Is this reason to rma 1080 Ti FE ?

Hi. I have weird little problem with my Gtx 1080 Ti FE. Mostly of the time , on load , clocks are 1550-1700mhz ,fan speed is on auto 49-50% ,temps of course are high 85C and GPUZ reporting THRM ( purple field ).



But.........sometimes when i run game or benchmark fan speed is on auto too but it is raising itself to 60% and temps of course are smaller and clocks are higher like 1800mhz and gpuz not reporting thermal limit



. I dont have installed any oc programs,like afterburner or precision. Only fraps,gpuz,cpuz. I will be very greatful for help here.

Please help.



Somebody said this to me:"Try the card in a different computer to see if the problem follows the card. If it does then it is time to start a RMA."

So i am worried now.



I am on 378.92 drivers.
Ah and ambient temp on the room is always the same. I am using card on the same case.

 

It's how GPU Boost 3.0 works. It increases clockspeed until it hits the thermal limit (85-86*C) and then throttles down until it stabilizes.

The whole reason is that the Founders Edition cooler is kind of crappy.

 

Fan just sometimes ramps up to 60% even on stock fan curve. And then GPUZ showing PWR limit ( green field ).
I am on stock fan curve. No programs installed. Stock all.

But mostly often is 49% and then gpuz tells THRM limit. It is doing itself without my control randomly. Any ideas why?

 

You just need a 3rd party cooler. The stock one with the FEs is pretty crap.

 

One time i run UNIGINE HEAVEN and fan speed is on 60% which resulting higher clocks and smaller temps.
Second time i run UNIGINE HEAVEN and fan speed is 49% which resulting on lower clocks .
Can somebody explain me this?


It is without my control. I dont have installed any programs like Precision etc.

 

Every GPU has limits set by the OEM via the vBIOS - those include thermals, power and voltage.

What happens is - your GPU starts boosting and then, if it hits the thermal limit (86 degrees centigrade), it starts downclock (thermal throttling) because of the Nvidia vBIOS setting.
If it hits the maximum power draw allowed in the vBIOS - it will start downclocking (power/TDP throttling) so as to conform with the vBIOS limits.

In short - this is normal behaviour. The thermal throttling is due to the FE cooler being crappy and the power limit throttling is there because otherwise GPU Boost 3.0 may end up overloading the card. Everything can be changed and tweaked via MSI Afterburner or EVGA Precision X.

EDIT: You cannot RMA your GPU as this is not considered a defect. It is widely known that the 1080 Founders Edition runs hot and throttles.

 

It hits thermal limit because fan speed is on 49% on auto. But sometimes fan speed goes itself up to 60% and there is not downclocking like was on 49%. So this i cant understand. Fan sometimes is on 60% without my control.
Testing on the same UNIGINE HEAVEN.

Ok. One time i run Unigine Heaven and GPUZ reporting all time POWER LIMIT ( Green Field ) with fan speed 60% auto.

Next day i run the same loop of Unigine Heaven and gpuz reporting after few minutes THERMAL LIMIT ( purple field ) but with fan speed 49% auto. This!

 

This is the last time I am going to repeat myself. TDP/Power throttling occurs due to the same reason thermal throttling occurs - you are hitting the maximum safe power draw specified in the vBIOS.

 

Last quesion. But why one day i run Heaven and all time in benchmark i have fan on 60%.
Second day i run the same loop in Heaven and have 49% fan. ?


I dont change anything in any programs. I dont have installed Precision or Afterburner.

 

Because you bought the Founders Edition. Next time, do research before buying a product.

 

Ok thanks for help

 

Ok i just catched a screenshot,so last time Sorry if i bothering.

ook i made today a test fire strike and look on fan speed , it shouldnt be 59-60% on load on stock: It was 59% fan speed instead of 49% like always ,look:



59% today. Yesterday was 49%.All is stock,default.

Somebody said that maybe bad fan controller?

 

For example. 1 June 2017 year - Running Unigine Heaven. Waiting 10 minutes and looking for fan speed. Its 49% and will go not higher till the end of benchmark.

3 June 2017 year ( this is example ). Running Unigine Heaven . Waiting 10 minutes and looking for fan speed. Now its 59% and will go not higher till the end of benchmark.


THIS i cant understand.

 

Just stop monitoring information that you clearly don't understand at all. Your card is running NORMALLY. That's it.

 

Yeah but i read in reviews that default fan stock curve is 49%. So why i have sometimes 59% on load ? I have stock curve.

 

There is NO default fan speed percentage. The cards vBIOS and temperatures will dictate what speed that fan runs at of left at default. The reason that can change daily and seasonally is because of ambient air temps, humidity, and it can also vary system to system because of overall case cooling and fan setups.

Again there is NO stock fan speed percentage, especially under load. Repeat this until it sinks in.

Like others have said the FE cards are pretty much crap air coolers and most buy them because they couldn't wait for the aftermarket cards or they want to buy a water block to cool them.

I suggest you run a manual fan profile and setup something like 70-80% fan speed while benching or gaming to keep the temps and therefore the clock speeds consistent and high. Go enjoy your card it does not at all sound defective.

 

Sorry if i am bothering. But i think i find similiar issue but with Gtx 1080 FE. Here is link,please read:

https://forums.anandtech.com/threads/i-think-my-gtx-1080s-temperature-target-malfunctioned.2494808/

 

Everyone already told you - you are looking for problems where there are none. If you continue down this path, don't expect anyone to waste their time trying to solve issues that don't exist in the first place.

 

Just read please

https://forums.anandtech.com/threads/i-think-my-gtx-1080s-temperature-target-malfunctioned.2494808/

 

No, I'm not going to waste more time trying to explain to you why you card is working normally.

