Power Consumption of 24/7 PCs

A forum just for SPCR's folding team... by request.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Power Consumption of 24/7 PCs

Post by MikeC » Mon Apr 07, 2003 5:23 pm

This occurred to me as I was writing up the Verax PSU review, which features Power Factor measurements (1st time in my PSU reviews): All of these PCs going 24/7 for the folding effort is not that cool from a power conservation / enviromental pollution POV.

The second page of that article has a brief worst / best case comparison about PSU real power consumption. If you have never considered it, it's mind boggling.

So here's a question: how many of you are running Active PFC PSUs on your PCs? Do you know the AC power draw of your systems?

Mine are both running modified Seasonic APFC PSUs right now -- Kill-A-Watt measures PF on each to be 0.99, which means the AC power consumption I measure going into each system IS in fact the real AC power the electric company is using. I just measured the AC wattage on both: 122W and 132W.

According to BC Hyro's Appliance Calculator, this works out to 2220 kwh and will cost CA$130 per year. We have low electricity rates in BC, I understand. I wonder what the same would cost in California? Also calculated what real AC consumption would be if I had a typical 0.65 PF PSU: 188W and 203W, which work out to 3420 kwh & CA$200 a year.

Food for thought...

dgouldin
Posts: 108
Joined: Fri Mar 14, 2003 1:31 pm
Location: Denton, TX
Contact:

Post by dgouldin » Mon Apr 07, 2003 6:10 pm

Yeah, but if you're going to have your computer on all the time anyway, how much is the cost really increased? Not $200 I'd bet.

... just think of it as a non-tax-deductable cancer research donation on the installment plan. :lol:

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Mon Apr 07, 2003 6:24 pm

It's not so much the cost that I am considering but the resources it takes to create the electricity required. Generally speaking, industrial pollutions (all kinds) are a big source of increased cancer incidence...

powergyoza
Patron of SPCR
Posts: 543
Joined: Mon Oct 21, 2002 1:01 am
Location: Vancouver, BC, Canada
Contact:

Post by powergyoza » Mon Apr 07, 2003 7:50 pm

MikeC wrote:It's not so much the cost that I am considering but the resources it takes to create the electricity required. Generally speaking, industrial pollutions (all kinds) are a big source of increased cancer incidence...
Hmmm...that makes it more complicated.... it's kinda counterproductive then to consume nrg which, in its production, can lead to higher incidence of cancer or other environmental maladies. But's it's also a waste of all the energy put into manufacturing all this cpu power (think aggregate of ALL cpu power on earth) and not take advantage of it all. Imagine the fewer number of # crunching servers that would need to be produced if the existing silicon was exploited for parallel-computing-solvable calculations. Ie. Fewer computers have to be built b/c we're taking advantage of our extising computational power -- in theory it makes a difference doesn't it?

Okay, that was way too much for a forum posting -- should be left for a theoretical paper or something.... I will definitely consider getting an ActivePFC PSU though. And turning down the heat. And turning off the lights... etc... etc. etc.

jinu117
Posts: 241
Joined: Sun Jan 26, 2003 1:46 am
Contact:

Post by jinu117 » Tue Apr 08, 2003 2:53 am

Now, I feel extra good about the 3.42 Ghz machine I am running (well, at least I am saving some with my LCD I guess...)

jamoore9
Friend of SPCR
Posts: 152
Joined: Thu Jan 30, 2003 5:58 am
Location: Fairfax, Virginia, USA

Post by jamoore9 » Tue Apr 08, 2003 7:47 am

Generally speaking, modern power plants aren't really culprits in industrial pollution. The only exceptions are those of us in the east and midwest who still get power from fossil fuels (coal and oil -- might be more wide-spread, this is just my impression). Those are obviously producing pollutants due to combustion (just like your car). But your computer's share (at CAD200 or USD135) is atronomically minimal.

