What is Digg?119 Comments
- macbwizard, on 10/11/2007, -27/+89"never performs any work"
No kidding. Work is force times distance. :-p - unibomber999, on 10/11/2007, -2/+64The remaining 50% is wasted on my co-workers, who use it to browse fark all day.
- gcnaddict, on 10/11/2007, -4/+65Then again, it's not like you're doing any work either.
- afruff23, on 10/11/2007, -5/+64Correct me if I'm wrong, but electricity has a voltage, AKA EMF. EMF=electromotive force. So if the potential changes place, then work is done.
- razrielle, on 10/11/2007, -7/+39"You go to hell, you go to hell and you die"
- inactive, on 10/11/2007, -4/+34"No kidding. Work is force times distance. :-p"
Go Go Gadget 8th Grade Physics! - toppgun, on 10/11/2007, -1/+30with my macbook pro sitting on my lap I would seriously like them to improve on this.
i think I can cancel my winter heating plan... and chances of having kids. - n344853835, on 10/11/2007, -1/+21HUKI365, maybe he meant the bloat Dell installs. Being as it's talking about _Dell_ turning on a feature in their Vista systems. Why _wouldn't_ he be referring to the software that Dell pre-installs? Next time, just think before you turn everything into an OS war.
- gcnaddict, on 10/11/2007, -10/+30In Capitalist America, power corrupts you!
- Roger, on 10/11/2007, -3/+22Simple - don't use you computer for a month and then look at your power bill.
- BigglesPiP, on 10/11/2007, -1/+19This is based on a cheap 50% efficient PSU, My Tagan sits around 75%.
- bIuebonics, on 10/11/2007, -0/+16@HUKI365
i take it you've never owned a dell computer. i'm pretty sure the bloat he's referring to is all that craptastic trojan-in-a-box software dell preloads onto its machines. - nicksauce, on 10/11/2007, -1/+15@dooms13... those are all expressions for power, not work
- Satanael, on 10/11/2007, -1/+13How can I find out my PSU Efficiency?
- nerditup, on 10/11/2007, -1/+13Intel was one of the companies who stated that this is an issue.
- Nocterminus, on 10/11/2007, -0/+11"The problem is cost, said Pat Gelsinger, senior vice president of the Digital Enterprise Group at Intel. Making a PC more power efficient in this manner adds about US$20 to its retail cost, and it adds about US$30 to the cost of a server."
This is really a no-brainer. I would happily pay an extra $20 for a power supply that saved electricity and generated less heat that needed to removed from a room with several boxes running...... and I did ;) - spartacus51, on 10/11/2007, -1/+10MOST will say it right on the side of the PSU. Otherwise, figure out who makes it and look on their website, it should be listed. If it's generic and 2 years old it's probably 63%. For some reason an awful lot were 63% as a standard. Even today's 'high efficiency' is rated 80+, most running ~82%. I've never seen a 90+% (doesn't mean it doesn't exist, it's been a year since I looked at PSUs)
I do like that this has been a 'major' news story. I've seen it now in 4 RSS feeds today. If intel and google are really just figuring this out they should be embarrassed. Like I said before, older power supplies were 63%. Obviously more is lost to heat with at least 4 decent size heatsinks in just about every modern computer. Plus ram spreaders becoming more popular. Any heat is electricity 'lost'. And of course, with such a scientific number like 'half' it's merely an estimate on their part anyway.
If anyone is interested in replacing their power supply might I recommend silentpcreview.com. They have a ton of great information on this topic. - HappyScrappy, on 10/11/2007, -0/+8silentpcreview.com
Basically, unless you bought your PS specifically for its efficiency, it has crappy efficiency. We haven't reached a point yet where high efficiency is "bundled in" to power supplies. It's either listed as a feature that you're likely charged more for, or it isn't there.
spartacus51:
Yes, Google seems to just have figured this out. Finally. RAM spreaders don't do anything for efficiency. How would a piece of metal reduce the actual heat output of a system. It can only help dissipate it. It's like saying if you put a bigger radiator on your car, it'd get better mpg. - bsdfree, on 10/11/2007, -0/+8Look into the AMD K8 Cool-n-Quiet feature - it does exactly what you want, and it came with every socket 939 processor.
- Roger, on 10/11/2007, -1/+8The electrical energy is just the integral of i(t)*v(t) over the time of interest. You want to be thinking in terms of power though.
- HappyScrappy, on 10/11/2007, -2/+9100% of the power turns to heat.
The power taken from the wall has to be converted into another form of energy. Polygons/second and Excel recalcs aren't recognized as forms of energy.
So except for the little bit of power that gets turned into light for the power LED or gets driven out on the ethernet or video interface, ALL the power gets turned straight into heat.
