Is this reason to rma 1080 Ti FE ?

Hello again Somebody said that this can be caused by faulty gpu fan controller or faulty card. I post this words:"My card doesn't do this, but by using Afterburner you could determine if there is a problem with the fan controler, and if you do not want to use Afterburner or some other tool to get around this issue, you could still send it into RMA afterwards. If you have just bought it, i would consider sending it back to the seller.:

 

If the problem was the fan controller, your GPU fan would either stop spinning or not go past 20% fan speed (which IIRC is the idle setting). The fan controller is not faulty. Again, stop listening to people who have no bloody idea what they are talking about. They are either ignorant or lying to you.

 

Ok i promise it will be my last question. I am worried a little because somebody said this to me:"Hey,

I cannot think of any other solution other than sending the card back to the retailer. All cards should throttle at various points, be that limitation to power, temps, voltage etc. At first my guess was that the card isn't actually physically in-tact, which would be the root cause of thermals suddenly spiking up (misalignment and poor contact of the cooling block), but even then, the card would be perfectly able to throttle itself back to safe limits.

The card doesn't seem to work even in stock configuration - that's a good a reason to contact the retailer and ask for a replacement.

Hope this helped :kahvi:"




So what to do now? Listen to him?

 

No because he has no idea how a GPU works.

If the GPU is not intact - the system WILL NOT BOOT.
If the GPU is not making proper contact, the system will overheat and SHUT DOWN

The reason why the card behaves like that at stock is because Nvidia's cooler design is not good enough to keep boost clocks high. Remember, you are still above base clock. You are ONLY guaranteed that the card will work at base clock. Nobody makes any promises as to what boost clock it can or will hold.

In short, sending the card would be a waste of money and time because there is nothing physically wrong with it.

 

Thanks DON i am listening to you

 

Ok last question. Somebody said this to me:"Huh, why can't you send the card back if you've just bought it a few months back?

Could you link any forum post/website/etc. where people have discussed this issue?

Admittedly this is the first time I've ever head a card ignoring it's predefined temp limit. It also doesn't make any sense to me why things suddenly reset back to normal when you reboot your computer neither it is commonplace for a graphics card to arbitrarily exceed its temp limit (which is predefined in the BIOS of the graphics card, not in the driver).

If this was something that you've done to the card yourself - modifying the BIOS settings, physically tampering with the card or somehow forcing higher temp limit through software, I would understand the confusing behavior, but like you said, you run the card at stock configuration.

It honestly is beyond me as to why this happens, that's why I suggested RMA'ing the card in the first place."


Anyway what you think about that words don? What answer to him ?

 

I do think I understand what you've been saying in this thread - you card is sometimes ignoring the temp target. That's weird. You've got a choice, you can either RMA the card if they allow you to, or you can keep the card. If you keep the card you can either create a custom fan profile to control the fan speed vs temperature (in something like MSI Afterburner, or you run manual fans at a constant 70% or whatever level you're happy with. It's your choice.

 

As i said, after reboot the card maxed out at 83°C under same conditions. WoW at 200% renderscale, 3DMark and Overwatch.

I checked my GPU-Z logs. The moment the card was at 87°C there was no temp limit warning, and the card was almost at full GPU boost speed, that's the thing that bothered me.

Same tests after reboot, the card was at 83°C and boost clock was lower and temp limit was reached.

That's why i am asking here what can cause that the card is ignroring the given thermal limit of 83°C.

Now my question is, what in the world was that ? Boost clock is controlled by driver, isn't it ? Maybe the driver went nuts.

i dont think it was an airflow issue. The card never reached 87°C even under maximum load at 4k resolution. It was the same Setup, i didn't change everything.

As i said, after reboot the card maxed out at 83°C under same conditions.

report delete reply

 

Who told you that 83*C is a thermal limit!?!?

The official Nvidia thermal point for the boost clock is 86*C - 83*C has NEVER caused throttling on desktop cards.

 

84C default temp target

 

Temps sometimes go past 84C to 86-87C and i dont have THrm limit drop in clocks in gpuz. Gpuz stating Pwr limit or voltage limit.

Temperature suddenly goes past 84°C to 85-86-87C and the fanspeed goes up.

Gpuz shows Pwr Limit or VREL,instead of thrm limit. This i cant understand why it doing this sometimes?


Default temp target is 84C. So on 84C gpuz should post thrm limit.

After pc reboot , again normal, 83-84C max and gpuz showing thrm limit on 84C like it should.

 

Yeah, doesn't sound like it should be doing the things you're describing, so something is 'off' with your GPU or system or something. If you're not happy with the card then RMA it, otherwise create a custom fan curve that prevents the GPU getting to the high temperatures that you obviously don't like to see.

 

It's normal for the card to still go beyond 84C even though that it's the thermal limit even if only temporarily peaking. It also sounds like the card is working perfectly and spinning the fans up higher which is again normal and needed.

From Anandtech review of the founders 1080 Ti

"The story is much the same under FurMark. The GTX 1080 Ti settles at 84C here as well – though it did peak at 86C before reaching equilibrium – showcasing that regardless of the workload, the card always levels out at its thermal throttling point."

Stop relying on GPUZ for your investigative work. It could be a bug or the program itself. It reporting correctly. The card is obviously fully functional and is correctly spinning the fans up when it hits 84C.

Stop posting the same thing over and over again or this needs to be reported to the mods for spamming. You're literally arguing with our members here and posting non sense or cross posting out of context information from other web boards. It needs to stop. Either take our advice or RMA your card and move on.

 

"One last question, I promise!"

 

This is consistent with our internal testing as well.

 

Didnt go through every page... but for those looking for aftermarket cooler solutions for GTX 1080 Ti FE, I bought the Accelero Hybrid III 120mm cooler (980 Ti Heatsink) and it works flawlessly.

Waiting for EVGA on theirs takes forever...

 

That's meant to work a treat on the TI. I was prepared to get one if my temps were bad.