|
On June 16 2011 00:56 LolitsPing wrote:Show nested quote +On June 15 2011 22:16 Womwomwom wrote:
That's probably just a case of coil whining from the GPU/PSU/motherboard. My GPU does it too whenever I boot up STALKER or Crysis when those intro scenes get like 1000FPS. actually now that I think about it, I've been hearing some click noises when I'm playing and it freezes temporarily before it completely freezes. I'll open the case and try and listen for the noise.
Clicking noises could be HDD trying to fail. Download a diagnostic tool from the HDD's manufacturer website and run a SMART check.
|
Hmm I tried it and it says that the test is not available... hmm
EDIT: Tried it with a different program. Apparently, my HDD is not SMART compatible...
|
On June 15 2011 17:01 Womwomwom wrote:Show nested quote +If your GPU is hitting 73c @ 1280 x 1024, the temps will be even higher at higher resolutions How would it. If a card is at load, its at load and its going to put out the same temperatures no matter what resolutions are provided the load stays roughly the same which it should.
You're not serious are you?
It's like saying, "If a CPU is at load, it's at load and its going to put the same temperatures no matter what"
but a CPU @ 3Ghz w/ 100% load will produce less heat than a CPU @ 4Ghz w/ 100% load when you run Prime95 or IntelBurnTest
Same thing with a GPU, a higher resolution is a higher stress level, therefore a higher temperature.
|
On June 16 2011 09:49 yks wrote:Show nested quote +On June 15 2011 17:01 Womwomwom wrote:If your GPU is hitting 73c @ 1280 x 1024, the temps will be even higher at higher resolutions How would it. If a card is at load, its at load and its going to put out the same temperatures no matter what resolutions are provided the load stays roughly the same which it should. You're not serious are you? It's like saying, "If a CPU is at load, it's at load and its going to put the same temperatures no matter what" but a CPU @ 3Ghz w/ 100% load will produce less heat than a CPU @ 4Ghz w/ 100% load when you run Prime95 or IntelBurnTest Same thing with a GPU, a higher resolution is a higher stress level, therefore a higher temperature.
Actually, you're only half right. A GPU at a higher load will have a higher temp, but it's not guaranteed that a higher resolution will cause a higher load. If nothing caps frame rates, and there's enough effects/eye candy being used to be using similar memory bandwidth, the card can hit similar loads.
And your CPU comparison is so off base it's not even funny, unless you meant to say identical CPU's at different clocks requiring different voltages.
|
Yes, I meant identical CPU's, sorry I did not specify. And yeah, the analogy is a bit iffy.
But it still stands. 2 identical GPU's will have different temperatures pending on the resolution and both on full-load.
I've ran 3DMark benches a ton @ different resolutions and have always seen different temps.
A/B comparisons of the same card is what I normally bench for.
edit; both stock, overclocked, and the only aftermarket gpu-air cooler i've tested was the thermalright spitfire. different resolutions at full-load has always led my tests to different temps.
|
That's because you ignored part of what I said. If they're both under full load with identical memory usage, they're going to hit the same temps, or very nearly. The issue is you frequently won't be hitting identical VRAM usage at different resolutions. I would expect it to normally take synthetic conditions, but I put nothing past the power of PEBKAC either.
Bear in mind, a lot of non-reference cards have memory cooling, not just GPU, that's why the VRAM saturation makes a difference to temps under load.
|
I think we're talking too much with too many unknown variables. Lets take PEBKAC out of the situation, and say you yourself were doing the tests.
Let's say, one computer with a 1gb 5870 running SC2 @ Extreme on 1280x720
And then the exact same computer but it's running SC2 @ Extreme on 1920x1080.
In both situations, a 4v4 with 200/200 armies from all 8 players attack, occurs.
Which 5870 will have higher temps?
edit - stock everything. same ambient temps.
|
On June 16 2011 10:43 yks wrote:
I think we're talking too much with too many unknown variables. Lets take PEBKAC out of the situation, and say you yourself were doing the tests.
Let's say, one computer with a 1gb 5870 running SC2 @ Extreme on 1280x720
And then the exact same computer but it's running SC2 @ Extreme on 1920x1080.
In both situations, a 4v4 with 200/200 armies from all 8 players attack, occurs.
Which 5870 will have higher temps?
Are we assuming CPU isn't a limiting factor, are we using vsync or some other form of framerate cap? It's just too complex of a question for a simple answer, that's the point I'm trying to make.
Maybe we're getting too in-depth and ridiculous for the question though...
|
I don't think it's that complicated at all.
If you run 2 sets, the first pair with v-sync, the second pair without v-sync. And in all scenarios the CPU isn't a limiting factor. The answer will still be the same.
Any A/B comparisons you want to make with the any settings you want on both computers as long as the settings are identical and the only variable is the resolution. The temps will be higher on the higher resolution. That's the point I was trying to make.
edit; just saw your edit, yeah we probably are. lol!
|
|
|
|
|
|
EPT
