Sign in with
Sign up | Sign in
Your question

The Power Saving Guide, Part 2

Last response: in Systems
Share
June 1, 2007 11:21:29 AM

This time we analyze the impact on power consumption of RAM speed and timings, graphics card generation, hard drive type and form factor and of typical add-on cards.
June 1, 2007 2:43:02 PM

If I can save just £100 from my power bill a year - thats more to spend on beer.

Its worth it when getthing new system parts.

Build more efficient parts and save money, no matter the cost!?!
June 1, 2007 3:51:48 PM

I think way too much time was spent analyzing the ram speeds. Different chipsets along with air compared to water cooling would have been time much better spent.
Related resources
June 1, 2007 4:07:25 PM

Quote:
I think way too much time was spent analyzing the ram speeds. Different chipsets along with air compared to water cooling would have been time much better spent.


Different chipsets.
I would have liked to see this. A good mother board holds it all together.

air compared to water cooling?
Do you mean power draw for fans and other cooling kit?

I guess a reason for not doing fans would be people can have an unknown number and there are a lots of makers. too many to cover them all.

I'm sure water cooling would use more power than a fan.
June 1, 2007 4:14:55 PM

I would have just like to have seen the power requirements of a board that is water cooling all the major components (cpu, vga, nb, vrms, and maybe ram) compared to a conventially air-cooled setup. Not wanting to see it broke down by fans etc, just a general air-cooled setup.
June 1, 2007 7:29:58 PM

I think the overall message I get from this article is that money should not be an issue, using the advice in the article can only save maybe $30 a year, which isn't a trade-off that really interests me.

The biggest issue for me is heat dissipation. As the warm summer months are approaching I can really notice the amount of heat my computer generates and it makes for sweaty, sticky computing.

So the two things I'm really considering is the efficiency of my power supply and integrated graphics.

A lot of websites will say "high efficiency" when referring to power supplies, does that really mean anything or is it BS? The PS isn't listed on 80 plus website so it can't be more than 80% efficient, so what the heck is "high efficiency"?

Lastly does anyone have a link to an article reviewing integrated graphics? Thanks
June 1, 2007 7:30:37 PM

A very decent follow up article. Looks like you took a lot of our comments about the first article into consideration. I look forward to the next in the series.

You mention spending *more* money to save electricity but there are really not many places you could do that, even if you wanted, and you didn't discuss them much. It would be nice if you did.
-PSUs
-CPU voltage (ULV and EE processors)
-mobile chipset motherboards
Pretty much everything else is spending less money to get a less power hungry device.

You also mention how it's probably not going to save enough money on the electric bill to be worth it for that reason alone but you didn't quantify it. About two paragraphs explaining how to calculate kWh per year and links similar to the ones I gave in the thread about the first article and a mention of devices like the Kill-A-Watt are in order. It's pretty simple and mundane but the article is incomplete without it.

At least a cursory overview of overclocking's effect on power consumption is certainly in order (although there are so many variables you will almost definitely get complaints that you "didn't test X, Y, or Z").

Would it be possible to, going forward, make energy consumption one of the datapoints on the regular VGA and CPU charts? I really want to see how the x1950PRO measures up in particular.

Why was HD power consumption only tested at idle? And how "idle" are we talking about? Were the drives spun down??? Was the OS installed on the drive and you're measuring the power consumption to run windows off the drive? A clean install of windows should allow for the drive to spin down though. www.iometer.org Max those puppies out with some 8k random read/writes and take another reading. And also mention the specs printed on the drives as those are the maximums and will include the amount of power necessary to spin up the platters which is where the major difference will be between slower, smaller, fewer platters vs. faster, bigger, many platters.

Memory matrix size would also be nice to see. Should be pretty simple to give an overview. Just compare a 512mb stick to a 1gb stick at the same speed and voltage. Probably not a big difference or very important, just to satiate curiosity.
June 1, 2007 8:34:35 PM

Still no PSUs?? What can we expect with Schmid on these pair of articles.

They probably spent more testing the 1066 v 800 RAM (BTW -- it's due to the 2.1V setting, try running your 800 @ 2.1V and compare that to the 1066) than anyone would ever save by using 800 over 1066.

Again, I'll illustrate an example: PSUs!

Let's say you have a PC that draws 300W (you have an 8800GTX or something, I don't care) under load. If you play games 4 hours a day, that would be load. Now, you also use the comp another 3 for non games and you idle at 150W.