One thing you might look into is whether you have the option of choosing a power generation company to be delivered through your power provider (ie, the means of production to be delivered through the electric company's grid to yuor home). In Pennsylvania we have the right and ability to choose a generator. One of the companies we can choose generates ALL of their power through windmill farms (that's who I went with). Their rates are about 43% higher, but it's a personal decision if its worth the money to you. It is to me.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Tue Apr 08, 2003 7:59 am

jamoore9, straight-out pollution is not the only downside of electricity generation... Since we're veering wildly from our usual obsessions, here's a worthwhile link:
http://www.microconsultants.com/tips/pwrfact/salmon.txt

riffst3r
Posts: 221
Joined: Tue Apr 08, 2003 8:43 am
Contact:

a scary thought

Post by riffst3r » Tue Apr 08, 2003 9:10 am

i'm a student here, in the caribbean. i've never actually thought how electricity was produced here in st kitts until MikeC brought it up.

so, i did a search and this is what i from the cia's website (about this country, st kitts):

Code: Select all

Electricity - production: 95 million kWh (2000) 

Electricity - production by source: fossil fuel: 100% hydro: 0% other: 0% (2000) nuclear: 0% 

Electricity - consumption: 88.35 million kWh (2000) 

Electricity - exports:	0 kWh (2000) 

Electricity - imports:	0 kWh (2000) 
http://www.cia.gov/cia/publications/fac ... os/sc.html

this scares me. i never figured we'd be still using fossil fuels in this day and age... makes me think twice about leaving the hall light on at night.

:(

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Tue Apr 08, 2003 9:29 am

riff brings up a good point. jamoore9 and I are both in the northwest (BC is more or less just a satellite of the US, no? :wink: ) where most of the power is generated by hydro, which is somewhat more benign than hyrocarbon / nuclear generators -- but obviously this does not hold true everywhere in the world. I am sure someone will correct me if I am wrong, but I am of the impression that more electricity in the world is created via fossil fuels than via hydro.
Last edited by MikeC on Tue Apr 08, 2003 12:42 pm, edited 1 time in total.

riffst3r
Posts: 221
Joined: Tue Apr 08, 2003 8:43 am
Contact:

Post by riffst3r » Tue Apr 08, 2003 12:30 pm

following MikeC's link, this is what caught my eye:
Colin Mattoon wrote:I believe that the salmon are also an important part of the economy, and for moral/ethical reasons I do not believe that humans should cause the extinction of species unless there is no alternative.
i don't think we have any right to cause the extinction of a species... not unless if we're ready to face the same fate. it may be just fishes now... but can you tell what could be next?

i'm sorry if that sounds a bit too harsh, but that is what the world we live in has evolved to.

but i digress. Active PFC PSU's in computers increases power usage efficiency. and just like this distributed computing project we're a part of, every little bit helps.

if you don’t have one, get one. you really should.

[pleading mode] do it for the fishes [/pleading mode]

:|



ps. is it just me, or is everyone else getting a gazillion invalid_session's ?

miker
Patron of SPCR
Posts: 798
Joined: Sun Aug 11, 2002 3:26 pm
Location: Akron, OH (The Rubber Capital)
Contact:

Post by miker » Tue Apr 08, 2003 12:32 pm

Pretty sure all my power comes from Perry nuclear power plant here in Ohio. Our fish glow. I need this cancer research for my kids 8)

SometimesWarrior
Patron of SPCR
Posts: 700
Joined: Thu Mar 13, 2003 2:38 pm
Location: California, US
Contact:

Post by SometimesWarrior » Tue Apr 08, 2003 5:35 pm

My geology professor today, conveniently enough, showed a table of where the U.S. power came from for 2002.

40% from oil,
25% from natural gas,
22% from coal,
8% from nuclear power,
5% renewable (I assume this includes hydro, wind, and solar).

I thought hydro power was a much larger contributor to the power grid. Perhaps this table was instead a "U.S. energy" table, including transportation and things other than just the power grid. I'll try to clear this up with him.

