Gaming PC vs. Space Heater Efficiency
Written on October 21, 2013 by Matt BachIntroduction
Winter is coming, and with the power bills stacking up it may be hard to justify gaming on your ultra-powerful gaming rig. But what if we were to tell you that, watt for watt, your gaming PC produces exactly the same amount of heat as a space heater? Suddenly you don't have to feel like your gaming habits are costing you tons of money during the winter. If the same amount of power is used either way, you might as well be entertained while your power bill racks up!
You may think that a space heater, which is created solely for producing heat, would be better suited to the task of warming up a room than a computer. But when you think about it, nearly everything that a computer does ends up creating heat. And if you have ever touched a high-end gaming PC after an intense gaming session, you know that computers can be very good at generating heat.
So to find out if a gaming PC is really just an oversized space heater with the advantage of providing entertainment, we decided to compare the heat output of a common space heater to a high-end gaming rig.
Test Setup
To compare the thermal output of a space heater and a gaming PC, we used one of the many small space heaters we have around the office and built a gaming PC that drew roughly the same wattage. The space heater we used was a $25 HC-0114T 1000/1500W space heater that can be found under various brand names. When we measured the power draw, we found that on the low setting it pulled 870-890 watts depending on how hot the heater was. This is a lot of power and we ended up needing a gaming PC with three NVIDIA GTX Titans in triple SLI running Furmark to match the power draw.
| Testing Hardware | |
| Motherboard: | ASUS P9X79 Deluxe |
| CPU: | Intel Core i7 4960X 3.6GHz Six Core |
| CPU Cooler: | Corsair Hydro Series H60 CPU Cooler (Rev. 2) |
| Video Card: | 3x NVIDIA GeForce GTX Titan 6GB |
| PSU: | Seasonic X-1050 1050 Watt |
| RAM: | 4x Kingston HyperX DDR3-1600 4GB Low Voltage |
| Hard Drive: | Samsung 840 Pro 256GB |
| OS: | Windows 7 Ultimate 64-bit |
| Chassis: | Fractal Design Define R4 |
| Monitor: | Asus VE248H 24 Inch LCD Monitor |
During our testing, we found that both the PC and space heater drew varying amounts of power depending on how hot they were and how fast the fans were spinning. To try to keep both the space heater and PC drawing the same wattage, we took regular power draw readings and used MSI Afterburner to adjust the GPU power draw on the gaming PC to match the space heater's power draw. With this method, we were able to match the PC's power draw to the space heater's within a few watts and ended the tests with exactly the same total kWhr of power used.
To compare the thermal output of both devices, we put both the space heater and PC (with the keyboard, mouse, and monitor) in a closed room and took continuous measurements of the air temperature within the room. This room measured about 10' x 10.5' with a total volume of just under 940 cubic feet. All incoming and outgoing ventilation was blocked off to keep it completely isolated from our office's HVAC system and the surrounding office spaces were held at a constant 70 degrees Fahrenheit.
Results
As you can see from the chart above, the gaming PC and space heater performed very close to each other. The times when the gaming PC pulled ahead were when the wattage draw of the PC climbed a bit higher than the space heater before we could manually adjust it down. We tweaked the video card power settings every 20 minutes, but due to how often the fans decided to ramp up or down the results were not quite as straight forward as the space heater results.
Even with these slight variations, the results are close enough for us to confirm that for all intents and purposes a PC and space heater will output the same amount of heat when drawing the same amount of wattage from a wall outlet.
Conclusion
We want to be clear that if your only goal is producing heat, purchasing a space heater is going to be many times cheaper than purchasing a gaming computer. Especially at the wattages a space heater operates at (usually around 1000 watts), a PC is going to be very expensive.
However, if your house is freezing and you already have a high performance gaming computer - or you simply want to know that your gaming addiction isn't actually increasing your power bill in the winter- our testing shows that a PC is just as efficient at creating heat as a space heater. In fact, you could even set the computer to mine bitcoins to make a heat generator that helps pay for itself! So when you wake up or get home to a freezing cold room, start up those games, folding@home programs, or benchmarks to get your room nice and warm. Its just as efficient as using a space heater, and much more entertaining.
