[SOLVED] RTX 2080ti performance far below average.

[BlazingSword]

Hello people,
I've been running a ROG-STRIX-RTX2080TI-O11G-GAMING for about a year and a half and recently I noticed that it's performance is quite far from what it's supposed to be. I've been running some benchmarks and comparing FPS with friends and some videos online, and while I do understand that these kinds of stats can be deceiving and far from perfect, the results tend to show about 30% performance loss most of the time, not a single result seems within tweaking margin, as in OC, Drivers and such.

Could you guys help me properly test my GPU and give advice on how to fix this?

Some examples of my testing include:

Unigine Heaven: Score 3754, Max FPS 347.5
Tested using standarized settings used in a thread in this website with 7 different people showing their results, people scored an average of ~5800 and ~470 Max FPS, with a minimum of ~5400 and 422 Max FPS.

UserBenchmark: GPU Scored 134%, while average is 168%

Black Desert Online: ~60FPS @1440p Maxed settings, while a friend with a nearly identical setup gets >90FPS
(only difference is an i9 9900k vs my 9700k)


My system specs:
Aorus Z390 Ultra Mobo
i7 9700k clocked @4.8GHz, 1.28v Max temp 67c
RTX 2080ti Asus Strix, stock settings Max temp 73c
Corsair Vengeance RGB PRO 3000MHz 2x8GB
Seagate Seasonic Focus Plus 850W Gold PSU

Please tell me whatever info you may need that i didn't add. TY in advance.

 

[Rogue Leader]

I have read through the thread and I agree with Darkbreeze, IMO there is possibly a temp issue with the card that you're not seeing due to a bad temp sensor or other problem. Its interesting because your clocks are higher than what that card is rated at (1350/1665), yet performance is not there. Could be memory temperature? Strange its also only showing 2 fan RPM, the card has 3 fans. Those 2 fans are screaming too at 3500rpms. Are all 3 fans running? I'm wondering if one of the fans is not running or not spinning up because of a sensor issue or there is some other cooling issue only affecting part of the card thereby overheating the memory or VRMs.

That would make sense based on Darkbreezes post above and the indicators pointed out. The card is throttling the memory or VRMs under normal conditions. The only other thing that can cause that other than a cooling problem is that they were defective to begin with.

 

[BlazingSword]

I have read through the thread and I agree with Darkbreeze, IMO there is possibly a temp issue with the card that you're not seeing due to a bad temp sensor or other problem. Its interesting because your clocks are higher than what that card is rated at (1350/1665), yet performance is not there. Could be memory temperature? Strange its also only showing 2 fan RPM, the card has 3 fans. Those 2 fans are screaming too at 3500rpms. Are all 3 fans running? I'm wondering if one of the fans is not running or not spinning up because of a sensor issue or there is some other cooling issue only affecting part of the card thereby overheating the memory or VRMs.

That would make sense based on Darkbreezes post above and the indicators pointed out. The card is throttling the memory or VRMs under normal conditions. The only other thing that can cause that other than a cooling problem is that they were defective to begin with.

All 3 fans are running and spinning up, i guess they're handled in a 2+1 configuration.
I know my gpu is advertised to run at 1665mhz on boost, but why is it that every video i see has the same GPU running at over 1900mhz? Something is causing my card to reach POWER limit and it seems to throttle a bit, since at the start of all tests and games it does run at like 1950mhz with good FPS, just to drop down over a few seconds. I had a suspicion that this issue was temp throttling because some of the first benchmarks to give higher scores were ran coincidentally when my case was open, however this didn't turn out to be the case since i tried running the benchmarks with my case closed and my room AC off, and it still scored much higher as long as i decreased the power limit.
It also may be worth mentioning that apparently Superposition benchmark REALLY values framerate stability since that's exactly what lowering power limit seems to achieve, a stable, low core clock speed, and that is reflected on the FPS counter: Without limiting power, the numbers go all over the place to the point its impossible to read, when limited however they remain within a ~3 fps margin, this would explain why reducing power limit has a positive impact on my scores. This is however not a solution for games, since everything seems to crash right away under these settings. The ideal would be to find a way to make my clock speeds remain stable since they are always going up and down like crazy during gameplay.
It almost feels like my card itself has a pretty low power limit, regardless of what is configured in Afterburner.
Also my card has a Micron memory, which i read soon after launch that almost all of the cards produced using this memory were bad somehow.