Not just half. - AwRy108, on 10/11/2007, -1/+7...and what percentage of the general population do you think you fall under? Glad to hear your PC is more efficient; but that doesn't mean the rest of the world is in the same category. The sad part is, I look at how many people in the US still drive gas-guzzling SUV's to work and back every day, and I know it'll take nothing short of either a miracle or a national disaster for the States to start being concerned about wasted energy. Meanwhile, places like the UK are making it illegal to produce devices that have a "stand-by" mode, which, surprisingly, has had a effect on decreasing power consumption in their area of the world.
- EntropyMan, on 10/11/2007, -0/+6I'd love to have a more efficient PSU, but I'd really want is a UPS that outputs DC directly to my PC without doing the whole AC/DC thing again in the PSU. Maybe call it a UPSU combo...
Or, in acro-speak: AC->UPS->DC->PSU->HD/CD/DVD/MB/CPU->USB->TLA=OD...
Anyone know of a cheap/reliable one? - Leffe, on 10/11/2007, -0/+6VIA processors, buy them.
- rjc1187, on 10/11/2007, -0/+6Well, my room gets hella hot when my pc is on. maybe this is the reason, besides the fans.
- 13B1303, on 10/11/2007, -0/+5heat> steam> power station> transformer> AC/DC conversion> CPU/wifi chips/video> porn> fap fap fap
That's pretty much it boys and girls, the porn heat engine. - geminitojanus, on 10/11/2007, -3/+8"The hotter a processor gets under normal load, the more inefficient it really is."
Or it's just got reallllly fast flipping transistors. Taking gate leakage aside, the amount of heat a MOSFET creates is directly related to how fast its switching. Having 200 million transistors flipping at 3GHz can do a hell of a lot more computational work than 30 million transistors flipping at 400MHz. Of course, if the company's were really interested, there's no reason we couldn't run at 400-800MHz and press a button and get 3GHz when we need it (or hell, determine the need in software and adjust accordingly), but they'd rather make you pay more for that should-be-standard feature... - Roger, on 10/11/2007, -2/+7@zdiggler (#7157788)
That doesn't tell you anything about the PSU efficiency though. - digtastic, on 10/11/2007, -0/+5I don't care what they say, AC/DC rocks!
- Soapdish, on 10/11/2007, -1/+6From my Thermodynamics text book:
Heat and Work are forms of energy transfer across system boundaries.
Heat is the form of energy transfered between two systems (or a system and its surrondings) by virtue of a temperature difference.
Work is any transfer of energy across a boundary of a closed system that is not heat.
Here's an example of the importance of system boundaries. I have a space heater in my room. If I choose a boundary that is my entire room including the space heater, energy is coming into my room in the form of electrical work. If the system doesn't include a space heater, energy is coming into my room in the form of heat.
When the article is talking about work, it means "useful work", which is a very different thing. If I wanted too heat my room in the winter, then 100% of the work in my computer is useful work. However, if the work I want is only performing operations, etc. then any energy not used for that is not considered useful work.
The efficiency of a system is (useful energy out)/(total energy in). It's kind of difficult to define what useful work is for a computer.
However, you can figure out how much energy is required to perform a task in the ideal case. You can then define the efficiency as (Ideal energy consumption)/(actual energy consumption). - sremick, on 10/11/2007, -1/+5My Seasonic is 88% :D
- HappyScrappy, on 10/11/2007, -0/+4A larger radiator never leads to better mpg. In fact, smaller radiators and raising engine operating temps from 165F to 185F or 195F in the last 20 years has increased efficiency.
And I don't agree a larger radiator allows a smaller pump. You need a certain amount of coolant flow to cool your car. Yeah, if the coolant were a little cooler it would in theory allow you to pump less of it, as it can absorb more heat. But in reality, too cool radiator fluid isn't good because it will give uneven cooling in the cylinder, and lead to parts of the cylinder wall (where the coolant is flowing from) being too cool. - geminitojanus, on 10/11/2007, -1/+5"If intel and google are really just figuring this out they should be embarrassed."
They're not "just figuring it out". They're "just now deciding that it's a necessary measure and thusly standardizing it".
The whole reason it's news is because Intel and Google are trying to get people talking more about efficiency in computers, which is usually a talking point people zoom right past and go straight for the GHz number. Who cares if your server can do 20% better than your competitors if it uses 50% more energy, it's going to cost you more to run than your competitor's machine. Unless your server is a hell of a lot cheaper, your product isn't going to survive the market.
It's really past time we cleaned up our energy act in computers. There's no reason they should belch fire or require eight fans or heat pipes and radiators. I miss the days when you could buy a regular Pentium box without even a fan in the machine and have it never make a sound (except when you drop a floppy in, or dialed up to the internet...), when the machine didn't warm up the room you were in, when you could actually keep a computer in your bedroom and not feel inconvenienced by the noise or the heat (or for that matter, the freaking blinking LEDs they insist on these days). Alas, those were the days... - HappyScrappy, on 10/11/2007, -0/+4My computer absolutely puts out as much heat as an 85W space heater (my computer idles at 85W). An 85W space heater doesn't do much. A 90W light bulb puts out almost as much direct heat as an 85W space heater. Do you notice your room getting a lot hotter when you turn on a 90W light bulb?