SometimesWarrior
Patron of SPCR
Posts: 700
Joined: Thu Mar 13, 2003 2:38 pm
Location: California, US
Contact:

Post by SometimesWarrior » Tue Apr 08, 2003 6:02 pm

My latest energy bill pegged me at US$0.13 per kWh, baseline. I'm running a dual-processor Athlon machine, probably sucking up 200W DC from the power supply, an Antec TruePower 400W.

I don't know how much AC power I'm drawing. Let's say 300W. Running my computer 24/7 would mean I gulped down 50kWh a week, or 2600kWh a year. That's US$340 a year! If I upgraded from Palomino XP's to T-breds, I could recover the cost of the new processors in a couple of years just from the power savings. If my TruePower powersupply isn't active PFC (or isn't efficient), an upgrade to a Seasonic could pay for itself in a matter of months.

This discussion is reason enough for me not to buy another SMP machine. And it's one more reason to fill my refridgerator with water bottles, and replace my incandescent light bulbs, and sell my car. :) My highschool physics professor used these tricks (he carpooled), and he was so energy-efficient, the power company had to put him on a flat-rate plan because he used practically no power.

MikeC, I calculated your 132W computer to cost US$150 if left on for a year in southern California. That's a big savings over my machine.

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Tue Apr 08, 2003 6:03 pm

Perhaps this table was instead a "U.S. energy" table, including transportation and things other than just the power grid
nah, if it was there is no way coal or nukes would be on the list. UNless GM or Honda have released some new model I know nothing about. :lol:
Our fish glow. I need this cancer research for my kids
:lol: :lol:

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Tue Apr 08, 2003 6:32 pm

...an Antec TruePower 400W
So happens I just measured the PF of a TP380: 0.69 -- or you could say 69% AC power efficiency. This is at around 250W AC power. Looking back at my Sonata 380S PSU review, I see that ~150W, its (AC/DC) efficiency is around 68%. So what this probably means:

It takes 220W AC for your PSU to deliver 150W DC, 70W lost as heat
For 220W AC to get to the PSU, 319W AC total is used, 99W lost to harmonics.

You're drawing ~320 total AC (assuming your machine is not too far off from Powergyoza's dualie).

Guest

Post by Guest » Tue Apr 08, 2003 8:42 pm

MikeC wrote:
...an Antec TruePower 400W
So happens I just measured the PF of a TP380: 0.69 -- or you could say 69% AC power efficiency. This is at around 250W AC power. Looking back at my Sonata 380S PSU review, I see that ~150W, its (AC/DC) efficiency is around 68%.
Thanks for the statistic. Powergyoza's machine was drawing 200W DC when running at full tilt. Assuming 70% efficiency and 69% PF, that means the power supply is delivering 285W AC, and the total AC power used is 411W. That means a year of on-time will consume 3.6MWh and cost me just under $470. Switching to a Seasonic (0.99 PF) would save 1.1MWh and $140 a year, or save me even more money if I go over my baseline charges.

(I'm going off-topic here, sorry.)

Now I'm in a bind: I'd like to upgrade to an active PFC power supply, but all the Seasonic models top out at 30A on the 5V rail, and judging from Powergyoza's measurements, I'll likely need more than that. The only other PSU on the recommended list with active PFC appears to be the Zalman one: am I mistaken? The Zalman 400W model delivers 40A on the 5V rail, but I con't see it for sale anywhere.

I'm tempted to just rip the second processor off the motherboard, since that would solve many problems all at once: power, heat, noise, and environmental impact. :grumble:

powergyoza
Patron of SPCR
Posts: 543
Joined: Mon Oct 21, 2002 1:01 am
Location: Vancouver, BC, Canada
Contact:

Post by powergyoza » Wed Apr 09, 2003 11:58 am

Anonymous wrote:Now I'm in a bind: I'd like to upgrade to an active PFC power supply, but all the Seasonic models top out at 30A on the 5V rail, and judging from Powergyoza's measurements, I'll likely need more than that. The only other PSU on the recommended list with active PFC appears to be the Zalman one: am I mistaken? The Zalman 400W model delivers 40A on the 5V rail, but I con't see it for sale anywhere.
40A on the 5V rail should only be necessary if you have the old-school Tyan Tiger MP 2460. If you have any other mobo, you should be fine with the Seasonic. I'm not so fortunate of course.... so your question is still appropriate. Any ActivePFC PSU's that can deliver 40A on the 5V rail?

