingly.

As it was getting late, I only had time to create the CD that night. This part was so quick and easy, and led me to believe the rest of the installation would be the same. I was planning to take the morning off and stay home to finish the install, but something came up at work. Fortunately the Mini is so small and portable, it was a no brainer for me to bring it with me. Well it turns out that the rest of the install wasn't so quick and easy as I thought, needless to say I wasn't very productive work-wise today... Also unfortunately I did not have a camera with me, so you'll just have to bear with my text-only recount of the installation.

As XP needs to be installed in a seperate partition, the next step involved partitioning the hard disk. With OS X's disk utility, this meant erasing the HDD first, wiping out the previous OS X installation. It is possible some other third party partitioning tool will allow you to do so without erasing, but this has not been tested with this solution. Also booting XP from an external disk might be possible, but again no working solution yet that I am aware of. Well how long can a partition and OS X reinstall take...

1) Backup HDD to external HDD using Carbon Copy Cloner (optional)
This is not necessary, however I wanted to do this just in case something got messed up, I knew I could still get back into OS X. Afterall, I have less than 20G, how long could a backup take. I had an external firewire (prolific chip) HDD enclosure with a 40G 2.5" 5400rpm Seagate Momentus HDD (actually it was the drive I took out of my old G4 Mini). I'm not sure if it's because CCC runs under Rosetta, or maybe some indexing feature of OS X, but it took more than 45 minutes to complete which was slower than I expected.

2) Booting OS X from External HDD (optional)
Again not necessary, but just to confirm my backup works. In the System Preferences under the Startup Disk, simply select the disk to boot from. There have been some users having trouble with this, however it seems it's dependant on the type of firewire interface. Fortunately mine worked without any problems. It was quite slow though, over 1 minute to boot up compared with 25 seconds for the internal disk.

3) Erase and Partition the HDD
I split my HDD:
----Partition 1 = XP = 30G = MS-DOS
----Partition 2 = OS X = 25G = Mac OSX Ext (Journaled)

I decided to make the XP partition a little larger, figuring I have more XP software than OS X, plus OS X can read off the XP partition, but not vice-versa.

4) Reinstall OS X from the installation DVDs
Ooh boy, I remember I waited quite a while reinstalling OS X on my old G4 Mini, but didn't realize OS X install is really so slow...

First the installation utility does a check of the DVD. I assume this is not essential and there is even a SKIP button. I should have opted to skip, but wanting to be sure every i was dotted, I waited the 20 minutes for it to complete. Then the rest of the install (Disc 1 and 2) took nearly another hour. I suspect in the install screen there is probably some option to exclude programs you don't want installed and that might shorten the time somewhat. In total, my preparation took 2-1/2 hours, not an extremely long time and I did get some work done in between prompts, but certainly longer than I expected.

5) Finally, the XP install
Ok first the XOM.EFI file has to be copied to the OS X home directory and blessed, per the instructions. This nicely gives you a bootup menu to select between a crystal Apple Logo and a multi-color Windows Logo (I snapped these photos after I got my Mini home). In went the patched XP install CD I burned the night before...

A screen of text messages (white on black) from the XOM bootloader appeared while the optical drive chugged away. I was getting excited, especially after the several hours of prep work. Unforunately I didn't get past this screen and eventually the CD spun down and nothing happened.
Checking back on the onmac forums I found that a new XOM.EFI file had been posted specifically for the Mini with mention to solving freeze problem. Great, back in business. Unfortunately the same result again. Checking back on the forums again yielded different people trying different things with mixed success. This included typing g, ENTER, g, ENTER, or pressing F6, pressing ENTER, etc. What finally worked for me, thanks to some very useful advice from one of the posters (sorry can't find the thread to give proper credit) was holidng down the F6 key as soon as the text screen appeared (I actually started holding it down just before), and then to keep holding until the drive stops spinning. This time there was an additional line with "patching successful" or similar phrase. Then pressing ENTER after releasing the F6 key caused the CD to spin up again and successfully continue into Windows XP disk partition screen!!

For those of you having trouble, there might be some helpful tips that apply to your situation in these forums http://forum.onmac.net/ and http://forum.osx86project.org/.

I selected Fat32 instead of NTFS as that way I could both read and write to my Windows partition while in OS X. I also opted for the slow format just to be sure which meant an extra 15 minute wait, but again just to be extra sure.

I did have the resolution set to 1280x1024 in OS X prior to booting the XP install. It seems this helps keep the resolution correct for the XOM.EFI and also XP install screens, some users who didn't set this first had problems. Consequently this also meant all the XP install screens looked beautiful in hi-res and not the crummy low-res (800x600?) that we usually find on PC installs.

With each passing step of the XP install, my heart rate climbed. I was really quite excited and anxious at this point, and things were moving along smoothly!! I had to record this, my super low-res phone cam would have to do. Unfortunately it got further clipped during the transfer to my PC, but you get the idea...

6) XP Installation continued
XP install successfully completed it's first part, and continued after rebooting from the HDD! In the instructions, it mentions install hanging at the end and needing to manually power down the Mac, but in my case with the Mini, XP completed installation normally and rebooted itself. Well between trying different XOM.EFI files, searching on the forums, reading through posts, (also took time to search for my CD key as I forgot to bring my original CD with me), I lost track of how long the install actually took. However it was about 3 hours from the time I first put in the XP CD until the install finished, but the actual install itself probably took less than 40 minutes.

7) XP Boot
The moment of truth... (again photos taken after I got home)

Selecting Windows from the boot screen...

The same text screen seen during the install (maybe some lines have changed)... Windows is loading... it must be...

YES!! SUCCESS!!
(ok this shot was taken after I had already booted XP several times and installed some stuff, but it was very exciting the first time)

Without any drivers installed, the boot time into XP is around 25 seconds (hey just like OS X!).

Here is a shot showing the XP partition from within OS X. OS X now recognizes the formerly named "XP" partition as "NO NAME", but I don't think I'll try to change it.

So What Now?
Well being a PC silencing enthusiast, let's see what some PC silencing tools can do.

SPEEDFAN
Unfortunately v4.28 is unable to detect any temperature (other than HDD), fan speed, or voltage sensors. It does pick up FDC37n972 super i/o chip and has two speed controls, but changing their settings doesn't seem to have any effect.

CPUZ
CPUZ identifies the 65nm Yonah Intel Core and displays a nice Coro Solo inside logo. Also shows speed step (correct name?) automatically cycling the CPU from 6x (1Ghz) @1.004v to 9x (1.5Ghz) @1.404v under load.

CRYSTALCPUID
CrystalCPUID seems to be working at least partially. The multiplier can be set between 6x and 9x and the respective frequency is verified in CPUZ. Voltage can also be set from 1.004v to 1.404v, also with the same being displayed in CPUZ, however I am not quite convinced that vcore is actually being changed. CPUZ is displaying the selected value verbatim and not a fluctuating reading normally found on AMD boards, casting doubt if this is the true vcore. Unfortunately without any voltage or temperature sensors, it is quite difficult to verify. Running maximum clock at minimum voltage 9x (1.5Ghz) @ 1.004v under Prime95 seems perfectly stable, which can be read either the chip undervolts perfectly at this level, or the vcore is not being set. There is a lot of hot air being blown out when running Prime95 but I could not notice any difference between the different voltages.

Here is a screenshot at idle with CPUZ, Speedfan, and CrystalCPUID open.

And under load with Prime95.

With Prime95 running, even with the lowest voltage setting there are no signs of instability (Prime was only run 30 minutes). Although a voltage lower than 1.004v can be selected in the listed options, CrystalCPUID doesn't let the set voltage go lower than 1.004v.

HEAT AND NOISE
Well so far I don't notice any difference noise-wise between booting into XP or OS X. The fan seems to be operating the same in XP. Running Prime95 even after an hour, I don't notice the fan spinning up, but certainly the hot air coming out the back can be felt. The top of the case does not seem hot either. Possibly the fan is spinning up gradually, but it is hard to tell.

Despite people's different preference for XP or OS X, I think most people will appreciate the option to dual-boot.

ISSUES
This is just the first day with XP on my Mini so it's hard to say what issues will come up when I install more applications. It seems most of the drivers (lan, wifi, bluetooth, etc) are working, however people are having limited results notably with the video driver. This will probably be solved shortly, especially for the Mini since it uses Intel's own onboard GMA950 solution. Some minor problems I noticed so far (note that I have not installed any of the drivers, which may or may not correct this)

- occassionally the system boots into XP or OS X with power LED remaining off
- rebooting from OS X to XP with resolution lower than 1280x1024 causes XP to be improperly sized (you only see a corner of the XP desktop); rebooting from XP to OS X also seems to set resolution at 1024x768 even though it was set differently last time (this might be monitor related)
- XP shutdown, drive spins down and screen goes blue, but must power off manually
- Standy option is greyed out from within XP

These are only very minor points at this time, I'm sure there will be more interesting and exciting discoveries along the way...

Pico PSU and PW200M DC-to-DC power supplies

2006-03-17T22:27:00.000+08:00

DC-to-DC power supplies for PCs
Power supply solutions using a DC-to-DC adaptor have been around for a while, but traditionally they were used mainly in Mini-ITX cases and car PCs. I have been pondering the idea of using this in a SFF PC since a fellow SPCR forum member "lm" posted such an idea over a year ago. However articles, reviews, and even fellow user experiences remained scarce, and it wasn't clear how well they would work in a desktop system. On top of that they weren't readily available and cost more than good quality ATX power supplies.
However after my good experience with the ASUS Pundit's external power supply setup, allowing it to run virtually silent, I decided it was time to step up and give these DC/DC adaptors a try.

