×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

The Power Consumption of Modern PCs

Zonk posted more than 7 years ago | from the keep-your-lights-off-to-compensate dept.

Power 122

janp writes "The power consumption of modern PCs has skyrocketed the past few years. Hardware.Info has done some fairly extensive research on the power usage of various configurations. It turns out the a high-end gaming rig can easily use more than 400 W, and that putting a system in stand-by isn't as saving as you might think. The article has some interesting tips to save on power costs."

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

122 comments

I agree (-1, Flamebait)

Anonymous Coward | more than 7 years ago | (#17905830)

PC's shouldn't consume power. That's my leftwing opinion.

Re:I agree (0)

Anonymous Coward | more than 7 years ago | (#17906022)

PCs should be made with their own coal-fired generators (nuclear reactors for the high-end). That's my rightwing opinion.

Re:I agree (0)

Anonymous Coward | more than 7 years ago | (#17906176)

PCs should be powered by a digester fueled by the blood of our enemies.

Re:I agree (0, Flamebait)

spun (1352) | more than 7 years ago | (#17907528)

If only government got out of the way, PCs could be powered by watch batteries. That's my libertarian opinion.

Re:I agree (1)

spun (1352) | more than 7 years ago | (#17910104)

If only people would stop saying things I don't agree with, I wouldn't have to mod them down. That's my slashdot moderator opinion.

Re:I agree (0)

Anonymous Coward | more than 7 years ago | (#17911216)

Humor, no?

Re:I agree (1)

spun (1352) | more than 7 years ago | (#17911414)

Yes, I was going to do a few more, but hey, if the halfwit mods can't take a joke, why bother?

If only the government made all our computers, we would all still be using a 486 that barely consumes any power at all. That's my communist opinion.

If everyone made their own power, and their own computers, this wouldn't be a problem. That's my anarchist opinion.

If we set up government programs to buy more efficient replacement computers for everyone we could reduce power consumption. That's my socialist opinion.

If only we FNORD the FNORD FNORD, we'd all FNORD. That's my discordian opinion.

And so forth.

FWIW, here's a little program for you. (0)

Anonymous Coward | more than 7 years ago | (#17911900)

Here's a little windows app that might help you save a little power. When your machine is idle, it throttles the clock back, which should save a bit of juice (and heat!) http://www.oldskool.org/pc/throttle/ [oldskool.org]

action meet consequence. (0)

tomstdenis (446163) | more than 7 years ago | (#17905910)

You mean buying the highest end power consuming gear raises the power consumption of the device as a whole?

NO WAI! Nice ad laden "story" though... [and yes, I flipped through it].

Tom

Conserving energy by not using applications (3, Funny)

Radon360 (951529) | more than 7 years ago | (#17905938)

I think the author of this article tried conserving energy by not using spell check.

Re:Conserving energy by not using applications (2, Funny)

D4rk Fx (862399) | more than 7 years ago | (#17907054)

He wsa jstu tyirng ot rerutn teh lteters to thrie natrual rnamdo staet ot icnresae entrp!oy

Get Laptops or smaller (5, Insightful)

Salvance (1014001) | more than 7 years ago | (#17905952)

If you're worried about power consumption, you're not going to buy a top of the line gaming rig. You'd probably buy a relatively low powered laptop (or even buy a very underpowered laptop similar to a OLPC machine). Gaming machines will continue to be bigger and bigger power hogs. More power consumption = faster and better gameplay, no way around it.

Re:Get Laptops or smaller (1)

gstoddart (321705) | more than 7 years ago | (#17906094)

If you're worried about power consumption, you're not going to buy a top of the line gaming rig. You'd probably buy a relatively low powered laptop
Except, in a short time, Vista is going to start causing everyone to need to upgrade to big, honking machines with high power consumption just to run the new interface.

Microsoft seems happy to be helping drive this trend towards ridiculous power consumption.

Cheers

Re:Get Laptops or smaller (1)

nxtw (866177) | more than 7 years ago | (#17906638)

Except Vista's new interface runs fine with all the fancy 3D effects on "underpowered" integrated Intel graphics solutions. When not charging the battery, and with the screen at full brightness and CPU at full speed, my laptop with the Intel GMA950 draws about 20-25w.

Re:Get Laptops or smaller (1)

mrchaotica (681592) | more than 7 years ago | (#17908496)

Microsoft seems happy to be helping drive this trend towards ridiculous power consumption.

Of course! The vast majority of Windows licenses are sold via OEMs. If people stop buying new computers (because the current ones are fast enough), how would Microsoft make any money?

Re:Get Laptops or smaller (3, Insightful)

tomstdenis (446163) | more than 7 years ago | (#17906172)

Not really. more crap in game IMPLIES more power, but compare say a 486 to a Core 2 Duo. The latter is much more efficient per MIPS than the former.

To put it another way, to match the power [in MIPS] of a typical 1989 486 desktop, you could do so with far less power consumption today. The problem is few companies write conservative software. Go ahead, make your application inefficient, a new cpu is always around the corner!

What people seem to forget is that we were doing word processing, vector graphics and all that on old school Mac IIs in the mid to early 80s. Those programs certainly didn't require hundreds of megabytes of ram or gigabytes of disk space. Of course people associate numerical requirements with quality. CPU has more megahurts? It must be better! Game needs a faster GPU? It must be awesomer! etc...

I'm personally impress with efficiency not bulkyness. Write me a competent word processor that fits on a floppy disk. That'd be a hoot.

Re:Get Laptops or smaller (1)

gfxguy (98788) | more than 7 years ago | (#17906958)

I agree - most of my apps run just fine on my machine, but some apps, like video editing, are still a nightmare. I'd like to upgrade and just have that app run faster, but you need a newer OS to take advantage of newer computer features (multi-core or more memory or SATA or RAID), which then requires an upgrade to the app, which, instead of bug fixes, has been fattened by unwanted extra features.

Try this site, though. [tinyapps.org]

Re:Get Laptops or smaller (1)

merreborn (853723) | more than 7 years ago | (#17911056)

What people seem to forget is that we were doing word processing, vector graphics and all that on old school Mac IIs in the mid to early 80s. Those programs certainly didn't require hundreds of megabytes of ram or gigabytes of disk space
Those applications also did far less. There's still a comparably light-weight word processing application on every PC. It's called notepad.

I'm personally impress with efficiency not bulkyness. Write me a competent word
processor that fits on a floppy disk. That'd be a hoot.
By sacrificing efficiency, you can develop software faster, cheaper, and with more features. The market at large has opted for this over efficiency. We want all the bells and whistles, we want them now, and we want them for $19.95. We don't give a damn if it fits on a floppy. Hell, Apple hasn't even sold a computer with a floppy drive in the better part of a decade.

Efficiency isn't free. And most people aren't willing to pay for it.

Re:Get Laptops or smaller (1)

MobyTurbo (537363) | more than 7 years ago | (#17912750)

Write me a competent word processor that fits on a floppy disk. That'd be a hoot.
CP/M and early PC versions of Wordstar would fit on a single floppy. That task has been done dozens of times before PC-XT-compatibles introduced people to the hard drive. (Which was a good thing, because floppy drives have awful access times too.)