EDIT: I just answered my own question. svc.com has the Zalman 400 watt A-PFC for $99 USD.

powergyoza
Patron of SPCR
Posts: 543
Joined: Mon Oct 21, 2002 1:01 am
Location: Vancouver, BC, Canada
Contact:

Post by powergyoza » Wed Apr 09, 2003 12:10 pm

Oops, and the silenx.com website is up and taking orders for their 14db PSU's too (A-PFC). Fortron-based like the Zalman. In another thread it was mentioned that new stock will be coming in a few weeks.

aphonos
*Lifetime Patron*
Posts: 954
Joined: Thu Jan 30, 2003 1:28 pm
Location: Tennessee, USA

Post by aphonos » Wed Apr 09, 2003 12:49 pm

FYI, here is the power consumption statement from the FAQ at the F @ H site:
They wrote:How much power/money is used by keeping a F@H running 24/7 on a computer?

Roughly, a CPU uses about as much power as a 60 watt lightbulb. Here's a report on computer power management from Lawrence Berkeley goverment labs, and there are other referencs on the web you can find. Although power supplies on most computers are rated at 250 watts, average usage is much lower. On average, a Pentium-type computer uses between 45-70 watts (I've read various different sources on this) while it is on. If the computer has no idle mode, it will use the same amount of energy whether it is running a program or not. If it is on idle, it will comsume around 25 watts. So, the daily difference between off and running F@H is about 24x(45 to 70) = 1.1 to 1.7 kWh. At $0.14 per kWh ( from PG&E here in California), this works out to about $0.15 to $0.24 per day, or perhaps $6 a month. The difference between an idled computer and one running F@H would be closer to $4 a month - and if the computer was already being used 8 hours a day, it would be closer to $3 a month.

In general, lighting and climate control use a much larger share of household power than computers do. So the best bet for cutting costs and conserving energy would be to turn off lights, turn off your computer monitors (which use more power than a CPU), and turn down the heat. And keep folding :)
Is the difference between their numbers and the numbers discussed here due to the fact that they are only counting the power output/wattage for the CPU?

DaShiv
Posts: 76
Joined: Tue Jan 14, 2003 12:32 pm
Location: Berkeley, CA, USA

Post by DaShiv » Wed Apr 09, 2003 1:16 pm

Yes--they're counting only the increased power consumption for a CPU under load vs. one at idle, rather than the power consumption from leaving the entire system running.

So it's pretty negligible to fold if you were going to leave your system on for other reasons anyways. But if you're leaving it on just to run F@H, then there are some power consumption issues to think about.

Me, I leave it running 24/7 anyways. :)

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Wed Apr 09, 2003 1:23 pm

It's dificult to say what they base their numbers on, but I dug up the original report, and it seems dated. The reference to Pentium CPUs is an obvious hint. Tests on the electrical consumption of P3, Athlon, XP, P4 PCs I've done over the past 12 mos. showed:

-- In idle, most single-CPU PCs draw less than 100W AC -- ie, the DC power is under 70W
-- when working hard, these PCs hit ~150W max, but usually not long term. The 132W I cited in an earlier post is more typical. That's a system with 1G RAM & dual HDDs.
-- The lowest power consumption was seen with VIA mini-ITX EPIA-bsed systems. W/1 HDD, 512M RAM, 200W PSU, the typical AC power draw was under 40W (I seem to recall) and did not exceed 50W.

ez2remember
Friend of SPCR
Posts: 809
Joined: Thu Nov 28, 2002 5:07 pm
Location: London, UK

Post by ez2remember » Wed Apr 09, 2003 1:39 pm

Maybe this is where laptops excel, they don't draw that much power. They are becoming increasingly fast, and almost comparable to desktops soon.