The Pico PSU and the PW200M are both DC/DC adaptors that come integrated on a small circuit board and can be plugged into practically any motherboard that utilizes a 20 pin ATX connector. These boards do need to be paired with an external "brick" AC/DC adaptor.

Let's have a quick look at these first...
Here is the Pico PSU. Yes that is the complete DC/DC adaptor. That's all that needs to go inside the PC.

It is indeed very tiny and the "Pico" name is well earned. It's not much larger than just the motherboard connector of a normal ATX power supply!!
And here is it's older brother, the PW200M. These are two seperate and independent DC/DC adaptors. The PW200M is slightly larger, but still very compact. We'll discuss their differences later.

In case the molex connectors don't give you a good enough idea about it's size, here they have their photo taken with a traditional ATX sized power supply.

So why would anyone want to install one of these inside their PC instead of a normal ATX power supply? Well of course we mean someone who wants a Small and Quiet PC...

1) Both the AC/DC as well as the DC/DC adaptors can be fanless, meaning they generate no noise.

2) In general such a setup will be more energy efficient especially at low power consumption levels where most silent-oriented PCs will tend to be. This also means the less energy that is wasted, the less heat that will be created. This should mean the system does not need as much cooling and fans can be run quieter.

3) The DC/DC conversion part should be more efficient compared with the AC/DC conversion, which also means most of the heat generated will be outside of the system again reducing the amount of cooling required inside the system. The claimed efficiency of the PicoPSU and PW200M is "greater than 95% at full load" whereas even the highest quality ATX PSU is rarely able to boast more than 85% (please note though, 95% only applies at full load and the AC/DC efficiency must also be considered when talking about overall efficiency).

4) With their tiny size, they only take up minimal space inside the PC. This becomes important for Small Form Factor PCs where space is already so tight and often airflow is not so good. Poor airflow will mean fans have to work harder. Removing the stock PSU should allow for much better exhausting of heat, keeping overall system cooler which again should allow it to run quieter.

So what's the catch, why aren't these setups more widely used? Well besides the ASUS Pundit P2-AE2 and the Mac Mini, there are actually other barebones/SFF that come with this type of setup. Shuttle has their Zen ST62K and more recent SD11G5, both very quiet SFF. Dell also runs their Optiplex ultra-small-form-factors like the SX260/270/280 or their current GX620 to name a few. ECS also did the same with their EZ-Buddie. However none of these manufacturers or other major brand are offering any DIY solutions you can apply to your home PC. Some minuses to consider.

1) Most people would choose to have their PC in one package and no external brick sitting on the floor.

2) Most of these DC/DC adaptors tend to be limited in wattage, 120W for the Pico and 200W for the PW200M. I am not aware of any larger DC/DC adaptor readily available. Whereas for ATX PSUs these are commonly in the 300-600W range, and some even going up to 1KW (in case you need to power all the PCs in your neighborhood :)
However what you can power with 120W or even 200W may surprise some of you; more on that later.

3) There is limited availability and costs tend to be higher. I was only able to find them in specialty Mini-ITX/Car PC retailers or otherwise resort to eBay. The Pico is generally priced around $45 and the PW200M can be found for $40. However adding in the AC/DC adaptor, depending on the wattage and where you get it from, could add another $35-80. Unlike lower wattage AC/DC bricks, high wattage bricks are a lot less common, especially ones that supply the 12v required by these two DC/DC boards. This compared to a high quality quiet ATX PSU like the Seasonic S12 330W which is sold for around $60.

4) Possibly other issues of electrical noise, durability?

Frankly for a regular sized PC, it's hard to justify the external PSU solution over a high quality quiet ATX PSU which can run very quietly with good airflow. For a SFF system however, the external PSU should have significant advantages... let's find out...

A CLOSER LOOK

Here is a closeup of the Pico PSU. The black and white wires is the 12v DC input. The other set of four wires provides the 12v/5v to the P4, molex, and floppy power connectors.
The two yellow squares and the tiny grey one on the back are metallic casings, possibly for isolating electric noise assuming this is a switched mode convertor, or to serve as a heatsink for dissipating heat? Again any engineers reading this are welcome to enlighten us by adding your comments.
From the side we see that the Pico is made of two double-sided circuit boards sandwiched together. Some ICs can be seen in between.

Here is the back side of the PW200M where it has the same three square metallic casings. In addition we see a series of capacitors not found on the Pico (CapXon 390uf 35v and 1000uf 10v).

A SMALL HICCUP
Here is the external brick AC/DC adaptor that I ordered with the Pico PSU. It's rated output is 12v / 8.5A which is 102W, but it claims a maximum 110W and was the largest wattage brick available from the store where I placed my order. Unfortunately it arrived defective. When I plugged it into the AC outlet, it immediately started ticking and the LED began pulsating. Measuring the output voltage showed a cyclic fluctuation from 10.2-12.2v.
The international freight cost was about the same as the cost of the adaptor so it really made no sense to return it for a replacement. I tried to see if it could be repaired locally, but unfortunately in our everything is disposable society, I couldn't find anywhere that repairs AC/DC adaptors. Manufacturing costs have come down so much that the replacement cost is less than the labor cost to repair it... My own attempts to repair it were thwarted as I couldn't even get the case open despite having removed all screws and trying to pry open the seams.

Well I'll say again, high wattage 12v adaptors are extremely hard to find. Replacement AC/DC adaptors are plentiful for notebooks, but these usually stay below 80W, and for the higher wattage adaptors that I did find, were usally more than 17-19v. In the end I found a second hand Lite-On brick rated at 12V 12.5A (150W) for a pricey $50. Anxious to test out my DC/DC adaptors, I bit my lip and took out my wallet.
Both DC/DC adaptors have a simple round 5mm DC input connector with a + and - connection. Unfortunately the LiteOn brick had a 4 pin din plug. I had no luck trying to find the pin out on the internet for this adaptor, but it seemed that some video equipment used a similar power connection. Then I read some AC/DC adaptors actually output two seperate rails, maybe that was the reason for 4 pins instead of just a simple two? This never occurred to me before, but if the 150W were split on two rails, then just taking one rail might leave me with 75W or less, which would not be enough power for the PC I was planning. So opening up the brick was in order, that plus I hadn't the slightest clue regarding the pin out.

Please remember opening up any electronic equipment especially power supplies can be dangerous even if it is unplugged there may still be residual charges. Having said that, the LiteOn brick opened up easily after the four screws were removed.

Fortunately my fear about seperate rails was unfounded; here we can see the red and white wires are soldered together to one connection, with the black, brown, and shielding wire soldered together to another connection.

The solder points can be seen better from the underside.
Ok let's put back together the brick and start splicing the wires.

Hurray a 12.2V reading! A quick check with the voltmeter reconfirms our understanding is correct. The red and white wires both correspond to +12v and the black, brown, shielding wires to ground (please be careful that other models/manufacturers could be wired/colored differently).
Off comes the old head and we splice in a round 5mm connector salvaged off some dead brick at the same second hand store.
No soldering yet, just some electricians tape for now so we can get to testing. Notice how much thicker the wire is coming from the LiteOn brick, I plan to purchase a connector and solder it on directly later, but for testing this should not be a problem.

TESTING
Finally we get to hook up the Pico PSU with a 12V input! A green LED on the front lights up to let you know it's receiving power.

Using a paper clip to simulate the PC ON switch by shorting the ATX connector's PowerOn pin. A nice little red LED on the back let's you know you have powered on.

Without any load, checking the various rails all seemed quite reasonable. The -12v was a little loose, but probably still acceptable.

Pico PSU (12.2v, 5.0v, 3.3v, -13.3v)
PW200M (12.2v, 4.9v, 3.25v, -14.3v)

I was however getting a very odd reading when the DC/DC boards were in power off mode. The 12v rail was reading 3.2v-4.5v instead of the expected 0v. This had me concerned and a little hesitant to test them in a real system. Maybe this was only very minor current leaking through (as the 12v rail is actually passed in from the 12v input)? Well I did have an old system I was ready to give away and wouldn't cry (not too much anyways) if something happened to it. Specs:

Athlon XP-M 2600+
ASUS A7V333
256MB DDR333 RAM
120GB Hitachi 7200RPM 3.5" HDD
GeForce Ti4200 VGA card

First let's check the system power draw using a regular ATX power supply. Unfortunately my Seasonic PSUs (which SPCR has tested efficiency curves) were not available and I had to use a Tagan TG380-U01. The maximum AC power draw measured was 150W. The Tagan's efficiency at load if I remember correctly is only 70%, and probably 65% or less at lower loads. That would mean the equivalent of around 100W DC draw. As the Pico is rated at 120W, this might be pushing it especially when the load balancing between the different voltage rails cannot be checked. So I decided to drop the processor speed to a safer lower consumption level. Initially it was set at 12.5x166 or 2.08Ghz, dropping it to 8x166 1.3ghz gave a more comfortable peak of 130W AC/90W DC (Just a note, despite the XPM mobile processor being able to run at very low voltages, the A7V333 motherboard I used did not allow voltages lower than 1.6v hence the higher power consumption).

THE MOMENT OF TRUTH
Ok, let's start with the higher rated PW200M. It's that tiny green card located in the middle of the photo. It's probably clearer in the closeup.

There was no problem fitting this board directly onto the ATX power connector of the motherboard. However it should be noted that the board covers the release clip making it inaccesible for removing. Fortunately there is a tiny hole in the PW200M that allows a thin rod to be inserted down to pry open the release clip.

