• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Fluctuating frame rate and not sure what to do

Joined
Mar 15, 2019
Messages
168 (0.07/day)
Processor AMD Ryzen 9 3900X
Motherboard Gigabyte Aorus Master Rev 1.0
Cooling NZXT Kraken X72 360MM
Memory G.Skill Trident Z Neo 32GB (2x16GB) DDR4-3600MHz CL16
Video Card(s) Galax GeForce RTX 2080 Ti Hall of Fame
Storage Samsung 970 EVO Plus 1TB NVMe M.2
Display(s) Alienware AW3418DW 34" UltraWide QHD 120Hz IPS 1900R Curved G-Sync Gaming Monitor
Case Lian Li PC-O11 Dynamic XL Black Steel
Audio Device(s) OnBoard Audio - ESS SABRE HIFI 9118
Power Supply Corsair RM850i 850W
Mouse Logitech G502 LightSpeed Wireless Gaming Mouse
Keyboard Corsair K70 RGB MK.2 Cherry MX Brown Mechanical Gaming Keyboard
Software Windows 10 Professional
Hi All,

I'm new here and registered purely to get help from the community. I hope that's ok.

All my specs can be seen on the snapshots provided.

I have had my Nvidia GeForce 980TI for some time now. It's never been a proble for me at all.
I don't overclock my hardware as I am happy with the defaut performance. Sorry if that upsets a few.

Anyways, today I noticed while playing Jurassic Park Evolution that my frame rate was dropping. Yesterday the game was 100% fine. no problem.

Today, the games FPS will drop suddenly and then climb back up and then drop and climb repeatedly. I have taken some video footage showing the problem.

So I figured it was a driver issue and updated to the latest nvidia driver. No improvement whatsoever.

So I checked other games and lo and behold, it was happening in Surviving Mars as well which led me to believe that the card might be the problem.

So then after doing some research, I saw people advise others to change the Power Management Mode to Prefer Maximum Performance. So I did that. no difference.

So then I downloaded GPU-z and checked how the card was performing during gameplay and noticed that the clock speeds for the GPU and the MEMORY were fluctuating every time the frame rate would drop and climb.

Then I noticed the PerfCap reason was set to Pwr which indicates that the card is hitting its power cap but I also noticed that this is not fluctuating when the clock speeds are fluctuating. So that baffled me.
I then downloaded MSI AfterBurner and increased the Power Cap limit and went back in to the game and my entire PC froze and I had to reboot.

During these GPUz tests, I created logs and I am hoping someone can assist me in trying to figure this out. It is escaping me and I would be SUPER bummed if my card has gone for a trip down trash lane!

See attached snapshots and logs.

Thanks in advance.
 

Attachments

  • GPU-Z Sensor Log(before_performance_mode_change).txt
    GPU-Z Sensor Log(before_performance_mode_change).txt
    256.7 KB · Views: 292
  • GPU-Z Sensor Log(After Performance_mode_change).txt
    GPU-Z Sensor Log(After Performance_mode_change).txt
    554 KB · Views: 278
  • fluctuations.PNG
    fluctuations.PNG
    55.9 KB · Views: 400
  • max-perf01.PNG
    max-perf01.PNG
    60.2 KB · Views: 456
  • specs.PNG
    specs.PNG
    77.7 KB · Views: 432
  • cpu01.PNG
    cpu01.PNG
    74.8 KB · Views: 433
  • memory.PNG
    memory.PNG
    61 KB · Views: 392
  • motherboard.PNG
    motherboard.PNG
    59.8 KB · Views: 415
the card may well be on its way out.. its certainly not performing as it should..

trog
 
Thanks but I was hoping for a little bit more than just "your card is dying".

I realise that the card is not performing as it should and that's why I came to this forum for some help.

Our country is currently experiencing power cuts and we had a power cut last night. Could that have caused this?

If so, exactly would would have died on the card or PC if this was the cause??

Any help would be appreciated
 
Card dying or PSU malfunction because of power surge?
 
Replace thermal compound on the gpu die, and give your case and fans, to included the videocard heatsink and fan a good dust removal.

Replace thermal compound on cpu too, clean cpu heatsink and fan (s)
 
What's the temps on your CPU?
The CPU may be overheating... The PSU may also be in its way out.
 
I'll give everything a good dust service tomorrow but I have to be honest, I've never touched the thermal paste on the card so I'm not sure why I would need to replace that. The card has been fine for over 2 years now. Its never been overclocked so it's never run hotter than it should so I doubt the thermal paste needs replacing.

I'm more inclined to believe that maybe the PSU us on is way out so I will check tomorrow if I have another port on the PSU for a second card that has never been in use to see if it will change things.

