GPU getting to hot Nvidia Gtx 690

[e6r6i6c]

So I have a GTX690 running it in Single GPU mode in the Nvidia controll pannel, and Im getting a decent framerate, but just now I found out using the EVGA precision tool that my GPU is getting very hot, and I really mean very hot, it says 98 degrees celcius. I have a big case, a Coolermaster Cosmos 2, and I have all the fan slots filled with some high rpm fans, so my airflow should not be a problem, I also just cleaned all the dust out. So anyone else having the same problem and same temperatures? The Gpu has a usage of 99% all the time, im hitting like 40 fps going down sometimes in the low 20's. I am not getting any artifacts or anything that looks like the GPU is failing.

[Salty_McSweet]

Originally posted by oXYnary:

Since your going to be a know it all and say everyone else is clueless, you might want to look in the mirror a bit more. 95+ under load is not normal. Especially if you admit its thermaling itself to not reach 100.

BAM!

http://www.anandtech.com/show/5805/nvidia-geforce-gtx-690-review-ultra-expensive-ultra-rare-ultra-fast/16

Thanks for giving the example of just what I was talking about. You have no clue yet you just keep talking. The link points to graphs that show the load temperatures they experienced under the specific loads they tested with. Show me that same graph with all of those cards running Ark and most likely you'd see many of the cards running hotter and the gtx690 pegged at 98c.

Do you even know what a thermal threshold is? I know those are big words and hard for you to process, but it's the point at which the card essentially is still safe, but will try not to exceed. That's when it will kick its fans to highest and start to clock itself down slightly until the temp drops. If nothing's broken it won't need to clock down much.

Check this link and look at the thermal threshold for the gtx690. GTX 690 Review[www.guru3d.com] See how it's 98c? Isn't that amazing! That perfectly matches the temp the OP stated. He said he clocked it down a bit and it started to run at 95 instead. That just means he's lost a few frames for no reason.

Take a look at this link, too, just as an example of what's "normal" under load. It's for newer cards so it doesn't include the gtx690. See the AMD R9 290? See how under load it normally runs at around 94c? GTX 690 Review[www.guru3d.com]

These temperatures are simply not "outrageous" as some clueless individuals will say. The chips are rated at over 100c, but they don't want to operate at their max temps all the time. That's why thermal thresholds exist. To keep them throttled and operating within safe limits.

[Sistermatic™]

You do know that thermal threshold is not the same as normal operating temperature right?

It's a safety mechanism, not an operational level.

It is the point where the card protects itself like a rev limiter on a car. Just as a car tacho has an area, painted red, to make it clear it is not safe to maintain that rev range over a sustained period of time. The rev limiter applies AFTER the red area has been exceeded. The limiter simply prevents terminal damage, it is NOT the level you should be operating at.

[oXYnary]

Originally posted by Alarmist:

Originally posted by oXYnary:

Since your going to be a know it all and say everyone else is clueless, you might want to look in the mirror a bit more. 95+ under load is not normal. Especially if you admit its thermaling itself to not reach 100.

BAM!

http://www.anandtech.com/show/5805/nvidia-geforce-gtx-690-review-ultra-expensive-ultra-rare-ultra-fast/16

Thanks for giving the example of just what I was talking about. You have no clue yet you just keep talking. The link points to graphs that show the load temperatures they experienced under the specific loads they tested with. Show me that same graph with all of those cards running Ark and most likely you'd see many of the cards running hotter and the gtx690 pegged at 98c.

Do you even know what a thermal threshold is? I know those are big words and hard for you to process, but it's the point at which the card essentially is still safe, but will try not to exceed. That's when it will kick its fans to highest and start to clock itself down slightly until the temp drops. If nothing's broken it won't need to clock down much.

Check this link and look at the thermal threshold for the gtx690. GTX 690 Review[www.guru3d.com] See how it's 98c? Isn't that amazing! That perfectly matches the temp the OP stated. He said he clocked it down a bit and it started to run at 95 instead. That just means he's lost a few frames for no reason.

Take a look at this link, too, just as an example of what's "normal" under load. It's for newer cards so it doesn't include the gtx690. See the AMD R9 290? See how under load it normally runs at around 94c? GTX 690 Review[www.guru3d.com]

These temperatures are simply not "outrageous" as some clueless individuals will say. The chips are rated at over 100c, but they don't want to operate at their max temps all the time. That's why thermal thresholds exist. To keep them throttled and operating within safe limits.