Re:Get Laptops or smaller (1)

habig (12787) | more than 7 years ago | (#17913374)

I'm personally impress with efficiency not bulkyness. Write me a competent word processor that fits on a floppy disk. That'd be a hoot.

Word Perfect v5.1. Could run from a 5.25" floppy, and had was featureful enough to write whole books or court documents with.

Of course, it was also the last major application written in assembly, so it also had to be a maintenance nightmare, but from a user standpoint it was pretty cool. Wrote a PhD thesis with it (shoulda used latex, oh well).

Re:Get Laptops or smaller (1)

tomstdenis (446163) | more than 7 years ago | (#17913466)

Yeah I was talking about ClarisWorks for the Mac in my post. From what I recall it wasn't huge [heck the macs only had like 80MB HDs at the time] but had all the nice on screen editing, fonts, drawings, and all that crap that the average Word user needs. Sure it didn't come with 3000 fonts or a paperclip but it was damn spanky and I'd use it over Word anyday.

As for LaTeX part of it's hugeness (other than the binaries which weigh in at several megabytes) are the scripts and metafonts. A TeX distro which used a zip container to hold all the files would probably save a decent amount of space.

I probably would want to run TeTex on a PC from the early 80s, but I think it's doable (well I know it is, since TeX was written in the 70s...)

Re:Get Laptops or smaller (1)

Sigma 7 (266129) | more than 7 years ago | (#17914192)

What people seem to forget is that we were doing word processing, vector graphics and all that on old school Mac IIs in the mid to early 80s. Those programs certainly didn't require hundreds of megabytes of ram or gigabytes of disk space
Word processors in the Mac II era were strict Word processors. There was no need to worry about anything more advances than basic documents.

Nowadays, word processors are expected to have:
- A Spell checker, and grammar checker: This will detect mistakes you cannot sea.
- Font support: The old processors don't have that "scribble" font that's in extremely high demand.
- Anti-aliasing support: Makes the fonts look nicer, and print out better.
- Printer profiles: For getting the colours on your printer correct.
- Forign language/unicode support: Effectivly doubles the size of the application.
- Support for many document types: Interoperability.
- Automatic text: For Table of contents and index.
- VBM support: For compatability with Word Macro viruses.
- Online Documentation: Informes pirates on how to use the product, complete with pictures. (Otherwise, the product lauded for lack of documentation.)
- Et Cetra: For things that nobody has thought of yet.

None of these would individually cause the word processor to become very large. However, the market demand wants everything put into one neat package, resulting in extremely large word processing packages.

What people seem to forget is that we were doing word processing, vector graphics and all that on old school Mac IIs in the mid to early 80s. Those programs certainly didn't require hundreds of megabytes of ram or gigabytes of disk space.
The scale, however, has increased. Many people wouldn't consider making a very large image, such as >2048*1536, but such an image managed to max out the memory on a 256MB system. (This was in MSPaint, and yes, the system did run out of memory.)

Re:Get Laptops or smaller (1)

tomstdenis (446163) | more than 7 years ago | (#17915992)

You're obviously very young or very sarcastic or very ignorant or all three...

ClarisWorks had a spell checker. I don't recall if it did grammar, but we were in school at the time. We were expected to correct our own grammar. It had "font" support in that any installed font could be used in a document. True, it didn't embed fonts, but it wasn't expected that there would be a huge market for addon fonts.

Anti-aliasing wasn't a big requirement given that we were using DOT MATRIX PRINTERS. Not exactly grace. As for printer profiles, you could use any printer that was install in the system under the control panel. That included dot matrix, bubble jet and lasers.

Unicode support shouldn't double the size of the application if you just design the application with that in mind first. UTF-8 for instance is very easy to support since it has ASCII as a SUBSET of the standard.

blah blah...

I don't few a lot of the new features as real features. If I wanted automatic text I'd use LaTeX and perl (oh wait, that's what I do). Not only is it easier but it's infinitely more customizable.

Point is, we were using a word processor for what most people today use a word processor. And yet, somehow, we didn't consume 400 Watts of energy in the process. Didn't need 200GB disk drives or multiple GBs of ram.... hmmm.

Tom

Re:Get Laptops or smaller (1)

Sigma 7 (266129) | more than 7 years ago | (#17916476)

You're obviously very young or very sarcastic or very ignorant or all three...

ClarisWorks had a spell checker. I don't recall if it did grammar, but we were in school at the time. We were expected to correct our own grammar. It had "font" support in that any installed font could be used in a document. True, it didn't embed fonts, but it wasn't expected that there would be a huge market for addon fonts.
I've used Speedscript for the Commodore 64, Wordperfect 4.1.2 for the Amiga, and the latest version of Wordperfect. Speedscript supported one font.

Wordperfect 4.1.2 fit on a pair of floppies and supported basic word processing features (spell, grammar, etc.) It ran on a 640x200x4 monitor (or 640x400x4 interlaced), and use a fixed-width font.

The latest version of Wordperfect runs on 1600x1200, with a real-time print preview with a variable-width font and plenty of other features that run while you type. Even if this doesn't require the latest and greatest system, it is not recommended to run this on what amounts to an Amiga 500.

If I wanted automatic text I'd use LaTeX and perl (oh wait, that's what I do). Not only is it easier but it's infinitely more customizable.
If you have to look something up, have to spend a year studying how to use the product, or have to personally write it, it's not easier. It's just like saying that rolling your own 3D-Modelling software is easier than using a prefab application such as Blender.

Re:Get Laptops or smaller (2, Insightful)

Banzai042 (948220) | more than 7 years ago | (#17906284)

Not always, granted if you want an 8800GTS/X you'll need a big power supply, but if you go with something like the upcoming 8600 Ultra (not the highest end, but should still have some pretty good performance in dx9 games), which has no PCI-e power connector (draws enough power from the slot), just about any C2D cpu, a single optical drive, and a single hard drive, you'll have a machine that's pretty light on power use. Just because it's possible to get a machine that needs a 750 Watt PSU doesn't mean that's your only choice in a desktop, even for gaming. It's all in the components you choose.

Re:Get Laptops or smaller (1)

kaaona (252061) | more than 7 years ago | (#17910384)

...or build your own system using energy efficient components. A year ago I put together a forensics workstation with a dual-Opteron mobo, 4GB registered DDR, and ten SCSI drives (1.1TB array). I used ball bearing fans throughout and a high-efficiency PCP&C p/s. My only concession to gaming was a pair of 6600 cards -- the only significant heat sources in the enclosure. At power-on the hard drives do a staggered spin-up. My Kill-A-Watt meter peaks briefly at 450W, then settles back to 330W. Running three virtual machines plus the host, it's effortlessly fast, reading room quiet, and very power efficient.

