• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Gaming PC vs. Space Heater Efficiency

We remodeled our basement in 2009. Before, it has uninsulated concrete-and-single-thickness-wood walls and single-pane, aluminum-frame windows. You could feel the heat being sucked out of you. I could run my servers 24/7 and it stayed cold - even with the furnace outputs all open. (The basement was a large open space at the time.)

After the remodel, (good wall insulation with an interior wall, high-efficiency double-pane vinyl-frame windows,) I decided to test. In November, I closed the door to the newly-walled-off office (10x20 room, no heat vents, two large windows plus a sliding glass door, all had been replaced with the high-efficiency,) and let my server (dual 130W Nehalem,) run SETI@Home for a 3-day weekend. Came home to find the home office HOT.

Now I can no longer run my servers at full-bore during the Summer. But I happily run them in the Winter - might as well get some useful computing done while keeping the basement at a reasonable temperature, while letting the furnace heat the main floor.
 
My little office easily gets up to 90 degrees if I have both systems running.

Because of the insane cost of energy and home heating this winter I'm seriously considering running a small fan run duct from my office straight downstairs over the balcony and pointing it at the living room. It makes way more sense than either opening a window and wasting the heat or running the in-window AC off and on to reduce the temp.
 
Kor said:
And this is why I like energy efficient parts :p
Click to expand...

It is actually amazing how much they have improved in that area. Still if you would be paying to run your heater anyways, might as well run PC and have them work on some noble project. Either way you will be paying the same to the power company/fuelsource. This way you can have a free byproduct ...
 
J So said:
I was going to write "In before better heatsinks or water cooling keeps your room cooler" but seems I'm too late.
Click to expand...
Ironically, more efficient cooling just increases the rate at which the surrounding environment's temperature increases (albeit very slightly) as heat energy is more effectively dissipated to the air. And water pumps typically consume between 5 and 10 watts themselves, adding only more energy as heat to the equation.

GoodBoy said:
Well, it's a legitimate question.. The computer does "work", computing, etc.
Click to expand...
The work computers (or, more specifically, their processors) perform is really just the moving of electrons from one place to the other. You have light-emitting components and components with motors and so forth doing their own work, but on the actual computation side, there's very little work being done.
 
A watt isn't a watt when it comes to heat output though. A 10 watt incadescent lamp will release more heat than a 10 watt LED for example, because heat is only ONE way that the energy can be transformed. Light is the obvious other one in that instance, and the LED will release more of that energy as light than the other lamp. Likewise, a water pump should be releasing little heat and more mechanical energy.
 
BTW, for those of us considering duct work in our homes to deal with this problem, perhaps going mostly water cooled and using a passive cooling tower in another room is the way to go. You'd then only need to make a very small passthrough in the wall, and I don't think that would be against any code because its just a tube of liquid... *shrugs* Hmmm...
 
Ducman69 said:
A watt isn't a watt when it comes to heat output though. A 10 watt incadescent lamp will release more heat than a 10 watt LED for example, because heat is only ONE way that the energy can be transformed. Light is the obvious other one in that instance, and the LED will release more of that energy as light than the other lamp. Likewise, a water pump should be releasing little heat and more mechanical energy.
Click to expand...

Conservation of energy. Energy cannot be created or destroyed in a closed system. So unless the light can escape the room, a 10W LED will output the same heat as a 10W incandescent, it's just the LED will produce a lot of light first, that light will strike objects, get absorbed and end up as heat (unless the light can escape the room or reflect off surfaces indefinitely). The incandescent will output most of it's energy as heat initially.

The same with a pump, unless you're pumping fluid outside the room, the mechanical energy eventually gets chewed up by friction which is heat.

If 100W of power goes in to your computer and your computer isn't pumping the heat to another room, 100W heat will come out of it, regardless of whether it's going in to pumps or fans or circuit boards or anything else (unless it's being stored as potential energy somehow).
 
Rahh said:
Sounds like some in this thread need proper cpu cooling installed.
Click to expand...

LOL, go back to school man. You obviously need to take some remedial science classes. It doesn't matter if you can keep your CPU at 0 degrees celsius, the entire point of cooling it to get the heat away from the CPU so it's going to heat up the room it's in anyway. How can you post on hardocp of all places and have zero understanding of cooling?
 
Not having read it yet, I'm guessing that electronic chips are probably pretty efficient at turning electricity into heat. They do very little work in the physical sense, in fact any work is an inefficiency of the silicon, so they probably are more efficient that typical space heaters which have limited materials selection in order to keep costs low enough that someone would actually buy a space heater, since nobody's going to plonk down several hundred on a simple space heater.

Now I'm going to read and see how my engineering assumptions fare.
 
So... wow, that was less technical than I had hoped, but basically results were no surprise to me.
 