It was with a mixture of anxiety and excitement that I turned on the power button, but fortunately everything went smoothly and the system booted up without a hitch! This was only a first test, so nothing extensive was done. The 12/5/3.3v rails all seemed reasonable both via the voltmeter and motherboard sensors. In addition I was happy to confirm that once the PW200M was connected to motherboard, there was no more voltage measurable on the 12v rail when the system was off.

Next up was the Pico PSU. Here it is installed, it really is very tiny! Look Ma, no PSU!

No issues here either, the system seemed to run normally during the 20-30 minutes of my testing.

INITIAL USE IMPRESSIONS
So far there didn't seem to be any stability issues, but of course the system was not run for too long. Both the Pico PSU and PW200M do get quite hot however. Within a minute of powering up, the square metallic casings get burning hot to touch, and after 20-30 minutes of use, the overall adaptor is too hot to grip for unplugging. As these are often used in fanless or low airflow Mini-ITX systems, I suspect it should not be an issue and the specs rate operating temperature up to 85C.
Overall power efficiency is better than the Tagan ATX power supply by 15-20% based on the measured AC draw. If our guesstimate of the 65% is correct for the Tagan, that would mean the DC/DC+AC/DC solutions are running at around 75-80% overall efficiency.

State------Tagan....PW200M....PicoPSU
IDLE-----108W.........88W..........90W
Prim95--124W........106W........108W

WHAT'S NEXT
Well the next step is to put them into their intended small form factors for some real testing. How much will they improve a SFF based on noise, temperature, and energy efficiency? Also how much can these DC/DC boards really power, and will they endure such use?

The PicoPSU is intended for my slightly dated HTPC. It's an AOpen XC Cube EZ18, running an undervolted XPM 2200+ chip, 1G RAM, nforce2 board with onboard VGA, 2.5" HDD, and a combo-optical drive. The maximum AC draw was only 87W last time I checked, so the Pico should be able to handle it comfortably.

And for the PW200M, how about an Antec Aria SFF case with an Athlon 3500+ Socket 939 CPU, 1G RAM, Asus A8NVM-CSM Nforce 6150 motherboard with onboard VGA, 250GB 7200RPM 3.5" HDD, a DVD optical drive, and let's add a 7800GT high end VGA card.
No I'm not kidding, even with the 7800GT (not pictured here) the maximum AC draw running Prime95 together with 3DMark is 183W. Assuming the Antec PSU is running at 75% efficiency, we're looking at less than 140W DC.

SilentPCReview did a detailed analysis of various systems' loading on their PSUs and a very similarly speced system running a more power hungry 6800GT measured a total DC power draw of 125W under CPU load (no 3D loading). As the article points out, most of the loading is on the 12v rail, which they measured at 7.8A. The PW200M's 12v rail is rated at max load of 12A with a peak load of 13.5A. With my 7800GT the difference in total AC draw between CPU load (no 3D) and CPU+3D load is about 52W, and if we assume this is all on the 12v rail, even at 80% efficiency this translates to 3.3A + 7.8A = 11.1A still within the rated 12A. However whether or not running a system so close to it's limits is wise or not is another matter.

Only time (and testing) will tell...
Also I need to get a second AC/DC adaptor, and preferably one close to 200W. So far I've only managed to find the 220W Delta 12V brick used in Shuttle's SD11G5, but with a $90 price tag, I'm still searching for other options.

I will post back after I've had a chance to run these PSUs more...

NOTE: For the defective AC/DC adaptor, upon notifying the online store they offered to immediately send me a replacement. However as the two-way international freight cost would be more than the adaptor's retail price, I figured I would try to get it fixed locally. Since that didn't work out, the store issued me a full refund on their own initiative without asking me to send back the defective brick. I do not have any affiliation with this store (other than having placed on order from them), but would like to add that their customer service has been very professional; they responded quickly to all my questions pre and post sales, shipment was sent immediately upon payment on the exact date they informed me (a Saturday), and resolved the defective item in a more than satisfactory manner. Thank you Carl at www.short-circuit.com .

Goto PW200M installed

Goto PicoPSU Installed

P2-AE2: Stealthed and Silenced

2006-03-07T22:54:00.000+08:00

After some initial silencing of my ASUS Pundit, the very quiet Samsung notebook harddisk became the main source of noise. Well thanks to some helpful advice on the SPCR forum from Alleycat who credits another forum member Cruelsister, comes the original idea of using coolpacks. Coolpacks are usually frozen and put in portable coolers for keeping things cool, like drinks, but as we'll see they can be useful even without being frozen. They are soft and gel-like making for a good vibration and noise dampening material. In addition they are able to absorb and transfer heat, making them suitable for packing inside an enclosed aluminium container with a hard disk.

Unfortunately I did not have access to any aluminum enclosures or electronics housings of the right size and couldn't find anything suitable in the local hardware stores. However searching in several stationery stores turned up this metallic pencil case.
No I have no preference for "Miffy", but this case seemed to be the most suitable size for enclosing my 2.5" HDD with some extra room for the coolpacks. The cost, well $3 ain't too bad and certainly worth a try.

Another problem was there were no coolpacks to be found (I guess picnics are not too popular in a metropolitan city) however I did find this "Head Cool" pack used for... head cooling. This also seemed like a good size, although the plastic wrap may be a little thinner than I would have liked. This cost $8.

Let's see if the harddisk will fit... Width wise it's perfect, allowing 1-2mm of room to spare. Length-wise it's a little bit on the long side, but we'll see how it fits inside the Pundit.
That tray on the right came inside the pencil case but won't be needed here.
With the IDE cable attached, closing the pencil case required a little extra care and some pressure. The pre-installed IDE cable linking the motherboard to the harddisk to the optical drive, was a little on the short side, it would have been easier if it were just 1/2" longer.

Well even though the pencil case is a little long (unecessarily blocking the case airflow) it manages to just fit. The height of the case is also just right, not even one millimeter to spare. The hard drive cage just manages to slide on top. Somebody from above is watching...

The system was clearly quieter. The majority of the HDD spinning sound was muffled, and the seeks remained inaudible. More on sounds and temps later, first I'd like to share where the Pundit has found it's perch.

STEALTH PC
Although the Pundit is a relatively small PC, it's horizontal orientation means it will take up too much valuable desk space if placed on top of a table. Unfortunately I don't have any shelves nearby, and I wasn't quite up to drilling holes on the underside of my desk to mount it upside down (although that still sounds like a great place for it).
So how about stealthing it, say... behind a monitor, a la iMac. Well actually the idea came from seeing a mountiing bracket for one of the Dell Ultra Small Form Factor Optiplexes mounted behind a monitor.
Well with it's small footprint, the Pundit had no problem hiding behind a 16" LCD. The Pundit is actually behind the screen in this photo.

Let's peek over the top, see it's really there.

Well ok, here you can see it better from the side. As there are no vents on the bottom, the Pundit could actually be moved right up next to the LCD depending on the design of the LCD. My Eizo screen actually runs very cool and no vents would be blocked if the Pundit were right next to it, however the LCD stand/base extends back quite a bit, keeping the Pundit about 2.5" away.
Between the Pundit and the wall, some space must be left as that's where the intake vents are (in my setup approx 2.5"). The optical drive and front ports are still accessible, although you have to reach around the side of the LCD.

Also the optical drive tray will be facing away, but maybe the optical drive can be remounted upside down. Noise-wise there is a big bonus, first the intake vents (ie where the noise comes out) is now facing away from you, and second the LCD provides further noise insulation being placed between you and the Pundit.

Well what about all that talk of horizontal orientation? Ok so you don't always have to follow instructions exactly, right. There are a set of side exhaust vents, now on the bottom, which would get blocked,

but using some adhesive rubber furniture feet easily allows the Pundit to be raised a few cm to allow airflow.
The system does run slightly warmer set vertically, but temps are still acceptable.

HOW QUIET?
With this setup, the Pundit became very very quiet, the quietest computer I have ever used (notebooks included). However the very very soft whir of the muffled harddisk could still be heard in the wee hours of the night when my room became very silent.
I was ready to close this project and publish my post, but coming so close to a real silent PC (I define as inaudible to my ears in the quietest environment I can create) meant that I just couldn't give up yet...

Well foam is one of many free silencing materials available to PC silencing enthusiasts. Here again recylcing and putting to good use some packing materials that otherwise would end up as landfill. The foam I found was not too dense, probably not extremely good at sound insulation, but since the amount of noise remaining was so small, it might just do.

I cut a small piece just large enough to fit in the pencil case. When closed it pressed firmly between the case lid and the top of the harddisk. I did a quick check by repeatedly running Sisoft's disk bench to be sure the foam wouldn't cause the HDD to get too hot and then proceeded to put the system back together.

Silent? I tried so hard to hear this system I found myself hearing things I never heard before. Ok first my subwoofer had to be turned off, it has a very slight electric hum, but I should expect that as I can usually hear this when I have my head under my desk. Then I started hearing a hum, but it was coming from a battery charger's AC adaptor. I turned every electronic device off in my study. Hmmm... I still hear a very very faint hum, let's turn off the Pundit to check... no change! It was only the faint hum from the ringing in your ears when you don't hear anything. YES!! SUCCESS!! My first SILENT PC. I absolutely cannot percieve any difference in noise with the Pundit on or off, and this is with my head less than 2 feet away at the same level!!