Build your own (1)

SanityInAnarchy (655584) | more than 7 years ago | (#17910806)

I'm starting to find all kinds of limitations with this box, but it's still worth mentioning. Built summer before last, it's a socket 747 or somesuch -- 1.8 ghz Athlon64 (stupidly bought because I can upgrade it to 2.7, was mostly stable at 2.4, but really only rock solid at 1.8), pair of 250 gig hard drives in RAID 0, 2 gigs of DDR 400, GeForce 6600.

There's some vibration now that I'm trying to kill, but before that, it was as quiet as any water-cooled rig. Got a nice cool, quiet power supply, and the video card is passively cooled (heatsink!) despite being a nice PCI Express. Stock cooling all around, a nice big ThermalTake case with all kinds of fans -- most of them fairly large.

All in all, it's still a nice, solid gaming rig, and I could probably build another one today with similar power consumption -- I believe the power supply is only some 400w. The trick is to stick just behind the curve, in the sweet spot, where things cost half as much (I'm not making that up!) for maybe 15% less performance, and where they've figured out how to make it cooler, quieter, and more efficient.

After all, we all remember the leafblower video cards, which took up a whole PCI slot just for exhaust. The absolute bleeding edge is pretty much always going to be less efficient, and also somewhat cobbled together. The sweet spot is where it's at.

Or get all-in-one units... (1)

FellowConspirator (882908) | more than 7 years ago | (#17912910)

Another possibility is getting all-in-one PCs or some of the mini-form-factor PCs. All-in-ones are typically redesigned laptops with the battery support removed. A good example is the Apple iMac, even the 24"LCD model peaks around 120W when pushing the CPU and GPU at 100%, thrashing the disk, and powering the display at full brightness. Many of the MicroATX kits use low-power components too. There's a British company that makes UPS solutions where the PC can, via USB control, turn-off some of the outlets on it. You could automate the process and cut power to peripherals when not in use and power them up when needed. Hopefully something like it will be available in the states (the British company doesn't make 120V models).

No Kidding (1)

spotter (5662) | more than 7 years ago | (#17906076)

my 2 machines were easily averaging 500W. I thought a $140 electricity bill in the winter was a bit high, so bought a kill-a-watt and figured out that my computers were consuming 12 of the 18 KWH my apt was using a month. Now they are turned off except when I need them, and thanks to Wake-On-Lan, I can turn them on remotely as well.

Re:No Kidding (0)

Anonymous Coward | more than 7 years ago | (#17906466)

Uh, that seems a bit off. [Yeah, I'm going to assume you get a monthly electric bill]

$140/month for 18 kilowatt-hours? That's roughly 777 cents per kilowatt-hour. The worst I've heard is electricity at approximately 20 cents per kilowatt-hour.

Re:No Kidding (1)

ThosLives (686517) | more than 7 years ago | (#17906748)

I think he meant 18 kW-h/day, not for the entire month; that would be about 540 kW-h/month, which is only about $.27 per kW-h; probably not too far off considering taxes and the typical flat-fee "customer charge" that many companies bill even if you used 0 energy.

Re:No Kidding (1)

spotter (5662) | more than 7 years ago | (#17908476)

18 KWH a day. though the math was off a little. it was about 130 for electricity (20+ for gas) and it was a long month due to the holidays. and there were various taxes and fees that raised it. Though this bill it was actually 19KWH a day.

Re:No Kidding (1)

gEvil (beta) (945888) | more than 7 years ago | (#17906556)

Now they are turned off except when I need them, and thanks to Wake-On-Lan, I can turn them on remotely as well.

Just to clarify, your machines aren't actually turned off, are they? I was under the impression that WoL needed the machines to either be in Standby or Suspend mode, which means they're still on to some degree (though using much less power than a fully "awake" machine).

Re:No Kidding (3, Informative)

alexhs (877055) | more than 7 years ago | (#17906890)

They're as 'off' as an ATX can be 'off' with the power supply switch to 'I'. You need a cold boot, but of course you still use some power. It's G2 state [wikipedia.org]

Re:No Kidding (4, Informative)

ashitaka (27544) | more than 7 years ago | (#17907006)

No, they can actually be fully "shut down".

On a modern PC with a built-in motherboard you will notice at least one lit LED on the motherboard as long as the PC is plugged in. A tiny amount of power is being provided to the network adapter to listen for "magic packets" which, after being verified, will cause the machine to power up as if you pressed the power switch. This could be from standby or suspend but a cold boot is also possible.

No your math is wrong (2, Informative)

Darkfred (245270) | more than 7 years ago | (#17906632)

Your 2 pcs at 500w are averaging between 4 and 6c an hour. At full load without power saving, and turned on 24:7 the worst case scenario is $30-$40 a month. In a real world situation this would probably average around $15 dollars a month.
An actually meter on my computer (150 watt power supply, with power saving features) showed that I was averaging around $8 a month.

On the other hand, your "energy saving" refridgerator will cost many times this amount. Mine averages around $70 a month worth of electricity.

You should pick up a meter from home depot, you plug it between the computer and the wall it has a small window with a dial ticking off the KWh.

Re:No your math is wrong (follow up) (4, Insightful)

Darkfred (245270) | more than 7 years ago | (#17906830)

Obviously I bought a new refridgerator. A 10 year old refridgerator is just not efficient anymore.

But the scariest thing I found during my power audit was that each incandescent lightbulb was taking more power than my computer at rest. A single chandelier in my house accounted for 1/4 of my electrical bill.
By replacing all the lightbulbs with compact flourescent I was able to shave a 3rd off my monthly bill. (still quite high because of an old ac system).

In conclusion your computer is such a minor contribution to electricity that you shouldn't even be considering it before you fix the big offenders.

Re:No your math is wrong (follow up) (1)

Overzeetop (214511) | more than 7 years ago | (#17907604)

That's insane - you either have few lights in your house and gas heating and hot water, or that chandelier was lighting a football field. And your math may be bit off. If your lamp accouted for 25% of your costs, and you replaced the lamps with fluorescents at the same light output (75% energy savings) your electric bill should have only dropped by (3/4 of 25, carry the 2,...) 18.8%, or 1/6 of your bill. Still, to have a single fixture account for 25% of your electric bill is amazing. At my office - about 1300SF, the lighting accounts for 25% of my electric. Now, that's with 32W T8 lamps, but the light level is easily 2x that of a residential building, is on 10-11 hours per day, and lights the entire space. And my electric bill is only about $120 at the office. To do that in a single incandescent fixture would require 900W of lamps, run 12 hours a day, 30 days per month. Yikes.

Re:No your math is wrong (follow up) (1)

Bishop (4500) | more than 7 years ago | (#17907836)

Except a computer that is running 24 hours a day is not a minor contribution to your electricity consumption.

Re:No your math is wrong (follow up) (0)

Anonymous Coward | more than 7 years ago | (#17909954)

How much electrity do you use per year? 10000KWh? 20000KWh?