Space heaters usually START at 1000W. - supimscott, on 10/11/2007, -0/+4On Digg.com, horrible variations on jokes by unfunny comedians from the 70s post you.
- Kumeelyun, on 10/11/2007, -1/+5Whaddaya mean, never performs any work'? You wouldn't believe the money I save on heating during the winter when I let the computer heat the house!
- cyssero, on 04/18/2009, -1/+4It is a standard feature and they don't get you to pay more, geminitojanus. Intel SpeedStep automatically clocks up and clocks down when required.. maybe not as extreme as 3GHz down to 400MHz, but to a fair extent.
- HsoKinees, on 10/11/2007, -0/+3solution? we NEED energy efficient graphics cards!!
- 13B1303, on 10/11/2007, -0/+3rectify and roll!!
m/ - dougmc, on 10/11/2007, -1/+4`Is there a way to prevent this waste?....'
Well, judging from who wrote the article, the answer they'd probably like you to pick is `buy Intel and use google', though it's debatable how much that would actually help.
More useful would probably be to buy things that use less electricity. Did you know that your microwave oven probably uses more electricity keeping the clock going than it does microwaving your food? (Because the clock is on 24/7, but it's only microwaving your food for perhaps 5 minutes/day.) Wall-wart transformers are known for wasting lots of electricity. People complain about inefficient switching power supplies, but wall warts keep using power even if the thing they're powering is turned off.
Insulate your home better. Adjust your thermostat so the A/C and heater don't have to do as much work. Use thermostat timers to let the house get hotter (in the summer) while you're at work. Turn off all your computers if you're not going to use them -- or at least turn off the monitor.
Use an old computer -- they tend to use less power. The 60 MHz Pentium chip only used 13 watts -- compare that to modern cpus using almost 100 watts! (And people freaked out about that 13 watts -- it was unheard of at the time.) Older cpus use even less power. G4 PPC chips tend to use less power than their x86 equivilents (but the G5 ones seem to be more comparable.) Use a laptop rather than a desktop -- they tend to be more efficient. - dooms13, on 10/11/2007, -2/+5Whoops my bad...digg me down above, thats bad info...
Work = Charge * Potential - Yonson, on 10/11/2007, -2/+5You would be correct. For a parallel plate capacitor, W = qVt. I think there's a better way to express work done by an electric circuit but I can't seem to find it.
- valkyries, on 10/11/2007, -0/+3turning transistors off and on creates heat. during the state that its changing from off to on(also on to off) is when it creates the most heat.
- dimension128, on 10/11/2007, -1/+4@Juaquin,
Are you saying that the energy just kind of disappears? I think there is some kind of basic law in science that says thats not possible, you may want to look into it.
Let me provide one example, you inquired about the fans. Well first of all, the motor, or whatever drives the fan, does put out heat. The blades move air, thats friction. In fact, every bit of energy that goes into your 80mm fan eventually is dispersed as heat through friction. - coviecarbine, on 10/11/2007, -1/+4Not only that, I'm sure that plenty of electricity is wasted just by computers idling (even if they are pretty efficient when they aren't doing anything). I'm sure that I'm not the only one who leaves my computer on overnight.
- rabidjade, on 10/11/2007, -9/+11If the processor companies would worry about efficiency rather than the next biggest, baddest marketable processor, that would solve a lot right there. I forgot the number but there are so many wasted transistors in a newer processor, that it would be unheard of 5 years ago. Having to use a 10lb copper heat sink on a Pentium4 compared to the 8oz aluminum one on a Pentium 2 should be enough to show anyone. The hotter a processor gets under normal load, the more inefficient it really is.
- bbear, on 10/11/2007, -2/+4Peak efficiency is not achieved when a PSU is fully loaded. Peak efficiency happens around 70-75% load but it varies by power supply. Fully loaded is actually the most inefficient state that a power supply can be in. Some review websites list PSU efficiency under different load levels so you can see the difference. The average PC is using a PSU rated at 60% to 70% efficiency. On the other hand some Seasonic PSU are rated over 90%. Seasonic isn't the average PSU. Expect to pay a premium.
- ninjasquirrel, on 10/11/2007, -2/+4No, your PSU is not at 88% efficiency. PSU's are rated at their highest possible efficiency, which is almost always achieved when they are fully loaded. It is very unlikely that you have your PSU constantly running fully loaded, as that would seriously decrease the lifespan of all of your components (more load generally means more electrical "noise", which decreases the life of your components). The average PC, under average use, is probably 60-70% efficient at most.
- Rileyper, on 10/11/2007, -0/+2How else would you make use of your flux capacitor
- damentz, on 10/11/2007, -0/+2But its got electrolytes.
-
Show 51 - 100 of 117 discussions
The Digg Toolbar for Firefox lets you Digg, submit content, and keep track of Digg even when you're not on the Digg site. Download the official 
© Digg Inc. 2009
— Content created and posted by Digg users is