If you use a HE PSU (80%) then you will be drawing 375W load, 188W idle. Through your day, you consume 2.06kWH.

If you use a LE PSU (60%) then you will be drawing 500W load, 250W idle. Through you day, you consume 2.75kWH.

As we expect, we can save 25% of the energy costs ((80-60)/80 = 25%) by using 2.06kWH versus 2.75kWH. Your comp runs as fast, it's performanc is not affected. You didn't swap out a Raptor for a notebook HD that can pull 40MB/s. You didn't underclock. You didn't sacrifice gfx performance. You didn't dump an add-in card. You didn't run your RAM slower.

All that said, we need to build more nuclear plants. It's the most effective way to add a large amount of extra energy with no carbon emissions (they produce water... and spent radiocative material). Wind and solar are good in theory, but can only produce enough energy to replace coal on a massive scale (I mean, wind will be able to the power the US only if we convert Wyoming, Utah, and Kansas into wind farms -- a trade I'm willing to make).
June 1, 2007 8:55:19 PM

Quote:

All that said, we need to build more nuclear plants. It's the most effective way to add a large amount of extra energy with no carbon emissions (they produce water... and spent radiocative material). Wind and solar are good in theory, but can only produce enough energy to replace coal on a massive scale (I mean, wind will be able to the power the US only if we convert Wyoming, Utah, and Kansas into wind farms -- a trade I'm willing to make).


hehehehe

Water and spent radioactive material... it's green! (glowing green that is). Eh, we need some serious improvement in nuclear reactor technology and policy before we can build any more of them in USA. That and we need to get a handle on all the nuclear waste we already have sitting around in leaky barrels...

I am also in favor of turning all of those states into wind farms. They should still be able to grow corn and wheat, right?
June 1, 2007 9:28:24 PM

I agree with your comment about Nuclear power, and trust me it's on its way. The Nuclear Regulatory Commision the past few years has been saying that there is going to be a boom in nuclear power plants (we haven't had one open in 11 years).

I disagree that we need improvements before building new ones in the US, the regulations are so strict that we will never have a disaster like chernobyl and the worst case senario is three-mile-island which resulted in no deaths and no injuries. If new one's can be built then old ones can be decommissioned and I believe the biggest risk comes from old reactors.

I could type for hours regarding my opinion of future energy, and i'm qualified to do so from a science point of view. however, it seems the larger factor in future energy is more political than anything so talking science is a moot point.
June 1, 2007 11:45:11 PM

Quote:
I agree with your comment about Nuclear power, and trust me it's on its way. The Nuclear Regulatory Commision the past few years has been saying that there is going to be a boom in nuclear power plants (we haven't had one open in 11 years).

I disagree that we need improvements before building new ones in the US, the regulations are so strict that we will never have a disaster like chernobyl and the worst case senario is three-mile-island which resulted in no deaths and no injuries. If new one's can be built then old ones can be decommissioned and I believe the biggest risk comes from old reactors.


Have you ever heard of Hanford? It's an ongoing radiological disaster in eastern washington. Radioactive contamination is seeping into the groundwater near the Columbia river as we speak (they have wells that theoretically keep the amount of contamination that gets into the river at "safe" levels). A multi-billion dollar glassification plant project has been bungled multiple times (for instance: it wasn't supposed to be a multi-billion dollar project when they started it). The Yucca Mountain project is also encountering serious hurdles. Three Mile Island was never really investigated to actually quantify the number of injuries it may have caused, was it? They transported open containers of radioactive materials on barges down the river and didn't sound any alarms to warn people to stay indoors and no one was hurt? Highly unlikely. Hanford was investigated: http://www.cdc.gov/nceh/radiation/brochure/profile_hanf... Fission reactors create nuclear waste and the USA has never had any kind of a real working plan to dispose of nuclear waste safely. Until the mess that has already been made has been cleaned up and the capacity to handle more reactors has already been established any rational person should appose any kind of new nuclear reactor projects. Meltdowns aren't the only potential harm nuclear reactors can due. Their normal operation creates health and environmental risks that must be properly mitigated. Quite frankly even once they glassify it I'm not entirely convinced that putting it under a mountain and hoping people forget about it is a good solution.