 

Ok i figured out. SO listen. Its something like my card sometimes forget about thermal limit. And then clocks are higher like usualy, temps are higher,and fan speed raising to 60%. Its totally random.

Link i posted is about this. !

 

That is virtually impossible. A vBIOS cannot "forget" to trigger. Anyone telling you so is lying.

 

Dear don svetio,last question really. So why GPUZ sometimes not reporting THERMAL LIMIT ( purple ) field even if temperature goes to 86C? Its above temperature target.

 

Because at that point the GPU is thermal throttling in an attempt to lower the temperature of the core.

 

I ask in evga forum, and they said that i must rma card Here last post:

https://forums.evga.com/GTX-1080-Ti...seen-anything-like-that-m2689569.aspx#2690184

 

If you want to waste your time and spend 1 month without your card, feel free to RMA it. They won't swap it or fix it because it is NOT broken.

 

The guy is a troll or he regrets buying the FE card and he is looking for any reason to RMA the card. Luckily since there are plenty of FE cards to go around he will get another FE card, and with his attitude I hope he gets one that performs worse.

 

My card just sometimes ignoring temp target on stock. Its usually sits on 83-84C and thermal throttling. But sometimes when i run game temps go to 86C and there is no throttling thermal and clocks are higher. Any ideas? Drivers,GPU,bios? I am on stock settings. Even no installed Precision,Afterburner,nothing. Only gpuz,fraps,cpuz,real temp.

 

FE cooling is just barely "good enough". Either sell your card, take the loss and buy one with a better cooler, or get a water cooling kit.

 

He said IF nothing helps. Not that you MUST.

 

They've made cooling kits a lot easier to install for GPUs than last time I did it. It's definitely worth a try.

 

I've got the FE card and it actually cools better than I expected. Clocks usually settle in around mid to high 1700s with the fan around 50%.

@sew333 you're getting to hung up on values changing. That's exactly what GPU Boost 3.0 does.

 

Normally, card does fine and temp throttles when it reaches 83C under load. But card on rare occasion seems to disable its own temp throttle and just run itself full power ( without clock throttling ) until the temps reach in excess of 86C and the fans come to 59%. I've used GPUZ to try to monitor temps and confirm the readings so I know they're accurate. I I have noticed that rebooting the machine does seem to clear up the issue so I suspect this may be an obscure driver bug.

I have closely examined case cooling and found no issues there. I know the card does manage to run just fine most times with a constant temp around 83-84C when under load.

Has anyone else run into an issue like this? Any ideas?

Also found some few peoples with the same issue with FOUNDER EDITION cards.

 

because FOUNDERS EDITION cards are barely able to cool themselves - it has been this way since day 1 - FE cards are just hot

 

yeah but why sometimes i have 84C and clocks throttling which is normal. And why sometimes it forgetting to thermal throttle on 85C? And when i reboot pc it resolving this. So why rebooting helps?Aftter reboot again 83-84C and throttling at those temps.

 

If you've got it in a smaller case it may actually work better than a top venting one, with little cases the ability to vent out the back becomes an asset more than a liability.

 

That might be the the case. It's in a Ncase M1. I'm quite chuffed how this little build came together. It's my first SFF build.

 

Nice! What did you put into it?

 

So restarting pc fixing temp target. After reboot again temps 83-84C and throttling. Any ideas why this happening?

 

Is this after every reboot? How long until it starts ignoring the temp max?

 

Card on rare occasion seems to disable its own temp throttle and just run itself full power ( without clock throttling ) on 85C with fans 59%. I've used GPUZ to try to monitor temps and confirm the readings so I know they're accurate.

I I have noticed that rebooting the machine does seem to clear up the issue so I suspect this may be an obscure driver bug.

I have closely examined case cooling and found no issues there. I know the card does manage to run just fine most times with a constant temp around 83-84C with normal clock throttle when under load.

Has anyone else run into an issue like this? Any ideas?

Also found some few peoples with the same issue with FOUNDER EDITION cards.

 

Fans never go over 59%? are you using something to manage the fan curve?

 

nothing 0 programs like that. No have installed utilities oc.

I run now 3dmark ,temps go to 87C and no clock throttling, fan was on auto 63% lol. What is going on?

 

Try installing Precision X or something like that, you should be able to set a point where it throttles even if the driver/firmware isn't doing it automatically.

 

It's in a NCase M1 v5

 

That would definitely benefit from an FE or other rear-venting card. Very nice looking case too.

 

Default temperature target: 84C.F



Sometimes temp limit stops working on my card,fan goes from 49% to 60% on auto, and temps goes as high 87C and there is not throttling. Any ideas? Card is 1080 TI FE stock,no oc. No programs installed. Rebooting pc cure this.

I tried on 378.92 and 382 drivers

Card sometimes goes above temp limit which is 84C. Go to 87C and when i looking to GPUZ its showing PWR LIMIT ,not THRM LIMIT. Can somebody explain me this? PLEASE . Is this normal or not? I promise i will not ask again and not make next topic.

 

We've told you several times already - it's normal. The Power limit kicks in when the GPU tries to draw more power than the board can handle. These are all safety protections that keep your card operating in the correct conditions. Stop worrying about it and just remove the monitoring software - you don't need it

 

Sorry and my last question , Somebody said :" To me it sounds like a disconnect between the driver and the hardware so the hardware is left to fend for itself. A complete removal and reinstall of the drivers may help. If it has always done this then it could be a bad card."

It can be bad card like he said or nah ? Thanks for reply.

 

That is virutually impossible. Please stop listening to 12-year-olds who have no bloody idea how hardware works. If your driver crashes, you will see a screen flicker and your application will likely crash as well. In some cases of driver crashing, you can also see a BSOD. So no, a driver "disconnecting" is simply not possible. It has never been possible and never will be possible.