I use both of my desktops and server(s) to keep my apartment heated during the winter. Heck, I have to open up my door and windows regularly to cool my apt off.
exactly the same here, central heater is almost always of. my computer and server manage to make it more then hot enough for me
All you need is an AMD card
And yet, its an intel chipset they tested this with.
well he said card, implying AMD GPUs because they output more heat than others, but whatever
Through titans will heat more than fury x[due to stock water cooling]
If you use watercooling or not doesn't matter. The room will still be the same temp regardless of if you're using watercooling or not.
Actually, watercooling tends to warm up your room a little bit more than air
Correct. Because it's quieter. All the sound will eventually turn to heat via friction, but any sound escaping the room goes through that process outside of the room and never heats it. Therefore quieter will be slightly more efficient at heating the room you are in.
Can confirm my 7970Ghz got the job done during Christmas of 2013 while it was mining doge coins for me.
:D
Crossfire dual 390Xs FTW!
And what about summer?
Love this article!
this article is full of win.
My favorite "To try to keep both the space heater and PC drawing the same wattage, we took regular power draw readings and used MSI Afterburner to adjust the GPU power draw on the gaming PC to match the space heater's power draw. With this method, we were able to match the PC's power draw to the space heater's within a few watts and ended the tests with exactly the same total kWhr of power used."
It actually took us a bit to come up with this idea. At first we just tried changing the benchmark we used and adjusting the settings, but we could only get to within 10-15 watts of the space heater. By using MSI Afterburner to set the maximum power for the video cards to 94% then tweaking it a percent or two during the test we were able to get the PC to almost exactly match the space heater throughout the test.
It would be interesting to see the comparison given actual gaming scenarios instead of matching the power draw. As this XKCD article points out, when you match the power draw you are always going to get the same heat generation: http://what-if.xkcd.com/35/
Yep, thats exactly what we expected, but we have never seen anyone actually test it before. As for actual gaming scenarios, its all going to depend on how large your power draw is.
Anyone that has ever been in a tiny dormroom with 2 or more gaming setups could have old you this easily lol. Its the the nerd equivalent of a ritualistic sauna.
Or anyone with an AMD rig *waits for the hate train*
*cough*fermi*cough*
i couldnt agree more. every ati card ive ever owned has been a heater for sure. in fact i swore them off after having to run my last one at 70-100% fan just to stay below 80c under any load. nvidia runs so much cooler its just crazy and ill gladly pay the price difference for this alone
I laughed a little when I saw the email topic. Can a gaming PC kick out as much heat as a heater? Maybe a tiny little heater, sure. Since the dawn of NVIDIA SLI or CrossFireX systems, I've never had a 2-way or 3-way gaming system put out the same BTUs. However, I did find myself not worrying about turning on the heat quite so much when the system was down on the floor blowing heat under the desk. If nothing else, it's more of a personal area "warmer" and much cheaper than turning on your heat because you KNOW you're going to game no matter what. :)
Awesome job!! I have always wondered about the heat a desktop gaming computer could produce, really liked the whole investigation and write up. Keep up the great work!!
Those little heaters like that are very inefficient. The larger oil heaters heat up a room much better while using less energy.
This is not the case, oil filled heaters are still resistive heating devices. A heat pump, however, is more efficient.
Pshhhh guess how much electricity my volcano needs? I recommend you invest in getting a lava well drilled if you really want efficiency. I highly recommend a sprinkler system, and home owner's insurance.
How can a heater not be 100% efficient? The only way it could not be is if some of the input energy went into another form of energy like sound or light, and I am pretty sure little heaters do not make make much noise or light. They also do probably not vibrate much either.
Therefore all heaters have nearly the same efficiency at turning electrical potential energy into heat and this article should only be surprising to those who don't know about the law of conservation of energy.
I wanted to comment that heat pumps can have values of efficiency above 100%. While this appears to break the laws of thermodynamics, it actually doesn't. It's somewhat complicated, but the base take-away is that heat pumps do not *create* energy, but rather *move* energy. With this, you can 'move' more heat than the energy it took.
Consider the reverse idea of an air conditioner. It takes a LOT of energy to cool the air, enough that it's not '100% efficient' to cool the air, per se. However, an amount of heat is also being produced, by moving the heat from the inside to the outside. That same heat is benefitting from the inefficiency of cooling, and resulting in a >100% efficiency in 'creating' heat. If, that is, you compare it to direct heat generation, as opposed to moving the heat itself. You're still not creating more energy than you started with, just moving energy from one spot to another. However, in energy terms, you're moving more heat than you could have generated with the same amount of energy on a 1:1 basis.