I lived near Hanford for several years. The mascot of Richland (a nearby state-created city that has incorporated, part of the "tri-cities") Highschool is "The Bomb". I kid you not. They have a model ICBM on campus and mushroom clouds and similar images adorn school paraphernalia. http://www.richlandbombers.org/ That's how desensitized they are to it. They seem to think it's cute in rural USA but somehow I don't think people in Japan would find it amusing that USA schoolchildren are still celebrating the greatness of the weapon that was manufactured at Hanford and dropped on Nagasaki. http://en.wikipedia.org/wiki/Hanford_Site

There are several huge changes that need to be made technologically, politically, and socially before nuclear programs can proceed with any semblance of safety in USA.
June 2, 2007 1:46:54 AM

Quote:
I agree with your comment about Nuclear power, and trust me it's on its way. The Nuclear Regulatory Commision the past few years has been saying that there is going to be a boom in nuclear power plants (we haven't had one open in 11 years).

I disagree that we need improvements before building new ones in the US, the regulations are so strict that we will never have a disaster like chernobyl and the worst case senario is three-mile-island which resulted in no deaths and no injuries. If new one's can be built then old ones can be decommissioned and I believe the biggest risk comes from old reactors.


Have you ever heard of Hanford? It's an ongoing radiological disaster in eastern washington. Radioactive contamination is seeping into the groundwater near the Columbia river as we speak (they have wells that theoretically keep the amount of contamination that gets into the river at "safe" levels). A multi-billion dollar glassification plant project has been bungled multiple times (for instance: it wasn't supposed to be a multi-billion dollar project when they started it). The Yucca Mountain project is also encountering serious hurdles. Three Mile Island was never really investigated to actually quantify the number of injuries it may have caused, was it? They transported open containers of radioactive materials on barges down the river and didn't sound any alarms to warn people to stay indoors and no one was hurt? Highly unlikely. Hanford was investigated: http://www.cdc.gov/nceh/radiation/brochure/profile_hanf... Fission reactors create nuclear waste and the USA has never had any kind of a real working plan to dispose of nuclear waste safely. Until the mess that has already been made has been cleaned up and the capacity to handle more reactors has already been established any rational person should appose any kind of new nuclear reactor projects. Meltdowns aren't the only potential harm nuclear reactors can due. Their normal operation creates health and environmental risks that must be properly mitigated. Quite frankly even once they glassify it I'm not entirely convinced that putting it under a mountain and hoping people forget about it is a good solution.

I lived near Hanford for several years. The mascot of Richland (a nearby state-created city that has incorporated, part of the "tri-cities") Highschool is "The Bomb". I kid you not. They have a model ICBM on campus and mushroom clouds and similar images adorn school paraphernalia. http://www.richlandbombers.org/ That's how desensitized they are to it. They seem to think it's cute in rural USA but somehow I don't think people in Japan would find it amusing that USA schoolchildren are still celebrating the greatness of the weapon that was manufactured at Hanford and dropped on Nagasaki. http://en.wikipedia.org/wiki/Hanford_Site

There are several huge changes that need to be made technologically, politically, and socially before nuclear programs can proceed with any semblance of safety in USA.


If remember my history Japan attacked the U.S. first and the bombs we dropped on them were not ICBM's . If we sent an ICBM not much of Japan would have been left. Nuclear power can be and is safe. How we dispose of the waste, now that is another story. Would you rather we continue to burn coal and natural gas?

In my area of the wonder state of Maryland we have good old constellation power with two wonderful power plants that burn coal. In that area around the power plant they have a very high cancer rate. The company is expected to spend billions to clean it up. Too little too late IMO. We also have a nuclear power plant at Calvert Cliffs that has been there since the 1970's and there has been no negative effects to the envoronment from that plant and they are even talking about building another plant in that area. If you want to save power. Turn off your lights, TV etc that are not in use. Hello people this is nothing new. Remember the energy crisis of the 70's? If not read about it.


http://cr.middlebury.edu/es/altenergylife/70's.htm
June 2, 2007 3:20:52 AM

Er... a bit off-topic, guys? :wink:

Really would be very nice for a PSU comparison and guide. I just have no idea how to find a good one to run my computer, and it's tempting to go to generic. Tom's?
June 2, 2007 7:26:37 AM

Quote:

If remember my history Japan attacked the U.S. first

AFAIK no one celibates pearl harbor so I really don't see what your point it. I don't care "why" it was done, it was horrific and not something to be proud of. It's really not a very good justification for nuking two cities either ("they wouldn't surrender" being the generally accepted excuse for the first one. "but we made two of them" being the pathetic excuse for the second one). Pearl harbor was a military base that was attacked by an, albeit large and unexpected, conventional military force using conventional tactics and principals. Hiroshima and Nagasaki were major cities full of civilians that were flattened. Huge difference. "But he hit me first"? We're talking about nukes here, not playground fights.