Hey man, you started the flame war, I'm just showing you don't know what you are talking about since that's what you claimed disrespectfully sabout us. Like Sistermatic says, that's the upper limit of the card. NOT what it should run at underload. You very much left out that your card which you say is normal for 90+ ran under 85° under load in the review I linked and 75° under your link.

So you're still showing ignorance. The best thing to do is admit it and apologize, or continue flaming/trolling to show your pettiness.

[Strider]

Originally posted by Alarmist:

Originally posted by oXYnary:

Since your going to be a know it all and say everyone else is clueless, you might want to look in the mirror a bit more. 95+ under load is not normal. Especially if you admit its thermaling itself to not reach 100.

BAM!

http://www.anandtech.com/show/5805/nvidia-geforce-gtx-690-review-ultra-expensive-ultra-rare-ultra-fast/16

Thanks for giving the example of just what I was talking about. You have no clue yet you just keep talking. The link points to graphs that show the load temperatures they experienced under the specific loads they tested with. Show me that same graph with all of those cards running Ark and most likely you'd see many of the cards running hotter and the gtx690 pegged at 98c.

Do you even know what a thermal threshold is? I know those are big words and hard for you to process, but it's the point at which the card essentially is still safe, but will try not to exceed. That's when it will kick its fans to highest and start to clock itself down slightly until the temp drops. If nothing's broken it won't need to clock down much.

Check this link and look at the thermal threshold for the gtx690. GTX 690 Review[www.guru3d.com] See how it's 98c? Isn't that amazing! That perfectly matches the temp the OP stated. He said he clocked it down a bit and it started to run at 95 instead. That just means he's lost a few frames for no reason.

Take a look at this link, too, just as an example of what's "normal" under load. It's for newer cards so it doesn't include the gtx690. See the AMD R9 290? See how under load it normally runs at around 94c? GTX 690 Review[www.guru3d.com]

These temperatures are simply not "outrageous" as some clueless individuals will say. The chips are rated at over 100c, but they don't want to operate at their max temps all the time. That's why thermal thresholds exist. To keep them throttled and operating within safe limits.

Do you actually have any real world experience with working with such hardware? I mean at its actual component level? Yes it can get to 98c but that is its absolute top temp and a dangerous one to run at... Imagine 3.5 billion transistors running full whack at their limit in temperature with thermal expansion... Yeah they are gonna fry - Internal closed short. Anyone that knows anything about GPU's know that the GK104 series of GPU's run hot and can easily google that top temp.... It doesn't mean you actually know anything other than how to use google... If you want him to fuse his transistors or destroy his BGA connections on his GK104-355-A2 chips then by all means keep telling him it's fine... my suggestion is to not bother trying to be all so tech savvy when clearly you have no idea about any of it other than what you may possibly read!

I work with GPU's day in and Day out... I Reflow, Reball and change ANY component on such hardware... I can trace out shorts (Open and closed) and reverse engineer PCB's... I don't have my own business in doing this for not knowing what is what!

Wanna know something funny... I have a card sitting here that was running in the mid nineties and it doesn't display.... Why doesn't it display?... the BGA connections have shot because of thermal issues.... Guess what the said card is?... You guessed it... A GTX 690... Even has the ARCTIC Accelero Twin Turbo cooler.... Still due to incorrectly applied TIM by its owner it died from such heat! Now it is costing it's owner £30 for each GPU to be fully reballed.

[Degenere]

I was having this same problem and I found a thread at EVGA where the card was new and when they RMAs it the problem went away. So I kept doing research and turns out this happened to other GTX 690 users at different stages and different non-SLI games. Turns out stock heatsink gets damaged and stops doing its work even after replacing thermal paste (some people said they tired and didn't help). So I went with replacing the heatsink with this one: http://www.amazon.com/ARCTIC-Accelero-Twin-Turbo-690/dp/B009PK1TL0 which was what some other guy did. Now from 98c up to 101c once (never crashed). I got 67c tops and 29/31 idle.

[YenLoWang]

my 690 is sitting at 94 degC its 45 deg ambient, thats on core 2, core 1 is at SFA cos ark doesnt even use both cores on it. fans at 90% drivers are current. stats off asus gpu tweak, card has been manually oc slightly as well

[GG.Madnetics]

The NECRO is REAL!!!!!