I recently reduced power consumption of my server, which runs 24/7 by 20W (100W=>80W). That saves about 6.7% of my yearly bill. And now you tell me cutting power usage from 400W to a typical 70W for a laptop under load is not worth doing? :)

> In conclusion your computer is such a minor contribution
> to electricity that you shouldn't even be considering it
> before you fix the big offenders

Now do the maths on the high end system they cite in their test. 186W when idle. My laptop uses about 30W when idle and playing Neverwinter Nights 2 is more than possible with it. Thats 3,7KWh/day in difference. Even with 10000KWh/year of power (which I think is waaaaay too much for a normal household), you "only" use ~27KWh/day. Those 3,7KWh in difference between the laptop and the high end system will save you 14% of your bill. Plus the money you save when buying a normal laptop instead of a high end system.

Re:No your math is wrong (follow up) (2, Insightful)

HappyEngineer (888000) | more than 7 years ago | (#17911328)

Using a kill-a-watt I found out that my computer draws 600W. (3 monitors, dual cpu, high end video card, 4 hard drives, 8 fans)

I was able to reduce my power bill from $250/month to $100/month by turning it off every night.

The upshot is that people should buy a kill-a-watt and find out what the big offenders are. Guessing probably won't work.

What are you guys DOING? (2, Insightful)

potat0man (724766) | more than 7 years ago | (#17912622)

These power bills make me cringe. $100??? $300!?!?!

My one bedroom apartment with its occasionally used dishwasher, electric stove, fridge/freezer, 4 or 5 LED lightbulbs, 25"tv, router, modem, cell phone charger, electric razor, gaming pc & work pc costs me between $16-$24/month.

This is in Albuquerque, NM. I am pretty efficient. I...

1. Never leave things on when not using them and have everything plugged into power strips conveniently placed on TOP of my desks/tv stand so it's easy to flip off the switch so no standby power is ever wasted. It's just a habit now: Shutdown the pc, flip the power switch.

2. Use LED Bulbs

3. Have gas heat/hot water. Don't need AC. The air naturally circulates in the summer due to a heat chimney, plus my part of the building is shaded.

4. Keep my fridge/freezer packed to the brim with old milk jugs full of water.


I've never NOT done any of these things so I don't know what the bill would be if I didn't. Perhaps I should bite the bullet some month and try it out. But I can't imagine having a $100 electric bill. You guys all have hot tubs and the fanciest christmas displays in the county or what? I don't know if my bill would get that high even if I left everything on 24/7 for the entire month.

Am I an anomally? Is my meter broken? I don't know how I'd even get the bill that high if I wanted to.

Re:No your math is wrong (1)

poot_rootbeer (188613) | more than 7 years ago | (#17907130)

On the other hand, your "energy saving" refridgerator will cost many times this amount. Mine averages around $70 a month worth of electricity.

My total electric bill rarely climbs above $70/mo, and yes, my apartment does have a refrigerator in it (and not a particularly "green" one, either).

There must be some other factor at play.

Re:No your math is wrong (1)

Darkfred (245270) | more than 7 years ago | (#17907418)

Like I said in my follow up, the other major offender in my house (besides AC and refridgerator) was incandescent light bulbs. A single 4 bulb chandelier will take more electricity than a massive 300W power supply at full draw and 100W monitor. My biggest savings in doller/per KWh was replacing all my light bubls with 10 and 15watt compact flourescents.

I had an 8 bulb chandelier which turned out to be the single most expensive device in my house, who would have guessed.

Re:No your math is wrong (1)

apathy maybe (922212) | more than 7 years ago | (#17914190)

Mind you, having 8 100W light bulbs is just silly even if you weren't concerned about the power costs. How big is this room you are trying to light? You could always have used 8 60W (or smaller) incandescent light bulbs (not that I am suggesting that you do, fuck that use the more modern better tech).

Re:No your math is wrong (1)

spotter (5662) | more than 7 years ago | (#17908364)

uh. This is how I computed it.

Con Ed tells me I am averaging 18 KWH a day.

Con Ed charged me $150 for electricity (17-18c a KWH)

Computers are using over 500W (kill-a-watt, energy reader tells me this) steady if both are on (i.e. at least 12 KWH a day)

hence, computers were using at least 2/3 of my daily electricity.

Re:No your math is wrong (1)

Darkfred (245270) | more than 7 years ago | (#17908556)

Even at those insane rates, and assuming your computers ALWAYS use their power supplies maximum rated power (technically impossible) that is still only $60 a month. Most computers only draw half their power supply rating and even less when in power saving mode. Do you really leave the monitors on all day?
Kill-o-watt's spot metering is just not accurate for computers because their power consumption varies by 75% with simple application changes.
But my guess is that you don't have kill-o-watt, you are just using the power supply ratings (2x250) for the sake of argument.

Re:No your math is wrong (1)

spotter (5662) | more than 7 years ago | (#17912630)

No I have a Kill-A-Watt (bought from newegg 2 months ago).

There are 2 problems with your power supply argument

1) the power supplies take in more watts than they output (they are not 100% efficient)
2) My power supplies are both 400W antec power supplies.

I would guess power supply 1 is near it's max (dual athlon, radeon 9800, 8 HDs), while the other machine is probably using around 100-150W (single athlon, el cheap video card, 1 HD)

When I said how much it was using, it was the steady reading given by the kill-a-watt when the machines were doing nothing.

Re:No your math is wrong (1)

Darkfred (245270) | more than 7 years ago | (#17912806)

That is pretty scary. Its possible that your power supplies are just throwing away the surplus rather than balancing it. With the setup you describe a 300w power supply should still be sufficient. I am running a single core athalon w/radeon 9800 and 4 HDs. In idle (but not power saving mode) i am hitting around 120-150W, my powersupply is underpowered, but not by much. While running battlefield at full res I occasionally brown out. (reboot)

modern PCs or gaming PCs? There's a difference (2, Insightful)

grommit (97148) | more than 7 years ago | (#17906344)

The summary mentions modern PCs but it seems to be about gaming PCs. Posting a story saying that gaming PCs take up a lot of electricity is pretty much stating the obvious.

I'd be more interested to see the power consumption differences between an off she shelf Best Buy computer of 5-10 years ago compared to one of today. Brick and mortar electronics stores are where a good majority of people buy their computers so as far as home computer power usage goes, that's what matters. I'd like to think that with components like sound, networking and video being put on the mainboard and the ability of major manufacturers to set machines to go into a sleep mode by default that computers of today would actually take up less power than those of yesteryear.

Not having any machine of that type around, I can't really do any testing unfortunately.

Re:modern PCs or gaming PCs? There's a difference (1)

Vellmont (569020) | more than 7 years ago | (#17907036)


I'd be more interested to see the power consumption differences between an off she shelf Best Buy computer of 5-10 years ago compared to one of today.

It's no secret that power consumption of PCs has gone up steadily. I'll bet you hard money that a Best Buy special made today is going to consume more power than a Best Buy special of 5 years ago. You might save a couple watts by having on-board LAN, but it's going to be more than taken up by higher electrical usage of the processor. As a real world comparison, I happened to plug in my old Circa 1996 PPro 200 to a power meter about a week ago. It ate up about 60-70 watts at idle. My Circa 2002 AMD XP2000+ machine with onboard sound, and LAN eats up about 110-120 watts at idle.