 

[Rogue Leader]

Yes they did switch memory vendors due to quality issues. Also the behavior you are describing where lowering the power limit says to me the VRMs may be overheating at max power. Something you won't see in HW Info or other monitoring software, but exposes a cooling issue.

I think you need to warranty your card.

 

[Phaaze88]

I know my gpu is advertised to run at 1665mhz on boost, but why is it that every video i see has the same GPU running at over 1900mhz?

You need to ignore the advertised boost clock. That's a guaranteed number in the worst case scenario.
10, 16, 20 series all easily boost higher than advertised if most parameters are 'ok'.

It's hitting the power limit because it's hitting higher boost clocks. But if the power limit is lower due to the VRMs are overheating...
https://www.techpowerup.com/vgabios/207409/asus-rtx2080ti-11264-181122

https://www.techpowerup.com/vgabios/207794/asus-rtx2080ti-11264-181019-1
Those are vbios files for your specific gpu - one with a limit of 325w, another with 313w. In the hwinfo samples you've provided, the gpu is never seen hitting either, and it should be doing that with ease.
But if the VRMs are throttling, 313-325 isn't gonna happen.

By the way, the temperature readings you do see with that card are an average of 3: the gpu core, memory, and VRMs.
As for why Nvidia did it like that, I've no idea.

 

[BlazingSword]

Well, I guess there's nothing we can do about it then, sadly my reseller's warranty expired about 8 months ago, i will try to contact ASUS to see if there's anything they can do about this since i live in Mexico and although my card was sold and bought here through an official reseller, the manufacturer's warranty policies/customer service may not be the same.

Thank you everyone for your time and patience, especially Darkbreeze for following my thread so diligently. Have a good one.

 

[Darkbreeze]

So, just put a limit you can live with on your boost clocks using Afterburner (Or whatever utility you prefer), and that should completely fix the issue, or at least put a legit band aid on it. You won't get the full boost the card was intended to be capable of, but play around with the boost settings manually to see where the fine line is between performance and where it starts goofing up. You might even want to manually configure the fans so that they run a bit higher and a bit sooner than what they do on the default cooling profile to try and avoid the throttling behavior altogether.

Could even be worth tearing the card down and repasting the various heatsinks.

AND, it would be a very good idea to contact ASUS anyhow, especially on their forums, because there is OFTEN somebody who has had the same problem with the same card that knows a work around already PLUS many of their tech support people wander around on their product forum as well so you might actually catch tech support assistance there as well.

 

[Darkbreeze]

IDK man, looks like you're not the only person. MAYBE there is a driver glitch or bug going on with the latest drivers.

[SOLVED] - Gigabyte GeForce® RTX 2080 SUPER™ WINDFORCE stuck at 1650hz.

Gigabyte GeForce® RTX 2080 SUPER™ WINDFORCE. Aorus Engine is set to 1830hz, but my NZXT cam app and FurMark states that it stuck at 1650hz.

forums.tomshardware.com

 

[BlazingSword]

IDK man, looks like you're not the only person. MAYBE there is a driver glitch or bug going on with the latest drivers.

[SOLVED] - Gigabyte GeForce® RTX 2080 SUPER™ WINDFORCE stuck at 1650hz.

Gigabyte GeForce® RTX 2080 SUPER™ WINDFORCE. Aorus Engine is set to 1830hz, but my NZXT cam app and FurMark states that it stuck at 1650hz.

forums.tomshardware.com

Hmm that's interesting, I sure hope its a software issue. The throttling really is instantaneous and it kicks in as soon as theres any kind of load, even with the card power-limited and running at temps under 60c. It almost seems like an energy-saving feature or a problem with GPU Boost, but i digress.