Temperture wise it does run a little warmer than before because in addition to the pencil case and foam, the fans also had to be run slower now that there is no more HDD noise to mask it. NEW LOW = CPU fan @ 23% 850rpm, SYS fan @30% 661rpm. NEW HIGH = CPU fan @35 1350 rpm, SYS fan @45% 1400rpm. The noise is completely inaudible at LOW, however at HIGH there is a moderate amount of wind noise coming from the fans, but at a very smooth tone (cutting a larger vent for the 80mm CPU fan would probably help). The weather has also become a little warmer and the ambient has risen to 24C

STATE ---------- FANS --------- CPU - SYS - HDD
IDLE ------------ NEW LOW -- 31C - 36C - 39C
PRIME+HDD - NEW HIGH - 48C - 39C - 51C

Not everyone will be happy with a HDD temp of 51C, but this is within the operating temps of Samsung's specs. Running electronic components at higher temps may shorten their life span, but probably still allow it to function long enough until it becomes obsolete first. Please keep in mind these temps were achieved by continuously running Sisoft's HDD bench utility in addition to running Prime95 to load the CPU, which purposely stresses the system to it's limits, and is not something that you encounter in normal use. Of course if you live in a very hot climate, you would probably have to increase the fan settings to reach cooler temps and sacrifice a little on the noise side.

CONCLUSION
Being able to stealth the Pundit behind my LCD made it perfect for my already cramped study; I only had to sacrifice a little desktop space (sliding LCD forward 3") and a very tiny bit of floorspace for the AC adaptor. Then I found the Pundit was very easy to make quiet using some software utilities, and the final step to an inaudible PC (which is very difficult to achieve) didn't turn out to be any harder. None of the mods were difficult or required any fancy skills, and with a little scavenging the cost for the upgrades were very modest at $16 ($5 for the Zalman fan, $3 for the pencil case, $8 for the coolpack, and the foam/twist ties were recycled for free). For a small form factor PC, the Pundit is not expensive to begin with (my system including all components came in under $500).
However the Pundit does not have fancy styling, no fancy features, basically no expandability (beyond what you can add via USB/Firewire), and moderate processing power. It's perfect for office work, email, surfing, even decent at multimedia (and doing it all silently!), but probably not a good choice for a HTPC and certainly not for heavy gaming or CPU/data intensive applications.

Back to Silencing Update

Back to Overview

Mac Mini 2 - 1st update

2006-03-06T01:26:00.000+08:00

After only a few days of light use on the new Core Solo Mini, the biggest difference from my old Mini my old G4 Mini is the remote control. If you use the Mini as a desktop then this may be nothing more than just a novelty, but since I use my Mini as a secondary computer hooked up to my bedroom TV, it's wonderful and leaves my bluetooth mouse in the dust. Even though the bluetooth mouse offered cordless freedom, trying to operate a mouse on the bed wasn't the easiest and even simple interfaces like DVD Player required considerable concentration and dexterity.
Of course the remote and Front Row isn't perfect. Some of the time it's not very responsive, pressing a button could mean several seconds of waiting before anything happens (with no indication the command was received). Sometimes this can be explained the first time you select a function and the Mini has to load up the program, but other times it's not clear why there's this delay. However overall it's still very usable and beats a mouse anyday for the functions that it can handle. Of course there are only certain applications and limited functions accessible using the remote and you still need a mouse to do other things.

TV SHOWS
Apple's iTune's store now offer TV shows for purchasing and downloading. Although you can't do any purchases (yet) via the FrontRow interface, you can access them once they've been downloaded. For those of you who are Sci-Fi fans, you may be familar with the new Battlestar Galactica series that Universal in their marketing wisdom, released a season 2.0 DVD set which only includes shows for half the season. This means those of us living outside the US are left hanging (and BSG actually has a storyline that is strung out across many episodes). Well iTunes wonderfully offers a solution, although at $1.99 an episode, and not exactly what you'd call DVD quality.
First step is to purchase it from Apple's iTunes music store.

And then easily accessible in FrontRow. The quality is certainly watchable on a TV set, but there's room for improvement without worrying they'd be competing with their DVD sales.

This is much easier than using a PVR and all the shows are nicely categorized and sorted by iTunes. This is great if you missed an episode on TV, however at a $1.99 a show it doesn't make for a very economical TV replacement. Maybe if Apple had a monthly flat rate with unlimited downloads (like Napster's plan)...
Well if you're not quite up to paying for your TV shows, you can always check out the free Movie Trailers that'll give you a taste of what FrontRow video watching is like... of course when the server is up.

(For those astute enough to notice, actually most of the screenshots were taken on a Dell widescreen LCD, however the Movie Trailers screen was off my old Sony TV set)

QUIET REVISITED
I never really noticed the big difference between the perceived level of noise in my bedroom compared to in my study. The same PC sounds significantly louder in my study than it does in my bedroom. I guess my bedroom has a larger window, slightly higher ambient noise level, is a much larger room, and has more sound absorbing material like a large bed with a soft comforter, curtains, clothes, etc. compared with my small study which has a large hard flat surface of my desk, the wall cabinets just above, etc.
I must correct what I previously said about the noise level of the ASUS ASUS Pundit barebone. Actually it is quieter than the Mac Mini, and by quite a bit in a very quiet room. I would rank the Pundit the quietest, followed by the new Core Solo Mini, and then the G4 Mini.

PERFORMANCE
In my previous post, I forgot to mention that my G4 Mini has OS X Panther installed, while the Core Mini has OS X Tiger. I also found out that Xbench actually has a newer version 1.2 which is updated to run on x86 Macs (ie Intel chip). I reran the new version and came up with the following:

TEST - G4 1.25 - Core Solo 1.5
CPU....... - 48 -55
Thread... - 56 - 86
Memory - 28 - 95
Quartz... - 57 - 61
OpenGL - 62 - 166
UI........... - 24 - 44/79 **
Disk....... - 20 - 41

** One peculiarity was on the User Interface test. On the Core Solo, the scores kept increasing as I reran the test, going from 44 to 53 then 79 and dropping back to 62. No other tests showed so much variance, maybe something particular to OS X Tiger?

Next Booting WinXP on the Intel Mac Mini

Back to Mac Mini Core Solo has arrived

Back to impressions of the G4 Mac Mini

Mac Mini Core Solo has arrived

2006-03-04T14:34:00.000+08:00

Today I came home to find a plain cardboard package sitting on the floor of my living room. A tell-tale barcorde sticker with the Apple logo told me it was my new Mini! Apple's online store had delivered earlier than their ETD, only two days after I placed my order, and three days after Apple announced this new model. That's one thing Apple does very is well is insure it's newly launched products are really available when they are launched (as they should be).

Inside the shipping carton, we find the Mac Mini box with it's carrying handle. Let's open it up...

Here it is next to my older 1st generation G4 Mini. Can you tell which is which? From the top, both look identical, only my older Mini looks slightly more yellowish and the grey Apple logo is slightly softer, but that's probably from being exposed to the sun for over a year now.

From the front, if you look carefully you can see some difference. The disc slot is wider on the New Mini, and hidden on the right side is a tiny IR receiver for the new remote. The tiny hole for the power LED is also very slightly smaller an has moved a little.

From the back the difference is more significant. Notice the reduced overall width of the exhaust vents. This allows an extra 2 USB ports to be added along with an audio in port. Both audio ports now support regular analog as well as digital signals. For those looking carefully, my old Mini has a RJ-11 port for the modem which was already dropped on newer G4 models.
Overall he bottom remains pretty much unchanged except the row of text that used to be at the top has now been moved to the bottom. Also minor changes to the formatting of their product identification label up top.

The external AC adaptor power brick looks exactly the same, but looking carefully at the spec it has now increased to 110W over the previous 85W. I assume this would be to accomodate the more powerful Core Duo model?
Here we see the little remote used for Apple's Front Row home media interafce. On the right is a iPod Nano (sold seperately).

The Nano is actually a little thinner than the remote, but slightly wider and longer.

FIRST IMPRESSIONS
Well I haven't had much time with the my new Core Solo Mini, but here are my initial impressions. Comparing both units side by side, the new 1.5ghz Core Solo Mini versus my old 1.25ghz G4 Mini, both with 512MB RAM, a 40G Samsung IDE notebook harddisk in the G4 and a 60G Fujitsu SATA2 notebook harddisk in the Core Solo:

PERFORMANCE
The startup time of the new Mini takes half the time of my old Mini (around 20 seconds faster). On my old Mini, there is a short pause of around 15 seconds before the Apple Logo will display which accounts for most of this difference. Opening up apps is also faster on the new Mini on the most part. Interface-wise (ie menus, selecting options, etc) though the new Mini doesn't feel any faster.
For 720p H264 playback (Ice Age 2 Trailer using Quicktime), the new Mini had no problems and was very smooth. The old Mini however had a lot of trouble, only managing 12fps out of the encoded 24 fps for much of the clip.
Running Cinebench, the new Mini was faster approx 60% on the CPU rendering test, 90% on the Cinema4D shading test, and 150% faster in the OpenGL Software lighting test. Interestingly the OpenGL Hardware lighting test showed the old Mini slightly faster by 10%.
I also had an old version of Xbench lying around, but the results seemed very misleading and I suspect that for such an old program, much of the tests especially involving low level commands would not be appropriate for the new Intel based OS X. However some of the tests which aren't so CPU specific maybe relevant. The disk benchmark gave the new Mini a faster 85 score compared with the old Mini 51. Also OpenGL on the new Mini was 229 compared with 108. User interface though was rated at only 92 compared with the old mini's 214. Maybe the added processing power of the Core chip makes the onboard Intel GMA950 video (using shared memory) acceptable in most usage, but having a seperate video processor like the 9200 with dedicated memory (although not a lot) in the old Mini still has some advantages?