Re:modern PCs or gaming PCs? There's a difference (1)

toddestan (632714) | more than 7 years ago | (#17915838)

On the other hand, if you are going to look at systems, I would guess that the power saved from going from a CRT monitor to a LCD monitor is greater than the extra power consumed by the rest of the system. I'm going to guess that the worst offender is going to be a system from about 2-3 years ago. Likely to still have a CRT monitor, and either a P4/Celeron or an Athlon XP under the hood.

Re:modern PCs or gaming PCs? There's a difference (1)

mrchaotica (681592) | more than 7 years ago | (#17908698)

I'd be more interested to see the power consumption differences between an off she shelf Best Buy computer of 5-10 years ago compared to one of today.

Ten years ago, I had a Packard Bell 486 desktop (yes, purchased at Best Buy). I don't know how much power it used, exactly, but what I can tell you is that the CPU was passively cooled by a heatsink machined with cubes -- not fins -- about 2mm on a side. On my newer computers, even the RAM has bigger heatsinks (and therefore dissipates more energy) than that, let alone the chipset, GPU, or CPU.

Re:modern PCs or gaming PCs? There's a difference (1)

camperslo (704715) | more than 7 years ago | (#17910348)

The summary mentions modern PCs but it seems to be about gaming PCs.

That's for sure, the power consumption I see on a general-purpose Core 2 Duo desktop system built last fall (excluding display) maxes out at about 93 Watts, much lower than their examples.
That's including 4 Watts for when the USB Eye-TV Hybrid NTSC/ATSC tuner is active, and with both cores of the E6300 kept maxed out (BOINC client always running) and the 1.86 GHz CPU overclocked to about 2.25 GHz by pushing the FSB speed a bit.
The CPU runs cool with the stock Intel heatsink/fan. Fairly fast memory was used so the voltage to it did not need to be increased at the faster settings. It is using an AS Rock Conroe 945G DVI motherboard. There's a single 400 gig SATA drive, and just the GMA-950 graphics provided by the Intel chipset. It's definately not for serious gaming, but it works great for HDTV and general use. The OS X effects work.
The BOINC benchmark shows about 1800 MIPS floating point, 4780 MIPS integer (per core). After 4 months, BOINC combined stats show world rank at 94.25% for total credit, 99.1% for recent average credit (scores are even higher for the individual project).
I believe the MacBook uses essentially the same 945G/ICH7 chipset, except the mobile version is designed for lower FSB speeds.

Assembly was considered based on an Asus board with the Intel 975x chipset which could have handled overclocking the FSB (and thus the CPU) much higher than the 945G allows, but saving $200 on a cheaper board, avoiding the cost of a high-end graphics card, avoiding the cost of enhanced cooling, and keeping power consumption low were all factors.

The rule of thumb I use for power cost is $1 a month for every 10 Watts (continuous).
That is based on about $.14 per kilowatt/hr

Saving 100 Watts full time works out to $600 over 5 years!

I just did some research on this actually (5, Informative)

casualsax3 (875131) | more than 7 years ago | (#17906388)

A few weeks ago I tested some power supplies to see if it's worth spending $70 on a power supply vs the crappy stock PSU that comes with a lot of cases you can find on NewEgg.

I used Kill-a-Watt power tester, which can test for a number of things - I used raw amps.

I tested 4 machines with 5 power supplies in 0, 1, 2, 3, and 4 drive configurations. I also took a reading of how much power the systems drew when I powered them on at 4 drives, which shows how efficient the power supplies become under serious load (it takes a good chunk of power to spin up 4 drives)

The machines were all tested with the same 1x1GB PC5300 RAM, and the same four Western Digital SATA drives. The Intel systems were LGA775 chips on an Asus, and the AMD's were AM2 - also using an Asus motherboard.

Here are the results (hosted by Voxel.net, so it should hold :) http://newyorkhatesyou.com/Power_Supplies.pdf [newyorkhatesyou.com]

Power supplies tested: http://www.newegg.com/product/product.asp?item=N82 E16817256001 [newegg.com]

http://www.newegg.com/product/product.asp?item=N82 E16817371006 [newegg.com]

http://www.newegg.com/product/product.asp?item=N82 E16817151022 [newegg.com]

http://www.newegg.com/product/product.asp?item=N82 E16817234002 [newegg.com]

In a lot of cases the stock power supply uses almost twice as much power.

In Brooklyn I pay $.19c/kwh, so 1 amp of power can cost around $20 a month - ((volts * amps) / 1000 ) * time (in hours). This means pretty plainly, that the stock PSU here would cost me another $15 per month on my one desktop that I always have on.

Now if an office switches all of our workstations to one of the three 80% efficient power supplies, we stand to save a few hundred per month. Add to that the fact that these power supplies generally have more stable rails, and they should last longer - and its really a no brainer.

Re:I just did some research on this actually (0)

Anonymous Coward | more than 7 years ago | (#17907154)

Nice dataset. You make me glad I just bought an Antec, mainly on reputation. What are the units in the tables? "raw amps"? And what does the color coding mean? Green is lowest in test-case (row) / red is highest? Also, the 1/2/3/4 drive stack is interesting. Looks like about 0.06 - 0.08 (amps?) extra per drive, with one oddity: AMD 3800 X2, Silverstone jumps 0.15 from 1 to 2 drives. Can you toss it into a spreadsheet and give some graphic pictures?

Re:I just did some research on this actually (0)

Anonymous Coward | more than 7 years ago | (#17907188)

Nice test, but I'm doubting some things about the methodology/results:

You're using peak current usage @ drive spin up? I'm not sure that's consistent enough from time to time to be really accurate, and I really wonder how you've accurately measured such a peak (not as easy as it sounds - the results could be WAY off!) Besides, accuracy at peak usage isn't really representative of normal operating situation. Perhaps some are better optimized for higher loads, and are less efficient when idle.

And using efficiency @ peak usage to calculate power savings is just wrong, unless your drives are always spinning up or such (always high load).

Besides, electricity rates vary quite a bit (under $0.10/kwh here), so even with more conservative efficiency calculations the savings wouldn't be quite as much as 15$/month (for PCs that are on 24h/day). But then again, if you live in a warm place, heat is a big deal. The AC for a PC that creates too much heat (especially with the new video cards) likely costs me more than whatever difference most power supplies would make.

I just replaced a 300w halogen lamp with a 55w energy star torchiere. Not only I'm saving ~250w, but that also means far less heat, hence lower AC costs during all summer.

Re:I just did some research on this actually (1)

casualsax3 (875131) | more than 7 years ago | (#17907564)

Sorry, forgot to mention that all of the spin up measurements are the average of three startups, though I can tell you that the numbers never varied greatly between runs.

Re:I just did some research on this actually (0)

Anonymous Coward | more than 7 years ago | (#17913414)

It's still a very flawed test in so many ways.

