Low PPD on some WUs

[Bryman]

I do have some case fans I could snap onto my case... but it would make more noise... I prefer just to have my video card about 5°C hotter and have the side of my case open with no case fans

I think your GPU fans make more noise at 100% than some case fans which you can even throttle from 12V to 7V.
You can also use a case fan and but it directly near the GPU using some wire tie and still keep your case open.

You can also change your GPUs fan curve using a software tool, so it does increase fan speed earlier.
I think MSI Afterburner allows that.

Eh... my video card is usually 80... that's perfectly fine... well at least the core temp is okay

Not sure how only having one fan running makes it

You definitely need to improve your card and/or case cooling

Heh... my motherboard is only reporting about 45... which means that the air inside the case can't be more than about 30-35

But my video card doesn't have power limit throttling... so when running folding@home plus stress testing it uses about 350 watts, and when running just stress testing it uses about 400

The TDP is 250

So... yeah


My video card went from 33 to 97 degrees in 45 seconds with 0% fan speed and 33 to 88 degrees in 60 seconds with 100% fan speed

I was running the video card fan at 100% for over 5 minutes with my video card idle before the stress test

With just stress testing if my card never throttled the core temp would go well over 115...

But the VRM would blow up before that happened... it would probably hit about 180°C by the time the core hit 115

[bruce]

Why does folding@home disable throttling when the VRM temp goes up...

I do not believe that FAH disables throttling. FAH officially does not support overclocking. That implies that their software doesn't talk directly to the drivers. They use the OpenCL interface in a way that's independent of the manufacturer of the GPU or code distributed by them.

My understanding is that the various limits defined by the hardware vendor and tweaked by overclockers all have a priority. If you reorder those priority settings you should be able to find a combination that achieves what you want. The fact that FAH stresses different components of the GPU in a different way than whatever you're using to benchmark your overclocking simply indicates that you're letting it exceed whatever limit should be doing the throttling.

... and Good Luck.

[Bryman]

Why does folding@home disable throttling when the VRM temp goes up...

I do not believe that FAH disables throttling. FAH officially does not support overclocking. That implies that their software doesn't talk directly to the drivers. They use the OpenCL interface in a way that's independent of the manufacturer of the GPU or code distributed by them.

My understanding is that the various limits defined by the hardware vendor and tweaked by overclockers all have a priority. If you reorder those priority settings you should be able to find a combination that achieves what you want. The fact that FAH stresses different components of the GPU in a different way than whatever you're using to benchmark your overclocking simply indicates that you're letting it exceed whatever limit should be doing the throttling.

... and Good Luck.

Well, all I know is without folding@home running the gpu throttles if it detects that the gpu can't keep the core temp below 97 just by forcing fans to 100% and throttles if the VRM temp hits 110, but with it running it never throttles with either

My video card does, however, turn itself off if core temp hits 101 or VRM hits 135

And I know if I set my card below factory clocks it'll actually set to that clock speed with a game or stress test, but if folding@home is running it'll only go down to 950mhz core 1425mhz memory, but I can overclock, just can't underclock

whatever you're using to benchmark your overclocking simply indicates that you're letting it exceed whatever limit should be doing the throttling.

What? I have power limit set to +20%.. I even tried +0%... didn't change a thing

It's using 400 watts... that's actually kind of dangerous.. it only has 2 8-pin PCIe connectors and with that combined with the 75 watts it can pull from the PCIe bus, would be 375 watts (the power limit would be set way lower than that)

Of course you can go above that amount and maybe be fine, but it's risking it


My PSU actually says that PCIe#1 (my video card) peaks at about 40 amps (480 watts)

But the core and the memory amps reported by the video card only says about 35 (about 420 watts)

With folding@home and stress testing at the same time, it peaked a little over 30 (about 350 watts)

The higher the temps, the more amps... it starts out at about 28 amps, then goes up to about 35 by the time the core hits 97

The temp starts at about 35ish... but one second after the stress test starts the temp hits about 50...

Every 1°C is about 2 more watts

At least, that's what my video card and power supply report



Found a thread talking about increase in temperatures = more power consumption

http://forums.anandtech.com/showpost.ph ... ostcount=2

It's about processors but I'm sure video cards are the same

[Bryman]

WTF... my video card fans keep going down all the way...