Quote:
and the bombs we dropped on them were not ICBM's.

...ok? Obviously you don't drop ICBMs you launch them and I said drop so again... what's your point? The ICBM I referred to was a model at a high school campus of a public school. Think "planet of the apes". It's bizarre.
Quote:

If we sent an ICBM not much of Japan would have been left.

you're confusing delivery method with payload
Quote:

Nuclear power can be and is safe.

It certainly could be if it were done properly.
Quote:

How we dispose of the waste, now that is another story.

No. No it isn't. It's the same story. You cannot have a safe reactor without a safe waste disposal system. They are one and the same problem.
Quote:

Would you rather we continue to burn coal and natural gas?

A good point in itself, but a fallacy of dichotomy as presented here.

Ok, I promise not to talk about nukes any more in this thread... :oops: 

Turning off an appliance when you're not using it: very reasonable idealogy
Deciding to not use an appliance at all when you want to because of an energy crunch: sucks
Being too lazy to turn off an appliance when you're not using it: unconscionable
Making energy efficiency a criteria when purchasing new appliances: also very reasonable
Putting pressure on governments and corporations to produce cheap, clean, renewable energy: need a lot more of this so we can all get back to having fun, which is what electricity is supposed to be all about
June 2, 2007 3:13:48 PM

Quote:
Quote:

If remember my history Japan attacked the U.S. first

AFAIK no one celibates pearl harbor so I really don't see what your point it. I don't care "why" it was done, it was horrific and not something to be proud of. It's really not a very good justification for nuking two cities either ("they wouldn't surrender" being the generally accepted excuse for the first one. "but we made two of them" being the pathetic excuse for the second one). Pearl harbor was a military base that was attacked by an, albeit large and unexpected, conventional military force using conventional tactics and principals. Hiroshima and Nagasaki were major cities full of civilians that were flattened. Huge difference. "But he hit me first"? We're talking about nukes here, not playground fights.

So that is a nice statement "you really don't care why" So that makes me think you really don't know what your'e tallking about


Apparently you need to go back and read your history books. The bombs where dropped to save lives. It was done it worked and that is it. If we invaded Japan we would have lots over a million men. It is sad to think that political correctness has gotten in the way of history. I can see in another hundred years liberal political correctness will have twisted history around and have us starting the war with Japan and not the other way around. Go read some books on the war and the atrocities that Japan commited during the war on the Chinese, Australia, and American troops during the war. Some of the treament was equal to that of the Nazsis.

I have relatives that fought in the pacific during WWII and they stormed the beaches with their comrades and they can tell you the truth about what happened. The Japanese fought to the last man in many of those battles. Do you really think that invading their country would have been a piece of cake?

I know what an ICBM is and I know that several war heads are attached to it so yes an armed ICBM is one powerful weapon and the little atomic bombs dropped on Japan pale in comparison.


And to stick with the topic of the thread....yes do a nice report on power supplies have the companies start to lable them just like my applicances so know what kinds of power consumption they have.


And i'm outta here :wink:
a b U Graphics card
June 2, 2007 11:45:56 PM

Based on the info in this review, we would need a 730W PS just for the watts needed by the RAM, video card, hard drive, and an empty PCI bus?

RAM 200
GPU 240
HD 145
PCI bus 145
Total: 730

And if you add in the watts needed for the motherboard, CD/DVD, fans, and CPU, then we would be over 1,000.

What am I missing here? :roll:
June 3, 2007 12:40:02 AM

That's watts on the DC side... What are you actually drawing at the wall? I take your point however, a 60% efecient PSU will always have a higher draw than an 80% effecient psu. Some of the better psu's are now getting 90% effeciency.
a b U Graphics card
June 3, 2007 3:40:54 AM

Quote:
Based on the info in this review, we would need a 730W PS just for the watts needed by the RAM, video card, hard drive, and an empty PCI bus?

RAM 200
GPU 240
HD 145
PCI bus 145
Total: 730

And if you add in the watts needed for the motherboard, CD/DVD, fans, and CPU, then we would be over 1,000.