You just can't measure things like that using a DMM (much less a non-true RMS one) - they're just not meant for measuring short peaks accurately.

Efficiency at peak usage is just that. It means absolutely NOTHING of how efficient it is at low power levels (if your PC is on 24/7, likely ~16h out of that it's idle). I'd rather buy one that's optimized for high efficiency @ low power (idle), which offers significant savings most of the time. Peak efficiency would be more of a concern for those who use their PC intensively for a few hours (like gaming) and then shut it off.

Your test fails to account for power factors and such (non 100% resistive load - switching PSUs are inductive). So your numbers are likely quite a bit off (not that your peak usage measurement was even accurate in the first place) They make specialized power supply testers for a reason you know...

Most good PSUs have a ~80% efficiency. For decent reviews and meaningful numbers, try http://ww.jonnyguru.com/reviews.php [jonnyguru.com]

Re:I just did some research on this actually (0)

Anonymous Coward | more than 7 years ago | (#17907552)

What's the noise level like on the Antec (80mm fan) compared to the SeaSonic (120mm fan)?

Re:I just did some research on this actually (1, Informative)

Anonymous Coward | more than 7 years ago | (#17907812)

If you want to calculate how much you're paying you should look at watts measured by your power meter and not amps. The simple watts=amps*voltage doesn't work for inductive systems like computers or most any other appliance. That said, these are interesting numbers, thanks for sharing.

Re:I just did some research on this actually (1)

Spoke (6112) | more than 7 years ago | (#17908230)

Nice data. I'm surprised to see the Antec PSU more efficient than the Seasonic as Seasonic is generally regarded as one of if not the most efficient PSU manufactures. In fact, Seasonic manufactures many PSUs for other companies, Antec included so I would not be surprised if the Antec PSU you tested was in fact a rebadged Seasonic unit.

There is a full review of the PSU [silentpcreview.com] at SilentPCReview [silentpcreview.com] who has many full reviews of PSUs including efficiency tests.

In general, any PSU with active PFC will generally pretty efficient, especially compared to any PSU without active PFC. An easy way to tell if a PSU has active PFC is to see if it can take 100-240v AC without a switch.

Re:I just did some research on this actually (3, Interesting)

Spoke (6112) | more than 7 years ago | (#17908736)

One other guideline when purchasing a PSU:

Buy the smallest PSU possible!

Many people out there have "SUV syndrome" when buying a PSU and incorrectly assume that they need that huge 500w (or bigger) PSU for their PC. Unless you really do have a high-end gaming PC with a high-end graphics card and multiple hard drives, your computer will almost certainly normally use less than 200w peak, and more typically 75-150w.

What does happen with an oversized PSU is in order to build a PSU to handle high current, it's efficiency at low current drops significantly. Typically the efficiency of a PSU starts dropping pretty quickly below 50% capacity and even faster below 25% capacity.

Finally, you can also look for PSUs which are 80 PLUS [80plus.org] certified. These PSUs have been independently tested to be at least 80% efficient at 20%, 50% and 100% loads with a power factor rating of at least 0.9 at those load points.

The Antec EA430 is part of Antec's EarthWatts series of PSUs which are all 80 PLUS certified.

Out of the other PSUs casualsax3 tested, the SilverStone SST-ST50EF is also 80 PLUS certified. I could not verify if the Seasonic S12-380 is 80 PLUS certified, but it does not appear to be so even though it is more efficient than the Silverstone in casualsax3's test. If the S12-380 is of the "S12 Energy Plus" series then it should also be 80% efficient. I wonder if Seasonic quietly started shipping Energy Plus S12s instead of the old ones...

Earth to Newegg: get with 80 PLUS (1)

schwaang (667808) | more than 7 years ago | (#17911106)

80 PLUS rocks. Why can't I do a search for 80 PLUS-certified power supplies on newegg.com? And for some PSUs they sell that are 80 PLUS, Newegg's product page doesn't even tell you.

For example, this Antec EA380 Newegg product page [newegg.com] doesn't even mention 80 PLUS, but clicking through to the manufacturer's product page [antec.com] clearly shows the 80 PLUS logo.

C'mon newegg! Get with it!

Re:Earth to Newegg: get with 80 PLUS (1)

Spoke (6112) | more than 7 years ago | (#17911638)

Good question. I'm going to send NewEgg an email about it hopefully it's something they can add to their site.

OT: $0.19/kwh ??? (1)

Luyseyal (3154) | more than 7 years ago | (#17908980)

You pay $0.19/kwh? Holy crap that's a lot. My peak rate last summer $0.074/kwh here in Austin, TX (on Austin Energy).
-l

Re:OT: $0.19/kwh ??? (1)

WillAffleckUW (858324) | more than 7 years ago | (#17911468)

You pay $0.19/kwh? Holy crap that's a lot. My peak rate last summer $0.074/kwh here in Austin, TX (on Austin Energy).

Here in Seattle we pay about 4 cents a KWH. I pay about twice that, because I have both the Green Power (wind-only, pays for construction of wind turbines) and Green Up (pays for schools and bus stops to have solar power) programs.

But it's still cheaper than yours. Even if we suffer under the clouds (except my Seattle neighborhood of Fremont, which frequently has rips in the clouds due to the nearby hills of Queen Anne and Phinney Ridge).

The major power usage of the average old PC is the CRT monitor. Using an old monitor can drastically increase power consumption, but moving to a flat screen LCD energy star monitor will cut power consumption about 40 percent from the old PC plus CRT.

I run a laptop, which uses batteries (man the fan kicks out heat! wish I had a lap board since I use it on the couch), but I connect to AC from a converter off of my UPS. My son uses a Mac Mini (fall 2006 Intel version) which gens close to zilch heat and has a giant flat LCD monitor (Sony, better power consumption than the Apple one). All my other PCs are not being used. But my office at work gens so much heat from our dual core and quad core CPUs - even though most have flat panel LCDs - that we have to open the window once spring rolls around or we bake. Since we run Linux, and sometimes have genetics statistical correlations running under Perl for weeks at a time (not a joke), we can't power them down, even if we can turn off the monitors.

Re:OT: $0.19/kwh ??? (1)

Luyseyal (3154) | more than 7 years ago | (#17914164)

Here in Seattle we pay about 4 cents a KWH. I pay about twice that, because I have both the Green Power (wind-only, pays for construction of wind turbines) and Green Up (pays for schools and bus stops to have solar power) programs.

That's pretty sweet. I should point out that that's my peak rate. My lowest this year was $0.068 in November. I have Green Choice wind power which, at the time we signed up, was higher for the fuel surcharge. It is now about the same or lower than the normal fuel surcharge. My fuel charge is fixed for 10 years (and saved us quite a bit after Katrina). The reason my per kwh varies is because Austin has a tiered rate and they increase the rates slightly during the summer months. The lowest tier/rate is up to 500 kwh. During the summer I exceed that.