I contacted ASUS and the support person also believed it was a software issue, even after i told him everything we tried. I might hold off on sending it for warranty though, thay said just the validation may take over 50 business days.

Also my card doesnt really get stuck at 1650, thats just where it seems to stay for longer at, but in reality the clocks sway drastically between ~1900 and ~1500.

Thanks for the heads up, I'll keep an eye out for more threads like this, perhaps we can find something out.

 

[Darkbreeze]

I'll keep my eye open as well.

 

[Rogue Leader]

IDK man, looks like you're not the only person. MAYBE there is a driver glitch or bug going on with the latest drivers.

[SOLVED] - Gigabyte GeForce® RTX 2080 SUPER™ WINDFORCE stuck at 1650hz.

Gigabyte GeForce® RTX 2080 SUPER™ WINDFORCE. Aorus Engine is set to 1830hz, but my NZXT cam app and FurMark states that it stuck at 1650hz.

forums.tomshardware.com

I have the same exact card as that post and haven't seen this issue at all, just updated to the driver released yesterday as well and no change (other than the card running a bit faster seemingly). I sit at 1930mhz all day long.

 

[Darkbreeze]

Curiouser and curiouser. LOL.

 

[Darkbreeze]

I don't recall, did you try testing it out with the side panel off to see if a lack of airflow was the problem?

 

[BlazingSword]

I don't recall, did you try testing it out with the side panel off to see if a lack of airflow was the problem?

Yeah i did, i tested with side panel off at ~19c room temperature and case closed with room AC off which brings room temperature to about 27c, no change.

I noticed yesterday that this "ghost power limit" is around ~72%, so my cards performance for some reason reaches its highest when i manually limit power right below this percent (manual limit 70%). I even got a ~35 fps increase in Black Desert at 1440p.

 

[Darkbreeze]

Well, it sucks that you have to do that, but it really does seem like it's software related. Did you try a clean install of Windows?

 

[BlazingSword]

Well, it sucks that you have to do that, but it really does seem like it's software related. Did you try a clean install of Windows?

Haven't tried that yet. I still have my old computer with the FX-9590 and a 970, gonna try swapping cards and see how they perform.

 

[BlazingSword]

Well, it sucks that you have to do that, but it really does seem like it's software related. Did you try a clean install of Windows?

Would you recommend flashing my vBIOS? I have a feeling this might help, but i don't know anything about this and it seems to be quite risky, also i believe it voids warranty, but i don't think RMA is an option.

Current BIOS:

 

[Darkbreeze]

Personally, I'm always in favor of updating MOTHERBOARD BIOS when there is a newer, stable version, but when it comes to graphics cards I'm not in favor of it AT ALL unless there is a CLEAR advantage to doing so AND a consensus has been reached with the card manufacturers engineers (Normally through tech support) that it is a problem likely resolved by a newer vBIOS.

That isn't to say that there aren't likely instances where just doing so, independent of what anybody else thinks or does, can't be a good idea, but I'd be inclined to have that discussion with the card manufacturers second or third level tech support people first, if they are talking to you about the problem WHICH they SHOULD be regardless of whether the card is still under warranty or not. If they are not, I'd be resilient in my attempts to get them to put me in touch with those people.

I do believe the ROG forum is a GOOD resource for getting in touch with the right people, as seen here in this thread with ex-employeed Bahz, who clearly works the ROG forum and has contacts with those people still. I'm not suggesting that THAT member is the one you need to contact, merely that the right people, or the people that can GET YOU to the right people, do frequent those forums.

https://rog.asus.com/forum/showthread.php?71345-Why-is-Asus-Tech-Support-so-bad

What I WOULDN'T do is take just any regular person's advice to flash your vBIOS or do anything else in the way of modifying things, until you've had a chance to at least speak with somebody with an above average understanding of the card's architecture and history.