What am I missing here? :roll:

roflmao :lol:  :lol:  :lol:  :roll: :roll: :roll:

Each of those are SYSTEM power requirements not requirements for each component. Man if RAM used almost as much power as a gpu...
a b U Graphics card
June 3, 2007 6:48:21 AM

Quote:
Based on the info in this review, we would need a 730W PS just for the watts needed by the RAM, video card, hard drive, and an empty PCI bus?

RAM 200
GPU 240
HD 145
PCI bus 145
Total: 730

And if you add in the watts needed for the motherboard, CD/DVD, fans, and CPU, then we would be over 1,000.

What am I missing here? :roll:

roflmao :lol:  :lol:  :lol:  :roll: :roll: :roll:

Each of those are SYSTEM power requirements not requirements for each component. Man if RAM used almost as much power as a gpu...

LOL! I figured I was missing something obvious, even though I read it carefully! I at least figured that it made no sense that an empty PCI bus would draw 145 watts. 8O :oops: 

So, they were measuring total wattage used to show the differences between what the different RAM HD and PCI card types used? That makes sense.

Oh well....
June 4, 2007 5:51:17 AM

Cute, but they didn't do one with a solid-state drive :twisted:

I ran my Toshiba Libretto off of a CF card and a CF to 44-pin IDE adaptor from ebay, plugged right in, zero vibration handheld PC, pretty good battery life too ;) .

I will need to check on the sort of tricks that where pulled on the CPU front, but I bet a Lima core @ 1000-1400mhz will rule for most mainstream tasks and be pretty darn low power while doing it. Second choice is a Geode NX (AKA super low "embedded" power Athlon, Socket A, 14 watts).

On the Graphics scene, try a 7900GS card, it is the 7900 die (smaller, less transistors) with 4 pixel pipes and 1 shader turned off, AKA just as much "rendering power" as a 7800.

Less power and better clocks too, also happens to be a steal at $120-150 last I checked.

PS, try an 8600 for a direct comparison of 7 to 8 series hardware on a fairly level performance field.

The 6600GT is well known as the best performance/heat/power/price ratio 6 series, don't pull 6800 crap on us.

If you want a properly set up power supply, cut the middlemen and just charge a 12-volt battery with a 12-volt solar panel and use a PC-Power 12v PSU, that way the motherboard will house all the inefficiencies as it takes 12 volts.

Middlemen in this case being Fossil Fuels that came from trees that "ran" on solar power, and the gas burning plants that convert this to electricity, and the 220v/110v distribution (not to mention the ultra-high-voltage lines for coast-to-coast power distribution), and then conversion back into DC 12-volts.

Better yet, find a stream and set up a water wheel with a 12-volt power generator, that will really fix the power problem.

I am thinking that a LED backlit or OLED display and a low power AMD or Cyrix/VIA C7 would be able to run usefully off of some solar panels, maybe try that instead of the "solar wii" joke.

I am serious, show me the money, use Benheck.com for guides on how to put LED backlights into an LCD monitor, then get a low-power processor, then challenge folks on the street to do their normal day-to-day tasks on the PC.

For a real kick, modify a motherboard to cut down on the voltage to components not under heavy use, underclock components that aren't a bottleneck (AMD hypertransport comes to mind, in fact the AM2+ is supposed to have power states and lane shut-off asynchronously up and downstream for this very reason, not that you can saturate 8x/8x links, much less the 16x/16x links that it has right now).

DDR3 anyone? Isn't this supposed to make the whole setup better? I am waiting for the new Motorola Razr 6 with integrated 800mhz Fusion and 512MB of RAM and 16GB Samsung NAND chip. XP on my phone!! That would be sweet.
June 4, 2007 6:21:55 PM

Well, I finally got myself a Watt meter, and I've done a some measuring of my two computers' power consumption. First, the systems:

A) Desktop
PD820 OCed to 3.36GHz, 1.35V
945P chipset
6600GT GPU
4x512MB DDR2 533MHz RAM
160GB WD HDD
400W ThermalTake PSU (XP480)

B) HTPC
X2 3600+, 1.9GHz, 1.1V
AMD690G chipset
2x1024 DDR2 800MHz RAM
320 Seagate HDD
430W SeaSonic PSU

Clearly, system A is going to have much higher power consumption. I built the HTPC from the ground-up to be low-power. Also, the PSU for the HTPC is higher efficiency.