POWER CONSUMPTION AND HEAT
We couldn't go through without looking at this. I used a clamp meter to measure the total AC power draw from the outlet.
Actually the power draw for the new 1.5ghz Core Solo Mini was very similar to my old 1.25ghz G4 Mini. During startup, peak consumption was around 38W. Idling for the G4 was slightly lower at 21W compared with the Core Solo at 25W. In Cinebench, both peaked around 41W (slightly lower for the G4). The only significant difference I found was in sleep mode. Here the G4 Mini has neglible power draw (below the 2W that my meter can measure), but the new Mini draws 7W. I assume this is to power the circuitry that will allow it to check the IR receiver and wake when the remote is pressed.
I'm actually not familiar enough with Macs to know the best tools to run for maximum CPU/system loading, and seeing Cinebench didn't even spin up any fans, the system probably wasn't being fully taxed. Also the HDD and optical drives weren't being loaded either, nor any external Firewire/USB devices connected . In anycase though, the 110W AC adaptor does seem to be overkill, at least for the Core Solo.

Heat-wise, both put out warm-to-hot air via the back vents. I didn't take any temp readings, but neither one felt hotter or cooler. The top of the case of the new Mini however did get warm (not hot) whereas the old G4 remained cool to the touch. Possibly the Core Solo is putting out a litte more heat, but with the very close power consumption figures I wouldn't guess the difference to be too big. The smaller vent size on the new Mini probably means the hot air is not as easily vented out so the case gets warm, or maybe Apple has managed to better make use of the case to help dissipate the system heat. Which brings us to the next topic...

HOW QUIET IS IT
Apple claims that their New Mini is "even quieter", is it true? Well from my initial impressions I will have to say it is. Maybe it's the reduced vent size managing to keep more of the noise inside, or maybe they have also slowed down the fan speed somewhat (maybe also why the case gets warmer). My old Mini has a louder whoosing sound, whereas the new Mini is a softer sound but with a very slight hint of whining when listened to with my ear next to it (similar to the kind of whining when you spin a small fan very slowly). I must point out though that my old Mini seems louder than I remember it, as before the fan/exhaust was not as audible as it is now. I will have to double-check this, maybe being placed on my desk is making it sound much different then it's usual perch ontop of a loudspeaker or something was jostled when I moved it?

A LOOK INSIDE
Well I haven't had time to put my putty knife to it yet, but others have done so. Macworld opens one up and also Applefritter's Take Apart Guide. For the RAM it seems that the DDR2 RAM is actually notebook sized and not desktop sized as I originally imagined, but there are two slots.

DOES IT RUN WINDOWS
Good question. There have been many people trying to get their Intel iMacs to boot Windows, but so far no success stories. I don't suspect it will be any easier with the Intel Minis, but given some time someone will figure out a workaround...

Macworld also has some more extensive benchmarking here.

Next - Mac Mini 2 - 1st update

Back to impressions of the G4 Mac Mini

Mac Mini 2 - Here come's the Intel one

2006-03-03T02:57:00.000+08:00

On Tuesday, Steve Jobs announced the new Mac Mini with the Intel Core processor. This new notebook processor has interestingly enough found it's way to Apple's desktop systems before even being available on Windows notebooks. Slowly more and more non-Apple notebooks are appearing on the market with Core Duo, but Apple has already been shipping it's desktop iMacs with Core Duo for over a month.

Following in the footsteps of the Pentium-M, Intel's new core processor is perfect for a small and quiet computer. Low energy consumption, the ability to run relatively cool, and decent computing muscle to boot. Fabricated using the latest 65nm technology this chip promises even better energy efficiency and eventually lower production costs (there is already talk of price cuts of up to 30% coming in May, but mainly only the higher end chips).

Back to the Mini, besides changing from a slightly under powered, old generation G4 CPU to the Intel Core, the new Mini also brings with it Apple's Front Row (remote control and interface for home media), optical audio in/out, faster DDR2 RAM with double the expansion capacity to 2GB, newer SATA HDD interface, and two additional USB ports.

I really liked my old Mini which is being used as a second computer in the bedroom. As it was the very first model it still has the older OS X Panther and iLife05. I've been wanting to upgrade to Tiger for it's new features like Spotlight and Widgets, but the $130 cost seemed a little too steep, afterall it's only an upgrade. Same for iLife06, some of the minor improvements IMO isn't worth the $80 either. However with the new Mac Mini which includes Tiger and iLife06 plus Front Row, with a brand new Core Mini, this seemed like much better value. That plus Apple's claim "In fact, the new Mac mini is even quieter than the already whisper-quiet previous generation." made it too hard for me to resist. My new Mini2 is on it's way from Apple's warehouse as you read this.

Apple has scrapped the lowest cost model (previously without wifi/bluetooth) and now only offer two models. The Core Solo model at $599 and the Core Duo at $799. The Core Solo is a single core CPU at 1.55 ghz and the Duo is a dual core CPU at 1.66 ghz. Certainly the Duo should be faster especially in processing intensive apps, but being my second (ok third, well actually...) computer I figured the Solo would be "enough" and should also run cooler. That plus the fact a 1.55ghz Pentium-M is no slouch and I would expect the same from the Solo, also it should be a good improvement over the G4. For the Core Duo model, Apple actually claims up to 4x improvement in speed over the G4, but we'll take that with a grain of salt.

I must admit I was very tempted to go for the Duo model, latest technology, "dual-core", fastest Mini, etc. but the price premium Apple loves to add on it's higher end models turned me away in the end. From Intel's price list, the difference between a 1.66 Duo and a 1.66 Solo (they don't list a 1.5, maybe Apple just downclocked the chip to make the distinction extra clear between the two models) is only $32. The only other differences, HDD capacity, DVDRW versus DVD/CDRW adds another $65. So total cost difference for what you get is less than $100, yet Apple is charging you $200!
Also if you want to upgrade from the standard 512MB to 1GB Apple charges you $100, but purchasing a stick of any other big-name brand 512MB DDR2 costs only around $40. Needless to say, I didn't order any upgrades, I expect upgrading the RAM yourself won't be difficult like the original Mini (I have my putty knife ready). That plus upgrades means you have to wait a few extra days for them to custom configure.

Hopefully my Mini2 will be arriving by next week, I don't think the courier service Apple uses here delivers on the weekend...

Next - Mac Mini Core Solo has arrived

Back to Impressions of My G4 Mini

No Batteries

2006-03-01T15:26:00.001+08:00

A flashlight that doesn't require batteries is nothing new, many of you have probably seen the early model which involves shaking the flashlight to power it's lamp. However for those who have tried it, know it's not very practical in actual use.

Well I stumbled upon a slightly improved version that is powered by a simple palm squeezing action. This make charge-as-you-use quite practical (unlike shake action) plus it's quite compact (about palm sized), relatively bright (3 LEDs), holds a decent charge for short use, and is quite inexpensive (I paid USD$3.50).
In addition to being enviromentally friendly, you don't have to worry about the cost of replacement batteries, or the batteries going dead. As a bonus it even doubles as a grip excerciser, although the boxes claim of "improve the health" and "make your hands flexible" seem a little far stretched. I have the slight suspicion this packaging is not made for export to an English speaking country, afterall who would want to buy more "manual pressure", we already have enough other pressures in our everyday lives!

Let's have a look inside:
Squeezing/pressing down on the lever turns a gear that drives the dynamo. The lever is spring loaded so that it pops back out after each squeeze. The dynamo, similar in principle to an electric motor but working in reverse, uses a spinning magnet's magnetic field to cut through a coil of wires inducing a current.

Here's a shot with the dynamo disassembled. The flywheel and magnet are on the right. Two very fine exposed copper wires run from the base of the dynamo to a switch. The windings of the copper wire can't be seen in this photo, but are located behind this base.

Soldered to the bottom of the switch is an IC labelled LT+ DF06M 5217. My guess is this is some sort of voltage/flow regulator circuitry inside. Oddly checking with a voltmeter, there is no measurable voltage at the input to this chip (coming from the two exposed copper wires of the dynamo),
but up to 3.8v can be measured from the output of this chip depending on how fast the dynamo is spinning.

With a degree in electrical engineering you think I would be able to easily understand and explain this, but alas no (too much time spent enjoying "other aspects" of college life) - I welcome anyone willing to offer some enlightenment. In the meantime moving on, one output of the switch is connected to a 3.6v 40mah rechargeable batter. That's not a lot of energy, but for low powered LEDs it gets by.

When turned on, the light will draw from the battery. Squeezing the lever will provide additional power directly to the light. Turning it off allows all power to be applied to charging the battery. Towards the back is an additional switch for locking the lever in it's down position keeping it compact.
So how well does it work? Well after five minutes of squeezing (that's about how long before I tired out both hands), it runs pretty bright for up to a minute. After that it slowly becomes dimmer and will linger on for several hours although at that point it's not providing any useful illumination. The battery seems able to hold the charge for a while as well (it remained charged when I first took it out of the box, unless of course the store keeper squeezes them everyday).
The peak brightness (same as when it is being squeezed) lasts only around 10 seconds, after that it dims slightly. The only way to maintain peak brightness for longer periods is to keep pumping. With more practice it will be less tiring (and you will develop more muscular forearms and hands) and it's pretty easy to keep the light beam relatively still, but there is the dynamo whirring and lever squeezing sounds to deal with ruling it out for any stealth operations :)

Googling "self-generating flashlights" actually turned up quite a few sites selling beefed-up versions of this flashlight with a hand-crank dynamo, up to 5 LEDs, larger batteries, and even the ability to power another device or charge your cell phone. Of course the cost for such units are higher, running from $20 to $40.
Some interesting projects, Mark Hoestra turned one into an iPod charger and later also used a converted a bicylce dynamo.