This is how heat pumps are more efficient than direct heat creators.
Why make it so complicated? A heat pump is "more than 100% efficient", because it takes heat from elsewhere (e.g. outside land or air), and adds it to your house environment. Simples.
"Those little heaters like that are very inefficient".
The only concept that this is remotely true might be in the heat output vs. the COST of the device to buy.
When converting electricity to heat, basically every device has the same efficiency (assuming no losses to other forms of energy - e.g. sound or light.
You can't play Crysis on a space heater. PC wins.
Because we all can afford 3 Titans...
1 titan= my whole pc setup :|
SOLUTION two r9 290x in crossfire will generate ATLEAST as much heat
Lol 3x GTX Titans.....
use overclocked 8350 and overclocked 7950 or higher, that will do the trick :) both for gaming and heating. price of entire rig: approx 700$
ever since Einstein shat out the equation e=mc2, it has been common knowledge that energy is always conserved, and never created nor destroyed....
given this, without even doing any experimenting, you should have been able to guess that 1000 watts of electricity equals 1000 watts of electricity. unless that electrical energy gets turned into something other than heat (like light or physical work/motion), it makes no difference what device actually converts it into heat; its still the same conversion process.
had your data collection been more accurate, you would have been able to actually measure the (small) energy loss due to light emitted from your computer screen. subtract this amount from the total measured input energy, and the remainder is how much energy is used to generate heat.
I'm not sure that e=mc2 has much to do with the first law of thermodynamics, and both the PC and the space heater also give off light and sound. However, I agree that this SHOULD be intuitive...but it is always useful to make sure the empirical testing results match your theory, right?
The light energy given out by your PC is absorbed by the surfaces in your room and converted into heat energy that warms the room.
The kinetic energy of the wind from the fans is eventually turned into heat energy due to friction with other air molecules, which heats the room.
The sound energy from your PC is, for a portion, absorbed by the sufaces of your room and turned into heat energy; the other portion becomes heat in the air due to friction between air molecules; all these warm the room.
True that sound and light both will eventually end up as heat energy. But what you're missing is that they're both capable of leaving the room to some degree BEFORE turning into heat energy. So any heat you create directly is 100% efficient, but any light/sound you create is not 100% efficient unless you've got a 100% blacked out and soundproofed room. So his point was that the light/sound does technically lower efficiency as far as turning electricity to heat in the room.
As someone who does not understand physics and doesnt know what E=MC2 actually means, I found this article very interesting and helpful. Thank you :-)
Conservation of energy was discovered long before Einstein in 1842. e=mc^2 equates mass(m) and energy(e) (c^2 is the speed of light squared), it says nothing about the first law of thermodynamics.
my old mining rig (gpus are useless now with fpgas on the market which are already obsolete, and asics the only thing that matters) kept my whole house warm. there was a paper published a few years ago about the heat efficiency per watt of ICs, which put them as vastly more efficient on their own (but lots of heat free idle transistors in any given IC) than any other electric heater. not to mention that they are much more expensive to manufacture. imagine if you had the same volume of ICs as in an electric heater? it would be like a jillion times more heat. you get something the size of a thumbnail putting out enough heat to warm a room, vs a baseboard heater or space heater. incredible
and yes, we all know energy is conserved. the light emitted (in the red/orange spectrum is not heat), nor are other conversion factors. the reason why a pure element is super close to efficient, but a transistor is closer relies on the light aspect. a space heater gets orange, wastes em radiation in the visible spectrum. transistors are kept non-visible, which is a higher percent. if my brain is working anyway. which there is a good chance its not.
AMD cpu/gpu would have been cheaper and more effective at heating.
Love this!
I've always picked components that have a good performance per watt because I have the opposite problem - I want to keep things cool. I live in Australia and I have to try and stop my room from turning into a furnace when I turn on the computer. This information isn't always obvious in reviews though.
I don't think the test might have been totally accurate, the energy going into the computer and the heater should be absorbed by the enviroment as heat. What might help the computer against a heater is that it uses fan to push the heat around and slightly better airflow in the chanber.