The last time I checked, Austin Energy had about the third cheapest power in the state. It was only surpassed by a couple power co-ops. The whole deregulation thing has been a complete scam. Commercial power has been way more expensive and the major provider, TXU, now wants to build 11 old-style coal plants within 5 years. Their utter disregard for the public, in pricing as well as health and environment is absolutely appalling.

I switched to LCD awhile back. We have 1 CRT which I plan on replacing when it is feasible. We keep our computers off except when we need them. I used to host a website, mail, etc. at home, but I figured out the electricity cost and it worked out slightly in favor of web hosting, so I switched to that. My workstation is a dual Opteron so it eats power. I don't have money to throw around, so I'll be using it for some time. The wife and the kid both have various Athlon parts.

The other thing that's nice is we switch off all the TV, stereo, satellite receiver, etc. electronics during the day. I haven't bought a Kill-a-watt yet, but I expect that that eliminates heat and some power and has very little downside (except the satellite needing to resync after reboot).

My power could be cheaper if the City didn't use the utility as part of general revenue... but then they'd offset it with higher sales or property taxes so it'd probably all work out the same. (Offtopic: interesting article on how Americans already pay a flat tax of about 40% per household [assetbuilder.com]).

What kind of Perl statistics do you do? (I write Perl and I'm interested in going into observational astronomy with a dash of computational thrown in for good measure.)

Cheers,
-l

Re:OT: $0.19/kwh ??? (1)

WillAffleckUW (858324) | more than 7 years ago | (#17914320)

What kind of Perl statistics do you do? (I write Perl and I'm interested in going into observational astronomy with a dash of computational thrown in for good measure.)

Most medical genetics linkage associations for genetic family trees of people with apoB data for Alzheimer's and Parkinson's and lipid studies.

One thing about our statistics is you can't use normal curves, as disease distribution has most people either very healthy or very sick, with few in between, so not ever a bell curve.

Re:I just did some research on this actually (1)

Q7U (1005025) | more than 7 years ago | (#17909616)

I am considering the 17" iMac, which is basically a laptop on a stand. SilentPCReview did a review of the older Core Duo (not Core 2 Duo) model and the most wattage they could get it to pull was 73 watts, and that was running Windows with 2 x CPUBurn and a ATI Tool stressing the video card.

http://www.silentpcreview.com/article594-page1.htm l [silentpcreview.com]

Another option might be to build a PC with AMD's new 65nm Athlon X2 CPUs, a motherboard with a built in GPU, a 7200rpm laptop drive, and LCD screen. You'd think that would be fairly miserly.

Re:I just did some research on this actually (1)

dr2chase (653338) | more than 7 years ago | (#17910444)

And don't forget, that extra power has to go somewhere; it gets turned into heat, inside your PC case.

Only going to get worse... (1, Flamebait)

Absolut187 (816431) | more than 7 years ago | (#17906692)

Now that Vista requires a mid-range video card to run, the power consumption of run-of-the-mill business PCs is going to take off.

Despite its flaws, Vista will probably muscle its way into the market eventually.
They'll probably stop supporting XP any day now.
Dell will stop selling XP soon I'm sure.

Then businesses will be forced to use Vista on new machines, and eventually they'll say "screw it" and get rid of XP altogether, and then we'll have blackouts.

Re:Only going to get worse... (1)

Joe The Dragon (967727) | more than 7 years ago | (#17906818)

some businesses have just moved to xp not that long ago and they will likely wait for sp1 before even thinking about vista. Also businesses run a lot older software that may not work in vista.

Re:Only going to get worse... (1)

nxtw (866177) | more than 7 years ago | (#17907062)

It doesn't require a mid-range video card.
Only if you want to run the Aero Glass interface do you need something decent; and even then, any recent ATI or nVidia card will suffice or even the Intel GMA 950 integrated chip that has been shipping on even laptops for over a year.

Why is parent modded troll? It's true (1)

ArghBlarg (79067) | more than 7 years ago | (#17907670)

.. if MS doesn't push out XP itself via arbitrary dead-ending of critical packages (already done for what, DirectX or Media Player right? I admit I'm too lazy to check), boneheaded hardware vendors will write drivers that only work for Vista, for no reason other than to reduce their own support issues. It's already happening to Win2k.

Re:Why is parent modded troll? It's true (1)

Absolut187 (816431) | more than 7 years ago | (#17907770)

Apparently someone with mod points thinks that you can have a new dominant operating system require substantially more hardware power without any corresponding increase in electricity usage..

Maybe they know something we don't know.
Like maybe Vista ships with a tiny cold-fusion reactor that powers your PC.
I dunno, I haven't seen Vista... THAT would make me say "wow".

Re:Why is parent modded troll? It's true (1)

yagisencho (930201) | more than 7 years ago | (#17908356)

Vista also provides improved power management services. And that 'mid-range' video card isn't exactly working up a sweat while driving the Aero interface. I expect that you were troll-slapped because of your hyperbolic use of FUD.

Re:Why is parent modded troll? It's true (1)

Absolut187 (816431) | more than 7 years ago | (#17908728)

FUD? There is no uncertainty here.

Vista requires more graphics power.
More graphics power requires more electricity.
Therefore, Vista requires more electricity.

What exactly are these "improved power-management services?"
Sounds like reverse-FUD to me.
You can't do much better than turn a computer off at night.
If it requires more electricity when its running, it will use more electricity.
I don't care how many sleep modes it has.
Anyone who cares about their electricity budget is going to shut them off altogether at the end of the day.

Windows is the dominant operating system, and Vista will eventually be adopted.
All those new faster-CPU, better GPU computers that businesses will buy to adopt Vista are going to draw more power.

Re:Why is parent modded troll? It's true: MS mods (0)

Anonymous Coward | more than 7 years ago | (#17911782)

Moderation courtesy of MS. Don't say anything negative about vista, it's unpatriotic!!!

Just swallow. And shut the fuck up. That's right.

It's A Troll Because... (1)

Petersko (564140) | more than 7 years ago | (#17912626)

"They'll probably stop supporting XP any day now."

It's a troll because Microsoft actually has a GOOD record of supporting their OS's for a few years after they've moved to new versions.

When did they end Windows 98 support? Wasn't it in 2006?

According to links from this page [microsoft.com] mainstream support for XP will end in 2009, and extended support in 2014.

So the original post made a completely false claim that cast FUD. Isn't that a troll by definition?

Mid Range card? Please (1)

D3m0n0fTh3Fall (1022795) | more than 7 years ago | (#17909246)

Anything above the absolute bottom of the barrel will run Aero just fine. Something like an integrated Gefore 6100, 6150, 6200, 7300TC or a radeon X200M or X1150. Those cards will barely sip a couple of watts more than the lowest end intel integrated graphics chip. Troll. This article wasnt about Vista

Re:Mid Range card? Please (1)

Absolut187 (816431) | more than 7 years ago | (#17910640)

Well if you're right about this, then I'm wrong about Vista's power consumption.

But wrong != troll..

Re:Only going to get worse... (not so fast) (0)

Anonymous Coward | more than 7 years ago | (#17909512)

> Despite its flaws, Vista will probably muscle its way into the market eventually.