Just a warning, the day before I bought a very similar flashlight with the same claims, but only when I opened it up did I discover that instead of rechargable batteries, it contained regular button cells found in watches. Some unscrupulous manufacturers will think of anyway to make an extra penny! I picked it up from a street stall, ok it was only $2.50 and had a working dynamo but still...

Mac Mini

2006-02-25T02:58:00.000+08:00

Mac Mini
With the popularity of the Mac Mini, the abundance of reviews, and chances that most of you have at least seen one, I'll try not to bore. However with it's super small size and quiet operation, it makes for a perfect Small and Quiet PC!
Granted it doesn't run the more common Windows, but OS X is actually a very friendly operating system and has quite a few neat features not found in Windows. We won't get into any Windows vs OS X discussion here, but each has it's advantages and for general computer use both are equally acceptable.
Although the old G4 processor is somewhat dated and underpowered, it's perfectly fine for day to day tasks of email, surfing, chat, skype, light photo editing, playing back MP3s, DVDs, etc. The base model starts at $500, but the $600 model is probably the better choice with built-in bluetooth, wifi, and a larger 80GB notebook sized HDD. Also included is a slim DVD/CDRW drive, DVI and VGA out (same port with an adaptor), 512MB RAM, OS X Tiger, and iLife06 media suite. It's not as cheap as Apple would like you to think, but considering it's the smallest widely available PC out there (minaturization comes at a cost) and it comes with Apple styling/branding, it's actually not too bad.

The Mini (seen here with the Nano and a 2nd Gen iPod) makes a wonderful addition as a second computer, and comes pre-installed with all the software to play DVDs, stream internet radio, play MP3s, dock with your iPod, download/edit/view photos, connect your bluetooth accessories, email, .. And it really is plug, and play.

With it's tiny size and low noise, it fits unobtrusively in the bedroom. An optional $20 DVI to S-video adaptor allows it to be connected directly to a TV with very decent image quality at 1024x768 resolution.

The Mini only has two USB ports (plus a firewire) and all located in the back, so adding a USB hub is probably a good idea. This $5 matching white unpowered hub works just fine.
Also a bluetooth mouse and headset (handsfree skyping and occassional late night DVD viewing) are easily paired. Apple's sleek looking bluetooth keyboard however was a little pricey and a little large, so a matching small sized silver USB keyboard was selected instead. It even has a backlight for glowing at night...

So how quiet is it? Well the Mini also uses an external brick PSU which helps keep out a good part of the heat and noise. It does have one fan that can get loud when the CPU is loaded, but I've had my Mini for a year now, and it's only a rare handful of times where I heard this (during some installations/updates and when compiling an iPhoto album). The only other source of noise is the notebook hard disk spinning inside. Depending on the HDD that comes inside your Mini (as Apple uses several brands/models) the noise level may vary. My Mini originally came with a Seagate Momentus that was certainly not quiet and had an annoying clicking noise every time it tried to park it's head. However replacing this with a Samsung notebook HDD made it very quiet. It's still audible late at night in a quiet bedroom, but the soft whir easily fades into the background.

How about further silencing to obtain the inaudible PC? Unfortunately with it's very small case and custom cooling solution, this makes any modding very difficult unless you happen to have good metalwork machining skills...

Mac Mini 2 - Intel Core Solo

P2-AE2: Silencing Update

2006-02-20T23:38:00.000+08:00

TWEAKING
Simply using some software utilities, it was very easy to get the Pundit to run very quietly without getting too hot. Using CrystalCPUID to replace Cool'n'Quiet, the Sempron 3000+ was undervolted:
-- State Multiplier VCore
-- IDLE 4x 1.10v
-- LOAD 9x 1.20v
Speedfan was then used to control the fans, using a LOW setting to keep fan speed down, and a HIGH setting once temperatures got hot under load. The LOW settings was selected so that the fans were virtually inaudible over the spinning of the HDD. The HIGH settings were selected based on the highest fan speed before the ASUS fans clicking became apparent from a distance of 2 feet.
-- LOW : CPU Fan = 20% 700rpm / SYS Fan = 40% 1200rpm
-- HIGH : CPU Fan = 30% 1200rpm / SYS Fan = 60% 2100rpm

With an ambient temperature of 22C, the following system temperatures were recorded when the PC was idle in Windows, and when put under load with Prime95:
STATE FAN SETTING CPU SYS HDD
Idle --- LOW ----------- 29C 38C 42C
Load -- LOW ---------- 50C 48C 46C
Load -- HIGH --------- 43C 41C 43C

During the day, the system is inaudible at LOW and even with my ear within 2 feet, I have to listen very carefully to hear it over the ambient noise. At HIGH the system becomes audible, but the sound is smooth and not loud.
At night however, in a very quiet room even the slightest noise becomes apparent. At the LOW setting, the main source of noise is the spinning of the 2.5" HDD. The Samsung MP0402H used is actually a very quiet single-platter drive, rated idle noise at 2.2Bel and tested by SilentPCReview at 17dBA from a distance of 1m. I would say this is quieter than many notebooks found on the market today, and very close to my Mac Mini in which I installed the same HDD (the Mini having the slight advantage of a more enclosed case to help muffle the sound).

FAN DIRECTION/REPLACEMENT
Both fans are setup to bring cool air into the case, with the hot air being forced out the side vents. Switching the direction of one or both fans did not improve the cooling. One problem with the ASUS fan was the extremely audible clicking especially at higher rpms. I tried setting the PWM frequency to the highest with Speedfan or using a rheostat fan controller but it didn't help.
The stock AMD fan seemed to be smoother, but when run at low rpms wasn't any better than the ASUS. The holes on the AMD fan didn't match up, but a little encouragement from a pocket knife made sure it fit nicely onto the heatsink. I also tried several other 70mm thin fans including a Sunon Maglev, but alas none proved any better than the ASUS fan.

Next thing was to try a 80mm thin fan. Despite it's high speed rating and dual ball bearings, the Zalman 80x15 fan is a very smooth fan that runs nicely at low rpms. No clicking or whining here. There was just enough room for a 80mm fan, but of course the mounting holes didn't match the 70mm sizing of the heatsink.

Well you know those millions of twist ties that come tied around the cables of every appliance/accessory, well here is a chance to recycle and put them and put them to good use. They handily allow the 80mm fan to be mounted securely onto the heatsink. They probably aren't extremely heat resistant, but they are tied to the heatsink mounting frame on the motherboard which shouldn't get too hot.
With the ASUS fan, I had to keep the rpms low to prevent the clicking from getting too loud, but with the Zalman it was only the wind turbulence noise once the cover was put back that limited the high setting. This allowed the HIGH fan settings to be significantly quieter than the ASUS fan, and overall temps ran cooler, notably the SYS temp by 5-6C.
STATE FAN SETTING CPU SYS HDD
Idle --- LOW ----------- 28C 33C 41C
Load -- LOW ---------- 52C 42C 45C
Load -- HIGH --------- 42C 35C 40C

Even at night, the HIGH settings were very quiet, producing only a soft whooshing sound. I tried to see if airflow could be further improved by adding the virtually silent Sunon 40mm, but unfortunately there wasn't space near the vents for even such a small fan.

For those curious about fanless systems, even with the case open and undervolted, the system temps would slowly rise to the point were it got too hot. However with just one 80mm Panaflo M at 5v and an open case kept temps at a relatively cool CPU 41C, SYS 35C, HDD 35C under load.

HARD DISK NOISE
Well with the HDD being the main source of audible noise from here to a truly inaudible PC, we couldn't let it go without a try.
First attempt was to suspend the HDD, so none of the vibrations will resonate onto the PC frame. Here some rubber bands are used, but this is actually a poor choice as over time they tend to crack/cut/break especially in a hot enviroment; in anycase this is just temporary. Four rubber bands were wrapped around the bottom mounting screws,

strung through the chasis and held together at the top by a handy (recycled) twist-tie. Unfortunately the 2.5" doesn't really vibrate much in the first place, and this suspension didn't make any perceivable difference noise wise.

Next thought was to enclose the HDD and muffle the sound, but at the same time not let it suffocate in it's own heat. With the plethora of external notebook HDD enclosures available for as little as $5, this seemed like an easy solution. The aluminum casing would allow some of the heat to propogate out of the enclosed space and at the same time keep in the noise. The drive did actually run 2-3C cooler, but unfortunately there was no perceivable difference in the noise level (I should have known this from using those extermely audible external enclosures). Stuffing a soft cloth to close up the opening where the IDE cable passed did help dampen the noise very slightly, but also caused temps to rise 1-2C. Maybe a thicker casing or adding some acoustic dampening material to the outside? Will experiment more with this later...

SYSTEM PLACEMENT
The HDD noise is less noticable from the front of the case, so placing the system with the front facing you will help. Also with it's horizontal orientation, locating the Pundit on a tall shelf or on top of a counter may make it virtually inaudible with the slight insulation help from the furniture plus with it not being at ear level. I also thought of mounting it upside down to the bottom of the table, but wasn't quite ready to drill any holes in my desk.
From the fixed legs and thin width, the Pundit wasn't designed to be orientated vertically, although it can be done. Here it's put up on some tapes to let air out the downward facing vents. The temps were slightly warmer but still acceptable. If you choose this orientation, you should lean it against a wall or side of the table and preferably fix it to prevent it from tipping over.