My video card temp hit 103 twice and shut my computer off.. fans never turned up.. they reported being at 100% but I can't hear them

Wtf...

https://www.youtube.com/watch?v=V0n3guwbmcU

Yeah... they decided to run AFTER the video card shut off... *sigh*



It was doing this yesterday too... but it wasn't severe enough to cause my video card to shut off by hitting 105°C...

OMG it's doing it again... my video card temp is 90 when normally it's 82

And it's still going up.. video card temp is 100 right now

[Bryman]

Ugh.. my video card fans keep going all the way down... it keeps hitting 105 and shutting the video card off...

The first 2 safety measures (forcing fans to 100% and throttling) are not working... it is setting the fan speed to 100%.. but the fans are at a very low rpm for some reason... and my video card doesn't throttle when folding@home is running

At least the 3rd and last safety measure (shutting off the video card) is working... but it's becoming extremely annoying



I don't know what broke but my fans keep going to 0%...

[Bryman]

Also.. the GPU usage starts doing this.. almost like it's trying to throttle but can't



That's always how it's been when the temp goes high... but normally the forcing fans to 100% works... it's still doing it, but the fans aren't responding... they are staying at 0%

[Bryman]

So... it happened again

I set fan speed to 100% and after a few seconds the fans became inaudible... I felt underneath the card and felt a small amount of air being pulled in... the fan was just spinning really slow... I slowed it down even more with my finger and instantly the fan speed went back up to 100%

[bruce]

Is it possible that it's a mechanical problem? ... like maybe the fans need to be replaced?

[Bryman]

Probably... the video card fan has always needed to be above 80% to have the second fan run

I tried using my finger to slow/stop the fans again and it didn't work...

https://www.youtube.com/watch?v=MQDtgIHuoFI

It just randomly starts up when it wants to

It never goes above the fan speed I have it set at, but it'll go down to 0% fan speed (which is technically just a really really low rpm)



I'm going to switch to my 7850.. I can't fold with this video card.. it keeps hitting 105 and shutting itself off

[Bryman]

So... I was able to get the fans started back up by stopping the fan... but it has to be stopped quick and for a split second for it to work

https://www.youtube.com/watch?v=8iqs0Eq2fog


So I also can't underclock my 7850 when running folding@home.. but it lets me overclock

My video card does, however, throttle when it reaches the power limit... unlike my 7970



Not sure if it throttles when it reaches really high temps with folding@home running... guess I could test that real quick



Okay... I forced the fans stopped with my hand... it never throttled and the temp hit 106 and the video card shut off...

It's interesting though... the 7850 actually shuts off the computer while the 7970 just shut off the video card but kept the computer on



Hmm... without folding@home the video card throttled when temp hit 102...



So those weren't just weird glitches from my 7970... it happened with my non glitchy 7850

[mmonnin]

You definitely need to improve your card and/or case cooling

I don't know why this thread is more than 1 page long. Start and end right there ^^

[Bryman]

You definitely need to improve your card and/or case cooling

I don't know why this thread is more than 1 page long. Start and end right there ^^

*sigh* I don't feel like buying new fans for a 3+ year old card

I guess I could go liquid cooling...

[bruce]

*sigh* I don't feel like buying new fans for a 3+ year old card

I guess I could go liquid cooling...

Contacting the manufacturer for a RMA might be easier....
(Unless it's time to upgrade your cards.)

[toTOW]

Here is a little story : a friend has a GTX 970 from Zotac with a dying fan, that didn't get accepted for RMA (maybe the guy who refused it was deaf, because he said that it wasn't broken enough) ...

Here's the card :


So we removed the plastic fairing and the broken fan :


We had a Noctua fan laying around from a spare CPU cooler :


There, we fixed it !

No cost (spare parts) ... and if the manufacturer finally accept the RMA, we'll put everything back together and send it back

[Bryman]

*sigh* I don't feel like buying new fans for a 3+ year old card

I guess I could go liquid cooling...

Contacting the manufacturer for a RMA might be easier....
(Unless it's time to upgrade your cards.)

Yeah I would.. But by the time I figured out one fan wasn't working properly, the RMA window had expired by about a month

It only had 1 year of warranty I think

Yeah, I'm just waiting for the new ones to come out

The 7850 is actually pretty good... I have it overclocked as much as I can and it seems to be pretty decent



I guess I could try putting a case fan on it or something...



They might of accepted the RMA since it was only a month past and since it was just the fan... But it's definitely too late now... I've had the card for over 3 years



It's both the fans that aren't working but if I put the fan on the left side of the GPU then that'll be enough... That's where most of the heat is generated I think