Perhaps.

> They'll probably stop supporting XP any day now.

Nope. They are supporting it until 2014.

> Dell will stop selling XP soon I'm sure.

True. Everything's Vista now on their ready-to-build machines. You can still find XP on their Outlet PCs and laptops.

So just don't turn on the heater... (2, Interesting)

SeaSolder (979866) | more than 7 years ago | (#17906960)

I used to live in a crappy little studio apartment. It was about 550 square feet. The winter I was there, our balmy Seattle weather dropped into the teens for a lengthy period of time, yet I never turned on my heater. The heat being put out by my refrigerator and my Pentium 4 was enough to keep me nice and toasty warm. If you're really concerned about power consumption because of money saving reasons, you could always move somewhere that electricity is cheap. Here I pay about 4 cents per KwH. Nice, huh?

Re:So just don't turn on the heater... (3, Funny)

spun (1352) | more than 7 years ago | (#17907698)

Here I pay about 4 cents per KwH. Nice, huh?

So that means you can afford huge sun lamps to combat the pervasive Seasonal Affective Disorder brought on by the constant clouds, eh? I keed, I keed. Seattle's a beautiful place with a lovely climate. All that stuff about rain is just a rumor spread to keep the Californians out. Really, it's sunny all summer long in Seattle. And if summer happens to fall on a weekend, everyone goes on a picnic!

Re:So just don't turn on the heater... (1)

textstring (924171) | more than 7 years ago | (#17908468)

I like to think of winter as mother natures gift of a few hundred extra MHz.

Re:So just don't turn on the heater... (1)

mrchaotica (681592) | more than 7 years ago | (#17908876)

The winter I was there, our balmy Seattle weather dropped into the teens for a lengthy period of time, yet I never turned on my heater. The heat being put out by my refrigerator and my Pentium 4 was enough to keep me nice and toasty warm.

Although it's great if the heat produced by appliances as a byproduct is enough to heat the space, it's important to note that you shouldn't run them for the purpose of producing heat, because a heat pump can accomplish the task using less energy. See Wikipedia:

When used for heating a building on a mild day, a typical heat pump has a COP [Coefficient Of Performance] of three to four, whereas a typical electric resistance heater has a COP of 1.0. That is, one joule of electrical energy will cause a resistance heater to produce one joule of useful heat, while under ideal conditions, one joule of electrical energy can cause a heat pump to move much more than one joule of heat from a cooler place to a warmer place. Sometimes this is inappropriately expressed as an efficiency value greater than 100%, as in the statement, "XYZ brand heat pumps operate at up to 400% efficiency!" This is inaccurate, since the work does not make heat, but instead moves existing heat "upstream"; otherwise, this would be a perpetual-motion machine.

Games (-1, Troll)

Anonymous Coward | more than 7 years ago | (#17907334)

The nascar, pro wrestling and SUVs of nerdom. double plus good, it forces folks to "stay stuck on microsoft". You read that a lot here. There's also the obesity epidemic in the western world, that starts with the kids unfortunately, who have gone from tens of thousands of years being physically active to in one generation being pretty passive, at the same time that getting food became relatively easy and inexpensive.

But of course, there's no correlation at all....just a coincidence. Same with violence and brain pattern conditioning, and forced drugging because of ADHD, just a coincidence.

Actually, My iMac is pretty good... (2, Informative)

frostilicus2 (889524) | more than 7 years ago | (#17907668)

My 1.83Ghz Core Duo iMac has a very low power consumption. See here [anandtech.com]. 64w under heavy load and 48w idle. If I put it in sleep I'd expect [dssw.co.uk] that it uses of the order of 5w. Which is impressive given that this is almost half of the power consumption of the most efficient system on test here.

The road to hell (1)

Fry-kun (619632) | more than 7 years ago | (#17907880)

I'd love to save some power!
I have a dual opteron 246 workstation. These CPUs don't support any kind of low power mode. The room gets a few degrees warmer when I run this computer. *Now* tell me how I can save some power while being able to use the workstation. Sleep mode, my ass..
And no, I'm not planning to shell out some $$ to swap the CPUs any time soon :P

Re:The road to hell (0)

Anonymous Coward | more than 7 years ago | (#17909430)

Most, if not all AMD processors support either the older AMD PowerNow system, or the renamed Cool n' Quiet. These will underclock and undervolt your processors when they aren't being used. The heat output by my older Athlon 64 drops by about 20C when this is enabled.

you thought it's surprise? (1)

mentem (1060302) | more than 7 years ago | (#17912828)

Nothing new for me. In future power usage will increase. Mobile system's power usage will increase too, but new batteries will be invented

Power consumption is falling... (1)

evilviper (135110) | more than 7 years ago | (#17913214)

"The power consumption of modern PCs has skyrocketed the past few years.
No, it WAS rising, several years ago.

When AMD switched Opteron/Athlon64/Sempron64, power consumption fell, and continues falling.

When Intel got off the P4 chip, power consumption fell.

When 80%+ efficient (consumer) PSUs came out, power consumption fell.

etc.

Power consumption is significantly falling. Unfortunately, many companies are sticking to the slightly cheaper, but vastly more power hungry components, like P4-based Celerons, cheap "500W" 50% efficient Asian PSUs, etc.

You can put together desktop systems today, which are more energy efficient than some of the larger notebooks computers. Unfortunately, only the premium (read: expensive) computer manufacturers are doing that.

Re:Power consumption is falling... (1)

toddestan (632714) | more than 7 years ago | (#17915986)

Actually, it's going back up again. It used to be that single core CPUs were using 75-100W or more at peak back in the Socket A and P4 days. They've gotten that back down to 30W or so per core - the only problem is that they are putting 2 and 4 cores on a chip now! We're essentially working our way back to where we started, though hopefully the budget (single core) chips will end up being very low power chips.

Re:Power consumption is falling... (1)

evilviper (135110) | more than 7 years ago | (#17917302)

We're essentially working our way back to where we started,
Not quite.

First, back in the good old days, CPUs didn't idle very well... With CnQ and SpeedStep on desktop CPUs now, they're using less power when idle, than ever before. Even if the peak is higher (which it usually isn't) your CPU is still largely idle 90% of the time, so there's tons of room for power saving with CnQ.

Second, that only applies to the CPU, while the rest of the system is falling... Specifically, with AMD installing the memory controller on the CPU, they've practically eliminated the northbridge, which was drawing up to 50% as much power as the CPU itself, as the bus got faster. That was specifically the situation on the Via KM600 chipset.

Turning off a PC isn't pracical (1)

GWBasic (900357) | more than 7 years ago | (#17913334)

Completly turning off a PC isn't practical. I recieve incoming calls via Skype, which requires that I leave my computer on.

Granted, with a few relays, we could make monitors & speakers more efficient. I wired up a 10-amp AC relay to my reciever, which I use to completly turn off my subs and TV when my home theater is off. It saves me about $5 / month! Those of us who are real power misers could just plug our PC monitors into a switched outlet for real savings.

Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...