THERMAL COMPOUND SCARE
(Photos removed for those with weak hearts)
Hoping to get better thermal performance, I attempted to replace the pre-applied thermal paste with some better performing Artic Silver/Ceramique, but found the heatsink stuck to the CPU and impossible to remove. If you change enough heatsinks/CPUs you may have had the gut-wrenching experience where upon removing your heatsink you find you have accidentally yanked out the CPU along with it (while the socket is still locked)! Well this time I had to forcefully do that as the only way to get the heatsink off (heating/cooling the CPU made no difference)!! Very fortunately no pins were damaged.
Even with the heatsink/CPU off the motherboard, I had a very hard time trying to pry the CPU from the heatsink, and when I finally managed, had to resort to some serious scraping to remove the remaining bits of thermal-super-glue. Shame on ASUS for using a substance closer to adhesive than paste, making it extermely difficult to replace a CPU on the Pundit without risking damage. If you plan on eventually upgrading your CPU in your Pundit, remove the pre-applied thermal-super-glue from the beginning and replace it with something like Artic Silver (non-adhesive type).
That being said, after all that work and scare, the temperatures were no different with the Artic Ceramique applied.

Next - Stealthed and Silenced

Back to Overview

HP Pavilion S7320n Slimline Media Center

2006-02-11T02:13:00.000+08:00

HP unveiled their new Pavilion Slimline s7300 Series PC on January 4th of this year at CES which according to their press release "gives consumers all the power of a fully loaded PC in one-third the size of traditional tower PC."

Interestingly HP has selected an Intel Celeron M processor, normally found in low end notebooks. This also means low power consumption, low heat, and potential for a very quiet machine.

Unfortunately this series seems to be avaliable only in the US at this time, so I can't offer any first hand experience. However from the limited user reviews online, all the feedback has been quite positive. Razzl mentions "does all of the things my recently expired full-size desktop did, only silently--no fan noise deafening you"; westernwoman26 mentions "extremely fast and quiet"; and even a SPCR poster Aglio mentions "This thing is very quiet. I'd be surprised if it's 25db at 1m; the loudest thing about it is HDD seek noise. My external HDD makes more noise." Even this Pc Mag review review rates it as "very quiet". Of course the ambient noise level and tolerance for acceptable noise is different for everyone, but initially this sounds promising.

The size although still large for a SFF, at roughly 9.75" x 4.5" x 13.25" can still be considered quite compact. It's roughly twice the volume of the ASUS Pundit, but it's vertical orientation means it could fit nicely under the desk or even on top without occupying too much desktop/floor space. Also the PSU is built in so there is no extra power brick. The more spacious interior also means the possibility of better airflow and more modding options.

An additional plus for this system is that it is very good value. For the higher end s7320n model, this is being sold online for US$550 (excl monitor) and HP is also offering a $50 mail-in-rebate. At only $20 more than the system I built using the Pundit barebone, this system includes a larger 200GB HDD, a faster 16X DVD writer with Lightscribe, and in addition has a 56k modem, 9-in-1 memory card reader, SPDIF connector, and comes with Windows XP Media Center Edition. If purchased seperately an OEM version of MCE sells for around $120, the Celeron M 380 1.6ghz CPU another $120, and around $150 for a 200GB HDD, Lightscribe DVDRW, and card reader. That leaves around $110 for the cost of the remaining barebone which really isn't too bad.

The PC Mag article reviewed the lower end s7310n with a slightly slower Celeron M 360 (1.4Ghz) processor and 512MB RAM. The Sysmark2004SE performance compared to a Sempron 3400+ desktop it was roughly 20% slower, but pretty much on par with a Celeron 340 (2.9 Ghz - non M) desktop. Also it seems that the one PCI slot in this system is occupied by a modem card, if you don't need that function potentially you can replace it with a low profile TV Tuner card or offer other expansion options.

Pictures from PCMag.

More information on the s7320n can be found on HP's PDF spec sheet.

P2-AE2: For the PC silencing enthusiast

2006-02-11T00:48:00.000+08:00

For those of us who spend time trying to make our PCs run as quiet as possible and expect anything described as silent to be practically inaudible even in a low-noise late night environment, some further details on the silencing potential of this machine.

FANS
The fan unfortunately does exhibit quite a bit of whining/clicking even at low speeds and can be heard when listened to carefully. Most of the whine is coming from the CPU fan, and although the smaller case fan also has some whine when listened to closely, the volume level is significantly lower. I guess part of this is due to the PWM and fan bearings. CPU fan.

Upon removing the fan from the heatsink, some whine/click can still be heard, but much less. The problem seams to be the fan vibrations causing the heatsink to chatter and making the whining noise. The heatsink is attached to the fans via a two seperate metal frames; the CPU fan to one frame that sandwiches the copper heatsink onto the motherboard CPU socket using 4 spring loaded screws, and the chassis fan to another frame that very loosely holds the other free floating end of the heatsink in place. This frame is secured to the back of the chassis by only one screw. Simply holding the fan firmly helps reduce the whine/click. I might try tightening the frames or looking for some heat resistant silicone sealant to dampen the vibrations and hold the heatsinks firmly in place.

Although the volume level of the CPU fan is quite low around the 1000rpm (low CPU load), the unsmooth sound even when decoupled from the heatsink makes it unsuitable for a silent PC. There probably are not a lot of choices for a good smooth 70x15 mm fan, let's see what we can find... I have some old fans I tried before on the AOpen XC Cube SFF which used a similar sized CPU fan, but I'll have to dig them up. A quick glance over at my Sempron retail box and I see the AMD stock cooler fan is the exact same size. Although this Delta fan is rated at 0.45A compared to ASUS 0.15A, it is much smoother, has less vibrations, and not noticably louder spinning at the same rpm. I will do further testing once I dig out my other fans and look for the best candidate. Alternatively there is also enough room for a 80x15mm fan (like the Zalman) although attaching it to the 70mm sized holes on the heatsink frame may require more creativity.

The smaller second system fan doesn't even spin up at idle and even under CPU although it also has some click/whine, it's noise level is much lower and you have to be within two feet to notice.

FAN CONTROL
The BIOS settings for ASUS Q-fan on this board unfortunately is limited to ENABLED or DISABLED. No setting of high/low target temps like some of ASUS's other boards. With Speedfan, I was very disappointed as I was unable to get this to control the fans via Windows. However I was very glad to discover later that unlike my other ASUS boards, disabling Q-fan in BIOS didn't automatically allow Speedfan control, I had to go into the Advanced Configurations and set the PWM mode for the ADT7463 chip which interestingly enough had much more options than most and even with Q-Fan ENABLED in BIOS, allowed Speedfan to take over the fan control
- Auto on CPU TEMP
- Auto on LOCAL TEMP
- Auto on REMOTE2 TEMP
- Always on full
- Controlled by hottest of LOCAL and REMOTE2 TEMP
- Controlled by hottest TEMP
- Disabled
- Manually controlled

The added control of Speedfan will certainly allow for much better flexibility to achieve a "silent" system. For example the quieter chassis fan can be set to take some of the load off the louder CPU fan.

TEMPS AND AIRFLOW
As far as temps/fanspeed read by ASUS Probe, under load CPU=45C/2000rpm SYS=36C/1700rpm, and idle CPU=30C/1000rpm SYS=37C/0rpm. This is with CnQ/Q-fan enabled and 24C ambient. These temps are quite similar to a same CPU/chipset system I have running at work in a larger mATX micro-tower case.

The two sides of the case is moderately warm, and I was initially puzzled why. Only looking more carefully at the system did I realize both fans are for intake and not for exhausting. This is completely opposite to the AOpen XC Cube SFF which also only two fans, both for exhausting. Maybe the absence of the PSU inside the case makes this setup work better? I will also try reversing the direction of the fans, maybe just the chassis fan for exhausting. I also happen to have a Sunon 40mm 0.8W fan, which would happily fit along the side vents. Although small fans are normally a bad idea, I found this particular fan very smooth and inaudible at low voltages. Of course with no additional fan or even power headers, it will need to share a fan header with one of the other fans using a y-cable.

POWER CONSUMPTION AND UNDERVOLTING
Another nice surprise was the very low power draw. Measuring AC power consumption from the electrical outlet using a clamp meter
31W idle
55W load with Prime95
60W load Prime95+DVD+HDD

For my work system using the same CPU/chipset/ram config, only a larger 3.5" HDD and SFX PSU, consumes around 50% more power!! (48w idle, 80w load). From various SPCR threads, it seems that these external brick PSU combined with a DC/DC convertor can perform around 80% efficiency under light loads meaning the Pundit system draws 25-48w DC . From the manufacturer specs, my Hitachi 7k80 80G HDD at work typically draws 5.7W at idle compared with the 0.85W of the Samsung MP0402 40GB notebook HDD in the Pundit, the difference is around 5W. Assuming power draw for all other components being equal, that would mean the SFX PSU at work is only running at around 40-50% efficiency which sounds a little too low, even for a cheap PSU that came with the case (I would have expected at least 60% efficiency). Maybe the Pundit's K8M800 board is running more efficiently, or the external brick setup is performing at an even higher 80% efficiency?
In anycase the Pundit's power consumption is very impressive - at full load running HDD and DVD it consumes as much power as a 60w light bulb, which really isn't a lot considering today's PCs.