I would like to see future [H]ardOCP reviews and benchmarks include room heating benchmarks against competing electric and oil heaters. This would help give us ideas about different configurations of hardware.

Unlike a heater, computers can actually accomplish real work. ;)
 
wonderfield said:
Ironically, more efficient cooling just increases the rate at which the surrounding environment's temperature increases (albeit very slightly) as heat energy is more effectively dissipated to the air. And water pumps typically consume between 5 and 10 watts themselves, adding only more energy as heat to the equation.
Click to expand...

More eifficient cooling will not affect the ambient temperature unless the cooling solution itself requires some outside power (more powerful fans, more powerful water pump, phase change cooling etc)

More eifficient cooling reduces the temperature that the chip can be at, while still dumping the same amount of heat into the environment.

You can make the argument that hot chips have different power requirements then cold chips, but I dont know much about that.
 
I was at one point wondering if it would be feasable for some company to make a self contained unit that plugs into power and an ethernet.

It would be marketed as heater, but also happens to do bitcoin/folding/boinc calculations. If it did paid calculations then either the customer could get paid every year for running it or the company would get the profits.
 
Concillian said:
So... wow, that was less technical than I had hoped, but basically results were no surprise to me.
Click to expand...

Indeed. I could easily attest to this as before I moved I was in a pretty small room yet all my electronics (plus a beefy full tower desktop) could heat my room to around 80F while it was around -5F outside.

Interesting test but already confirmed by thus of the [H] lineage.
 
serpretetsky said:
More eifficient cooling will not affect the ambient temperature unless the cooling solution itself requires some outside power (more powerful fans, more powerful water pump, phase change cooling etc)
Click to expand...
It will indeed. An enclosed space will not become hotter as a result of a more efficient cooling system, but it will reach a given higher temperature more rapidly. As I said, though, only very slightly.

This doesn't take into account the effect operating temperature has on IC power efficiency, but that's slight as well.
 
GoodBoy said:
Well, it's a legitimate question.. The computer does "work", computing, etc. So it's understandable people being curious how much energy that might use. There's probably some energy lost in the fans/wind turbulence too, but that helps distribute the heat. It's expected there to be some small amount of energy loss, and they tried to determine how much.
Click to expand...
Nope, power in is power out, if it's moving air with a fan that's moving molecules around (the definition of temperature). The difference is in the losses of heat to the environment outside the room/house, now it could be that an case fan blows that hot air farther away from you and as a result you don't notice the temperature as much, but that power is getting out there, none of it is "lost" by doing "work" i.e. computations.

The cool thing about their findings is, if they can find a way to engineer a CPU that doesn't waste nearly as much energy as heat (as current cpu's do), we can have really high powered computing that will run on just a few watts.
Click to expand...
A 100watt CPU will dump 100 watts as "heat" regardless if it's made 20 years ago with that technology or today. CPUs don't "make as much heat" because they are able to do the same thing with less power and that's it. As you said a CPU running a few watts, well then it's only a few watts worth of heat it'll make because it's only using a few watts of power not due to any "efficiency".
 
The monitor and speakers draw power too. Without those, the computer is boring unless mining bitcoins.

The could have given up o video card maybe.
 
Kongar said:
Efficiency is a funny thing depending on your definition. Ultimately, most things are 100% efficient at converting energy to heat. Heat = vibrations/motion of molecules. But realistically - a PC isn't 100% efficient at direct conversion of electricity to heat. Some of that energy goes toward fan motion, light generation on your monitor, battery and capacitor recharge, etc. But give it long enough and ya, all of that eventually translates into molecular motion somewhere outside the system boundary of the PC.
Click to expand...
If by "long enough" you mean a few seconds. :p How long does it take a fan to stop spinning and the air to become still after you turn it off? That's about how long it takes for the energy to go electric -> mechanical (fan) -> fluid (air) -> thermal energy. The reason the air stops moving when you turn off a fan is because the air has viscosity (friction) so that motion gets converted to heat after a few seconds.

Light generated on your monitor gets converted to thermal energy pretty damned fast, if just has to strike 1 or 2 objects and boom, it's now mostly thermal energy.

The only one is battery and capacitor charging, that's actually storing the energy as potential energy. But that energy gets released as well otherwise, ya know, the batteries and capacitors would pop.

The only real factor in non-heat pump electric heating is how aggressively the heat gets moved in to the air. That's what can make one heater feel good and another feel bad. Not enough fan blowing and the heat gets converted to thermal energy and dissipates close to the heater rather than where the humans are. Too much fan and the air gets stirred up too much which causes the heat to be dissipated faster throughout the room and to the surroundings.

Of course, heat pump heaters can actually be much more efficient, because instead of generating heat they pull heat from elsewhere, so they can actually dump more heat in to a room than they themselves require to run.
 
You must log in or register to reply here.
Back
Top