Compared to an electric space heater, a computer would do just as well, the efficiency is 100% and it doesn't really matter what you use.
Compared to a central heating (on gas for example) it is a bad idea to use a computer solely for heating, as the central heating has cheaper fuel and higher efficiency compared to how the electricity was made.
Gas has about 12 times more energy per cubic meter than 1kwh. while it is only 3 times more expensive. so using an electric heater instead of a high efficiency central heater, is about 4 times more expensive. Might be even more if you can get cheap gas.
Awesome, would love to get this one!
custom gaming pc
Established in Australia since 1992, Adelong is a Family run Electronics & Computer retailer. well located in the centre of the Sydney CBD, behind the iconic Queen Victoria Building,
You forgot to add that there are many -WAY- cheaper alternatives to heating than electrical...
Congratulations, you've validated the laws of Thermodynamics. You've could have just looked them up on Wikipedia instead. ;)
Nice! Do we win a thermodynamic prize?
No, he validated that most of the energy turns to heat while still in the room. If you can hear the fan from the other room, that's energy that escaped the room before turning into heat, and is heating somewhere else instead. If you can see the light from outside the window, then that's energy that escaped the room before turning to heat as well.
We know that all the electricity consumed will turn to heat...but we didn't know how much of it turned to heat while still in the room. Showing that the noise/light that leaves the room before turning into heat isn't enough to have a drastic effect on our purpose of efficiency is a worthwhile test.
Lucky B*s*a*d, three!!! w...t....f.... What do you need three Titans for?
Repeat the test with FX + CF 390
The HVAC went out in the office for a few days last winter... and my Quad Opteron (running thousands of iterations of IntelBurnTest) helped keep things reasonable :)
The cold is unbearable and while working at my pc, Egloo is
ideal to warm up with only heat of 3 candles! http://eglooinfo.it/
A pc is much less efficient at heating a room if consider a tiny space heater is on average 2000W and the average PC is 500W, and the PSU inside it is only between 50-80% efficient, so a space heater will warm a room up 4x to 8x faster than a pc based on just this. Furthermore, heat is stored and dissipated through solid medium such as heatsinks, metal casing and components, stay inside the pc as whole it is not efficient to heat a room this way. So it is therefore much more efficient and cheaper to use a space heater. I don't know how they can conclude otherwise. P.S. Bitcoin mining using GPU's in PC's did not pay off since 2011.
what're you smoking there is NO way 4 Titans have a PSU under 1000 watts
I said average pc. Not 4 titan. Can you read?
Or just get an older computer like the Dell OptiPlex GX280 and run it for a few hours to bump up the temperature a few degrees. (Which would be much less effective at heating.)
what if you used an intel pentium extreme overlclocked to 4.6 GHz
Alternatively, buy a high end bitcoin miner. They produce heat, and pay for the electricity used in bitcoins!
except you don't game on a heater
Oh great. So I have to factor a second air conditioner into this purchase.
Intel obviously read this article and decided to enter the space heater market with their new 7980XE.
Matt, its not clear how you are defining efficiency but I can state with certainty that both are 100% efficient as electrical heating devices. This just means that every watt of power that the units draw will be converted to heat, that includes the sound from the fans (turns into friction heat in air) and the kinetic energy imparted on the air that the fans put out (also turns into friction heating the air). What is not discussed in great detail instead of the hidden benefit of "free heat" when gaming in the winter is the penalty taken by doing this in the summer in climates where the house is mechanically cooled or just the penalty of leaving the computer on to idle. This happens to many people without them knowing what it's dumping into the space whereas no sane person leaves their electric heater on during the summer.
That is a good observation, Tyler - I can definitely feel the added heat in my bedroom when gaming during the warmer months. Thankfully it is nowhere near the ~1000W triple SLI rig in this article, but it certainly has an impact on the heating and cooling requirements in my home. At idle it isn't as bad, of course, but I suppose it is still a factor. I wonder if there is a handy rule about ambient heat increase per watt... probably needing to take into account the total volume of a room, etc. Something to think about, certainly :)
I have a stack of 6 HP pentium 4 desktops that keep me warm in the winter. At idle, they draw about 450 watts combined. That's about 1/3 of a 1500 watt baseboard electric heater. It is more efficient, and also a more practical use of electricity to heat using computers, game or not. And much quieter. And much safer.