Undervolting is not supported in BIOS, but thanks to CrystalCPUID we can control that from Windows on any CPU+mb that both support Cool'n'Quiet (for Semprons only s754 3000+ and above). With several other Sempron 3000+/K8M800 systems I had no problems getting the system to run stable at 1.1v (the lowest voltage allowed for these cores), however for this system I had to keep volts a little higher 1.2v. This drops AC draw a further 8W to 52W. This must be pretty near a record low for any SFF running a conventional desktop CPU.

OTHER POINTS
The way the system is oriented with the legs fixed on the bottom strongly suggests placing the system horizontally on a flat surface; like on your desk or a nearby countertop. Unfortunately with the fan vents facing up, this also means most of the fan noise is directed at your ears. There is no stand or legs for horizontal mounting, and the thin 2" height also means it won't stand up stable by itself. Maybe I'll try mounting it sideways to the inside legs of a desk or hanging it down from under the desk?

Some other silencing features include a soft rubber casket that goes between the HDD and the bottom of the chassis cage. However as there is direct coupling of the HDD to the side of the chassis, the benefits maybe limited. Using a Samsung notebook drive however, which was practically inaudible in this system did not allow me to test this. The BIOS also has a very interesting feature for directly setting the Acoustic management mode of the HDD without the need to use Hitachi's utility. This is the first time I have seen such a feature.

So far this is only day 2 with this system, but so far the silencing potential looks positive. Happy to receive suggestions/comments via blog or the SPCR thread.

Next - Silencing Update

Back to Overview

More information about silencing computers can be found at http://www.silentpcreview.com

P2-AE2: Special BIOS options

2006-02-09T02:38:00.000+08:00

The BIOS allows the Acoustic performance to be set on the HDD. Normally this has to be done using Hitachi's HDD Feature Tool utility which has to be run from a boot disk/CD, this is the first time I've seen it in a BIOS.
Checking with Feature Tool, the BIOS does in fact set the Acoustic Management level on the Samsung harddisk. For those not familiar with this option, this will help reduce the head seek noise on the hard disk. This maybe at the cost of some performance loss, the effectiveness of this mode depends on the HDD, and not all models have this. According to the Feature Tool utility, the range for AM setting is from 128-254. The BIOS settings result as follows:
PERFORMANCE=254
MEDIUM=130
SILENT=128

On Chip SATA BOOTROM nicely allows you to boot/install Windows directly on a SATA HDD without the need to load drivers from a floppy. Unfortunately there are still boards out there without this useful function.
The Q-FAN (automatic fan speed control) option unfortunately is very sparse only allowing ENABLED or DISABLED. Unlike ASUS's A8NVM-CSM board which nicely lets you set the high and low temperature targets so fans throttling can be customized.

Next - For the PC Silencing Enthusiast

Back to Overview

P2-AE2: Size compared

2006-02-09T02:23:00.000+08:00

This barebone measures roughly 11.25" x 11.5" x 2.25".And the external brick power supply roughly 6.5" x 2.5" x 1.5"

How does it compare with to the very small Mac Mini?

The chassis is actually about the same height, however the Pundit's legs bring it up a little taller than the Mini.

If we look at width and depth however, the Mini is clearly much more compact.
Next to it is also an AOpen EZ18 XC Cube, another SFF PC, albeit without an external power supply.

Everyone has different tastes, but personally the Mini and the AOpen appeal to me more looks-wise. The Pundit has a relatively nice clean look, but I find the look of the top vents and unstealthed drive somewhat distracting.

Next - Special BIOS Features

Back to Overview

P2-AE2: Other components

2006-02-09T01:39:00.000+08:00

My unit came with a Toshiba slim DVD-R/RW drive installed. However this is an option and I believe it depends what the reseller or local distributor decides to install. Some retailers mention a combo drive DVD player only with CDRW whereas others have no mention at all.

One problem I noticed is that the tight fit of the case opening would sometimes block the drive and prevent it from opening. Adjusting the frame where the drive is installed fixed this problem, but if you move your PC around a lot or have it where it is occassionally jostled, this may require some extra work. Also with the location of the power switch where you would normally find the eject button on most other PCs, you have to be careful not to turn off the PC by accident when you want to eject the drive door.

On the bottom of the hard drive cage, a soft rubber pad is included which should help dampen some of the hard disk vibration from being transferred to the chassis.
However this is for the standard 3.5" HDD, when using a quieter 2.5" notebook HDD this poses a little more of a challenge.
In interest to get the system running quickly, I'll look for a mounting solution later...

The external AC/DC power supply brick is a Lite-On rated at 100-240 V 50-60Hz input and 19v 6.3A 120W output.

The DC/DC conversion circuitry (seperating 19v in 12v/5v/3v/etc rails for running a PC) is fully integrated into the motherboard. So there is no hope of pulling this out to be used in another PC (a 120W AC/DC brick with a DC/DC convertor suitable for PC use alone costs around $80 online, more than a third of the Pundit's price).
The silver block is where the power lead runs into from the external brick.

The audio board was removed for taking the picture, here it is installed back in.

Next -> Size Compared

Back to Overview

P2-AE2: Cooling

2006-02-09T00:22:00.000+08:00

Both the case and CPU vents for air intake are located at the top of the case. A bright label nicely reminds you not to block these vents in 7 different languages. The exhaust vents are located along the full length of both sides of the case with similar dot openings. I initially mistook that the fans were used for exhausting air like on several other SFF.

There are only two fans in this system both located on a common heatsink. The heatsink is made of two blocks that are connected by what seems to be a heatpipe. The larger block (lower right in the photo) goes over the CPU. The smaller block (upper left) is actually just a grill that goes over the K8M800 nb chip. However this does not rest on the chip, but is held above it by a screw to the rear of the chassis. I assume this space is left so cool air can be more freely brought into the case and at the same time helping to cool the CPU heat transferred from the heatpipe.

Here the heatsink is detached and the bottom view probably gives a better idea how this works.
A layer of thermal paste comes pre-applied.

A Sempron 3000+ E6 revision CPU which supports Cool'n'Quiet is installed.

The main CPU fan is a 70 x 15 cm fan from AVC rated at 12v 0.15A (pictured). The case fan is a smaller 60 x 15 cm fan rated at 12V 0.1A.

Installed and looking from the side gives an idea of the clearance once the case is closed. This is around 7 mm so there is the possibility to upgrade to a thicker (and potentially quieter) 70 x 20 mm fan if one can be found.

Next - Other Components

Back to Overview

ASUS Pundit P2-AE2 slim barebone

2006-02-09T00:21:00.000+08:00

ASUS Pundit P2-AE2
Slim form factor barebone for AMD socket 754 with an external power supply.

The Pundit P2-AE2 is a moderately priced and very slim barebone using the VIA K8M800 chipset. ASUS previously released other versions of the Pundit, however this is the first time with AMD socket 754 support and an external power supply; both strong pluses for making a very quiet computer. AMD's s754 Sempron processors run very cool and consumes low power, meaning there is less need for loud fans to keep the system running at acceptable temperatures. The external power supply means there is no power supply fan, often one of the loudest components inside a small form factor PC. Moving out the power supply also allows the case size to be reduced, but at the same time this can restrict airflow and make the system more difficult to cool quietly.

OVERALL IMPRESSION
With basically no internal expandability, not even a PCI slot, it would seem this barebone is targeted more at the budget user, for low-powered home/office use. Firewire maybe the only nice extra not usually found on a budget board; there is no DVI, SPDIF, HD-Audio, or video-out making this an unlikely candidate for a HTPC. Serial and Parellel ports are also missing, although that is quite understandable as use for such connections are gradually fading out.
That being said however, the system in my opinion is pretty good value if you are looking for a basic and quiet system for the office or maybe a low powered home server. The system will even happily playback H.264 video at 480p which I am embarassed to say my current HTPC has trouble with (XPM2200+ Nforce2 system).
The price I paid was around US$220 for the barebone, but this included a slim DVD/RW which bought seperately would cost more than $80. Adding a Sempron 3000+ (Cool'n'Quiet enabled) CPU, 1GB RAM, and a 40GB Samsung notebook HDD the total system cost is around $480 (excluding monitor and OS). Although not cheap for a budget PC, when you consider this is a SFF PC it's actually a pretty good deal.

INSTALLATION
The quality of the chassis is very similar to previous Pundits. Not as refined or solid feel as the AOpen or Shuttle SFFs, however still decent and not something I would call cheap. With the open top, this was one of the easiest PC installations ever, not much different from installing onto an open motherboard maybe only the heatsink requiring a little more care. Excluding the time to take photos, the total hardware installation took less than 10 minutes.System integrators will love the ease of setup.

AND HOW ABOUT BEING QUIET
Using a quiet notebook HDD, the two fans become the only source of audible noise in the system. The fans are automatically controlled by the motherboard to slow down reducing the noise, and only spin up when the system gets warm under load. Compared to most any other out-of-the-box desktop, this system is certainly quiet under light CPU use. It is slightly louder than a Mac Mini (which only has one fan and a relatively well enclosed case). Under heavy CPU load (running Prime95) however the fans do get louder, and although not to an annoying level, certainly louder than what I believe most people would accept for a "quiet" computer.

Of course if most of your PC use only involves surfing the internet, typing email, chat, Skype, playing back MP3, downloading, etc then this makes a very good out-of-the-box-quiet barebone PC.

A closer look at:

1. Cooling
2. Other Components
3. Size Compared
4. Special BIOS Features
5. For the PC Silencing Enthusiast
6. Silencing Update
7. Stealthed and Silenced

More information can about the Pundit P2-AE2 be found at the ASUS website .