Intel Centrino is out! It's a new mobile P4 cpu with double L1 and L2 cache, increased to 16k and 1megabyte respectively. A 1.6Ghz is said to be faster than 2.4Ghz. This is the official word from Intel.

http://www.intel.com/home/index.htm?iid ... thin_Home& Scroll down and look for tools on the right hand side, click on 'compare performance' then a screen will load, click on 'notebook' tab and see it for yourself.

I have read reviews of the new Centrino and it does perform to it's claims. Best of all it's said to only consume 20watts or so. Battery lasted over 5hrs with light use on a Toshiba Tecra.

I can imagine how much a notebook would save on power comsumption and hence save you $$$ every year on your electricity bill. :D

jamoore9
Friend of SPCR
Posts: 152
Joined: Thu Jan 30, 2003 5:58 am
Location: Fairfax, Virginia, USA

Post by jamoore9 » Tue Apr 15, 2003 12:01 pm

ez2remember wrote:I can imagine how much a notebook would save on power comsumption and hence save you $$$ every year on your electricity bill. :D
Plus you can stick it in a docking station, drill a whole in the back of the drawer in your desk for cables, and stick the whole darn thing in the drawer! Now that's Silent Computing. (Me, one of my main machines at work is a PIII 800Mhz Dell laptop. Using it right now. Sits on the desk, whirring away all day long... Driving me insane...)

Zhentar
Patron of SPCR
Posts: 918
Joined: Thu Mar 27, 2003 4:00 pm
Location: Madison, WI, USA

Post by Zhentar » Mon Apr 21, 2003 1:24 pm

I think my notebook is an uber silent pc :) automatic underclocking, 1 fan, silent PSU (except the adapter can't quite keep up with a full load power draw and full brightness LCD). Not only that but I get easy selected optical drive speed, and I can't hear the hard drive except for the occasional click (that the seek noise?), its truly silent while idling.

Now for the whole power thing- I did a research project on nuclear power 2 years ago for my 10th grade english paper. Not to mention last friday we had some navy guy come in and talk to us about nuclear power in physics. World wide, nuclear power provides around 20% of the power, and fossil fuels about 70%. If you inlcuded transportation though, things would weigh even more heavily to fossil fuels. The US has the lower 8% from nuclear power since one or two dozen nuclear plants were built back in like the 70's, but none have been built since, because people think it will make their fish glow and their kids get cancer. The truth is living a few hundred yards downwind of a nuclear plant will expose you to less radiation than 1 x-ray, or the difference of a thousand or so feet of air filtering UV rays. Then of course people are afraid of meltdowns. There have been 2 nuclear accidents, Chernobyl and 3-mile island. Chernobyl had the problem compounded by operator error; 3-mile island was handled correctly and released less radiation than 2 x-rays. All of our reactors are safeguarded against accidents like Chernobyl; the actions that screwed things up there are not physically possible here. And of course there is one last problem- radioactive waste. 2 things to point out here- 1 century ago, nuclear fission wasn't even thought of- if we make as much advancement in the next century, handling the waste may be trivial. And a much bigger point- it is possible to recycle the waste and get 10 times more power from the same amount of uranium- but it produces plutonium, and people automatically think atomic bomb. the truth here is that the plutonium is not high enough grade for nukes and every country capable of refining it is capable of making plutonium themselves.

But anyways, if you really feel bad about your computers power draw go buy a solar panel.

Which reminds me, renewable energy may be environmentally friendly, but its not economically feasible- Powering L.A. would take solar panels covering all of Texas- of course it would be a good use for texas :) but not feasible. There are only a few wind farms across the US and they are more expensive.

So, there's only one option- Raise your kid to be a science genius and make him discover a means of effectively harnessing fusion :shock:

powergyoza
Patron of SPCR
Posts: 543
Joined: Mon Oct 21, 2002 1:01 am
Location: Vancouver, BC, Canada
Contact:

Post by powergyoza » Mon Apr 21, 2003 1:41 pm

Zhentar,

I don't want to get into a debate about nuclear power, but I do know that many scientists and activists - David Suzuki being one - argue that there is not enough resources on this earth to power our future needs. Not enough uranium nor fossil fuels in the ground. On top of that, experiments in Yucca mountain show that nuclear waste can be safely stored, but the models can't account for all the variables in the mountain.

