Curious

[don_svetlio]

What are your guys' GPU core and VRM temps under load usually? Just curious.

[TheKDub]

Around 60-72c.

[Nonco]

Unless if its over 80-85 C then got nothing to worry about.

[don_svetlio]

Around 60-72c.

So VRMs being warmer than the core is fine for both AMD and Nvidia? My open case results were 58-63 (core and vrm respectively)

closed case it's 69-72

[CostcoSamples]

Mine are usually 60 to 65.  Nothing goes over 70, but I have 5 x 140mm case fans for massive amounts of airflow.

[don_svetlio]

Mine are usually 60 to 65.  Nothing goes over 70, but I have 5 x 140mm case fans for massive amounts of airflow.

Yeah, I use to have 68-65 but that was before I had to reverse airflow cause of the CPU cooler not making nice with RAM.

[incarnate]

72 degrees is fine for vrm temps. a lot of vrms are able to handle temps of 90 or sometimes more. i've never checked the vrm temps on my graphics card though, however it is actively cooled by the cooler.

[don_svetlio]

72 degrees is fine for vrm temps. a lot of vrms are able to handle temps of 90 or sometimes more. i've never checked the vrm temps on my graphics card though, however it is actively cooled by the cooler.

Are they? I assumed that most cards had semi-passive VRM cooling since they usually have a different heatsink that gets cooled by downdraft from the main heatsink

[incarnate]

Are they? I assumed that most cards had semi-passive VRM cooling since they usually have a different heatsink that gets cooled by downdraft from the main heatsink

 

a lot of cards cool the vrm the same that they cool the vram. they don't usually have heatsinks on them and just have them cooled by any airflow that makes it through the cooler. the EVGA 970 SC and FTW for example, had no heatsinks on the vrms or vram, but they added a metal plate to act as a heatsink for them for the SSC and FTW+.

[don_svetlio]

a lot of cards cool the vrm the same that they cool the vram. they don't usually have heatsinks on them and just have them cooled by any airflow that makes it through the cooler. the EVGA 970 SC and FTW for example, had no heatsinks on the vrms or vram, but they added a metal plate to act as a heatsink for them for the SSC and FTW+.

So it's fine if they run 3*C higher under load when the front fans are exhausts.

[incarnate]

Yeah, they can usually go up to 90-100°. Anything higher and it is time to start getting concerned

@don_svetlio

[Badger906]

75 to 83c and then my 980ti throttles back

[dougdangger]

60s

[don_svetlio]

Yeah, they can usually go up to 90-100°. Anything higher and it is time to start getting concerned

@don_svetlio

 

75 to 83c and then my 980ti throttles back

 

60s

so 69-71 is okay, btw, what's more likely - furmark damaging VRMs (I did stress test my new PSU for 40-50 mins) or the reverse airflow pushing warmer air through their heatsink and thus 3*C higher temps.

[incarnate]

so 69-71 is okay, btw, what's more likely - furmark damaging VRMs (I did stress test my new PSU for 40-50 mins) or the reverse airflow pushing warmer air through their heatsink and thus 3*C higher temps.

Futuremark has a reputation of frying graphics cards. 3°c isn't going to do anything

[don_svetlio]

Futuremark has a reputation of frying graphics cards. 3°c isn't going to do anything

I don't think it fried anything, I hope. I've been told drivers throttle it.

[incarnate]

I don't think it fried anything, I hope. I've been told drivers throttle it.

Honestly I think it's just high overclock that cause futuremark to fry gpus

[don_svetlio]

Honestly I think it's just high overclock that cause futuremark to fry gpus

Mine hasn't been OCed beyond what Asus did and voltage as well as power thingie is on 0 and stock.

IIRC there is a buffer built into the VRMs