So now, the results:

System A
off: 9W
idle: 148W
gaming: 180W (ish)
folding@home: 235W

System B
off: 7W
idle: 60W
folding@home: 80W

I'm amazed at how energy-efficient the HTPC really is. The CPU has a TDP of 65W, but I think it's pretty clear that I have it using much less than this, even at full load. After the PSU power efficiency is taken into account, the actual computer electronics are consuming no more than 64W. Next I'll be seeing how much I might save by trying 1.0V on the CPU. I've read that some have had good stability even at this voltage.

As for cost analysis, in Quebec it costs me just under 40 cents for every Watt-year I consume. By this figure, it costs me $94 for the DT and $32 for the HTPC to run them 24/7 for a whole year. These numbers ignore that my heating is electricity for at least 5 months of the year, so really running the computers is free, as the electricity would have been used anyway. I air-condition for 2-3 months, so during these times the computer heat costs me extra. For simplicity, I'll say these cancel each other (though I think analysis would show favor to the heating savings side).

I think when it comes time to replace my DT, I'll be looking for something more energy efficient than an overclocked Pentium D. Luckily for me, that's pretty much anything! :wink:
June 9, 2007 3:45:06 AM

I would have been intereseted in seeing the same setups running Windows Vista. They gave us idle and load usage, but does XP idle consume more or less than Vista idle? I know Vista carries a heavier processor load, but does this carry over into the idel time when you aren't interacting with the OS?
a b U Graphics card
June 9, 2007 3:47:58 AM

They did a test on that once b4, when vista came out.
June 9, 2007 5:04:04 AM

I have an overclocked Opteron 185 , 2gb DDR500, a 7900GT, 5x HDD's, an X-Fi Fatality with f'n RAM, and a thermal electric peltier air cooler tower, and a power hungry DFI Lanparty nF4 SLI-DR Expert UT that requires an ungodly amount of amperage just to boot. I calculated my actual draw to be close to watts at load. I was thinking about adding an HD 2900XT or two just to see if I can pull 1kilowatt at idle for fun.

At night my computer astrally projects itself into the rain forests of the world and personally kills cute animals and chops down trees.

Honestly though, I've had a low power system build on my mind lately. I want to get a next-gen PC built that wont blow out my Seasonic M12 700watter. I hear AMD and Intel will have 65nm motherboard chipsets in these next gen boards they have coming out. I hope that lowers the motherboard power draw because I have one of the most inefficient mobos ever made.
June 9, 2007 5:08:58 AM

Lower power in the 3x Intel chipset series? That's certainly not the impression I've been getting. Do they have a chipset coming out, like the G33 maybe, that's supposed to be easier on power?

I'm sticking with the Pentium D until Nehalem comes out. The heating is only a problem in the summer time, and Quebec power rates are very low. I want whatever I replace this computer to be at least 5x more powerful so I can really appreciate the money I've spent.
June 26, 2007 9:33:21 PM

I think the article misses the biggest power savings. When you aren't doing something demanding like gaming, turn off your gaming system.

For example right now reading this forum you could be using a Via processor that was consuming 2W (webpages aren't going to load it, except maybe certain flash animations). You could be running Win2k or a stripped down XP and get by with 256MB of memory for the majority of web, office, email, etc applications. Make it 512MB if you like to do multitasking. You could get by with integrated video for these purposes, and your whole system is going to use under 60W, even much less if you are hardcore enough to tweak the voltage/speed thresholds to find a sweet spot. Also keep in mind that when your power levels are this low there's no reason to have more than one low-RPM fan in the system, if even that (but it can be more compact, less time consuming and cheaper to keep one fan in it).

What does this cost? I picked up PCChips board with a C3 on it for $30 as a newegg refurb. Already had a KVM, already had a small PSU (but make that $20 if you like), and given such minimal cooling requirements it can be put into some old case unfit for a modern system due to poor cooling related design). Add a free old hard drive or if you wanted to get fancy, put a $50 DMA capable compact flash card and CF-IDE adapter in and use WinXPe's EWF to limit writing to the CF card. At this point the system has a real potential to be silent, not just inaudible. If you get tired of using it as a desktop and picked a tiny case you can always turn it into a car-PC later by buying the automotive 12V-input PSU for it.
August 22, 2007 7:55:38 PM

Why not compare peripherals of the computer like the Monitor, the mouse, Keyboard, USB Devices like mouse/keyboard etc.

I have always noticed the monitor is a big energy hog. I have a fairly large 19 inch Samsung Monitor and it is probably a big energy hog compared to a 17 inch LCD monitor. If the monitor turns itself off that helps some.
!