Like you said, solar and wind won't cut it b/c there's not enough useable land to house all the panels & turbines needed.

I am hopeful about this technology that can convert anything into petroleum products. They call it thermal depolymerization: http://discover.com/may_03/gthere.html? ... atoil.html

But then we're still relying on fossil fuels! Or should I say turkey fuel? I like the sound of that better than carbohydrate fuel.

Zhentar
Patron of SPCR
Posts: 918
Joined: Thu Mar 27, 2003 4:00 pm
Location: Madison, WI, USA

Post by Zhentar » Mon Apr 21, 2003 1:48 pm

My personal opinion is that we should use nuclear reactors to hold us off till we can get a real good energy source, either fusion or solar cells, but both have a long ways to go before they can be a feasible large scale energy source. So, like I said, raise your children to be uber geniuses to solve all of the worlds problems. can we agree on that solution? :)

wussboy
Posts: 635
Joined: Wed Oct 23, 2002 12:34 pm
Location: Southampton, UK

Post by wussboy » Mon Apr 21, 2003 2:21 pm

I realize we're spiralling rapidly OT, but I firmly believe that the world is based on economics. They recycle like crazy in Germany, because it's cheaper to recycle than it is to find somewhere to dump that stuff. They don't recycle in central Canada (Saskatchewan) because they have a billion miles to dump that stuff. It's way cheaper just to throw it away.

And my point is thus: At this point no one does solar because it costs more than fossil fuels. At such time as the fossil supply dries up, it will become more and more expensive to put gas in your tank. The very instant this price becomes more than solarpower, people will switch to solar. And yes, at this time, fossil fuel tech is much more refined than solar tech, but that will all instantly change too, the very second it becomes more economical to use solar power.

I'm not worried. Humanity always fluxes and changes. We used coal before we used fossil fuels, and we used trees before that and we used horses before that. The sun is a limitless energy source. We will find a way to use it for cheap.

Alistair
*Lifetime Patron*
Posts: 125
Joined: Sun Jan 05, 2003 12:59 am
Location: Vancouver BC Canada

Post by Alistair » Tue Apr 22, 2003 2:08 am

wussboy wrote:I'm not worried. Humanity always fluxes and changes. We used coal before we used fossil fuels, and we used trees before that and we used horses before that. The sun is a limitless energy source. We will find a way to use it for cheap.
It is our cheap and disposable thinking that has eroded our atmosphere, and its never coming back. Its the attitude of too many that 'everything's going to be alright' that is destroying the planet.

riffst3r
Posts: 221
Joined: Tue Apr 08, 2003 8:43 am
Contact:

Post by riffst3r » Tue Apr 22, 2003 3:59 am

Alistair wrote:It is our cheap and disposable thinking that has eroded our atmosphere, and its never coming back. Its the attitude of too many that 'everything's going to be alright' that is destroying the planet.
i believe wind powered generators are the only source of energy that's clean, renewable, , efficient, and has the least (if not any) impact on the enviroment. this is, of course, side by side with solar panels.

wussboy
Posts: 635
Joined: Wed Oct 23, 2002 12:34 pm
Location: Southampton, UK

Post by wussboy » Tue Apr 22, 2003 8:17 am

Alistair wrote:It is our cheap and disposable thinking that has eroded our atmosphere, and its never coming back. Its the attitude of too many that 'everything's going to be alright' that is destroying the planet.
Oh, I'm not advocating environmental irresponsibility. But I think you and I both know that it's never going to happen on any meaningful scale until the economics are there. I'm not happy about the damage that has already been done either, but my point was mainly that the only reason there aren't any serious alternatives to fossil fuels right now is because it is still cheaper to use fossil fuels.

I'd like to save the earth, but I want a car too.

Post Reply