Is that an acceptable plan?
 
I'll give everything a good dust service tomorrow but I have to be honest, I've never touched the thermal paste on the card so I'm not sure why I would need to replace that. The card has been fine for over 2 years now. Its never been overclocked so it's never run hotter than it should so I doubt the thermal paste needs replacing.

I'm more inclined to believe that maybe the PSU us on is way out so I will check tomorrow if I have another port on the PSU for a second card that has never been in use to see if it will change things.

Is that an acceptable plan?

You asked for help, do it as suggested.
 
I'll give everything a good dust service tomorrow but I have to be honest, I've never touched the thermal paste on the card so I'm not sure why I would need to replace that. The card has been fine for over 2 years now. Its never been overclocked so it's never run hotter than it should so I doubt the thermal paste needs replacing.

I'm more inclined to believe that maybe the PSU us on is way out so I will check tomorrow if I have another port on the PSU for a second card that has never been in use to see if it will change things.

Is that an acceptable plan?
If the other port is on the same PSU you will probably have the same issue.
You may not need to replace the thermal compound and or it may need to be done regardless of if you overclock or not as thermal compound is only good until it's not.
Do what you can to monitor temps, if you're getting temps that hover at or above 70°c you may have an issue with heat.
 
@eidairaman1 No need to be so snarky about it. I'm not 3 yrs old so don't treat me like a child my friend.

I don't have to agree with everything that you say. That is my prerogative. If you can give me a good enough reason as to why the first thing I should do is mutilate my card to replace the thermal paste then maybe I would be inclined to agree but your first suggestion right of the bat was to replace thermal compound on the gpu before giving the machine a good dusting or trying other PSU ports or motherboard slots.

Your first suggestion sounds extreme. I would think that the first thing to do is to do the easy stuff first and then the more risky stuff.

I can see you are an experienced board member here but that doesn't mean I have to tolerate those types of replies You should take a page out of jmcslob's book on how he is approaching the responses.

@jmcslob I will definitely check the CPU temps tomorrow and see what's happening. Thanks for the suggestion.
 
i took the trouble to look at your posted images.. i posted my thoughts.. maybe not what you wanted to hear but i aint trying to humor you just tell you what i think the problem likely is.. :)

more so now you mention a recent power outage..

trog
 
@trog100 I only thought of the power outage after I made my post.
 
@eidairaman1 No need to be so snarky about it. I'm not 3 yrs old so don't treat me like a child my friend.

I don't have to agree with everything that you say. That is my prerogative. If you can give me a good enough reason as to why the first thing I should do is mutilate my card to replace the thermal paste then maybe I would be inclined to agree but your first suggestion right of the bat was to replace thermal compound on the gpu before giving the machine a good dusting or trying other PSU ports or motherboard slots.

Your first suggestion sounds extreme. I would think that the first thing to do is to do the easy stuff first and then the more risky stuff.

I can see you are an experienced board member here but that doesn't mean I have to tolerate those types of replies You should take a page out of jmcslob's book on how he is approaching the responses.

@jmcslob I will definitely check the CPU temps tomorrow and see what's happening. Thanks for the suggestion.

TLDR. @Wolvyreen

Not snarky, I am to the point.
 
@eidairaman1 call it what you like but from my experience, people don't respond well to other people that are abrupt or "uppity". Google that ;)
 
@eidairaman1 call it what you like but from my experience, people don't respond well to other people that are abrupt or "uppity". Google that ;)
No need to, i gave you what needs to be done regardless, as i said you came here for help, I gave a suggestion and you are now doubting the suggestion. It is known as covering the bases, the card by manufacture is 4 years old, card makers use cheap thermal compound.

@Wolvyreen
 
do you have '"prefer maximum performance" enabled in the nvidia control panel?
 
@eidairaman1 To be honest, you've already lost my respect so I'll take the other posters suggestions above yours as at least they aren't asking me to mutilate my card as a first step and then when I hesitate to their suggestions, they don't get all "uppity" about it.

@phanbuey that was mentioned in the first post. In fact it was in bold. ;)
 
the sad thing is i even looked for it.

It sounds like a dying card or PSU.

Can you put your PC power settings to high performance... have you installed any new software or updates?

EDIT: ALSO- Geforce experience is known to do this sometimes - can you uninstall that? or install new drivers without it active?
 
@phanbuey its Friday. I'll forgive you. No more beers for you mate. :toast:
 
@phanbuey its Friday. I'll forgive you. No more beers for you mate. :toast:

I think the current lack of might be the problem :).

Try to disable geforce experience. It has a built in framelimiter that uses memory clock like you're showing.
 
the sad thing is i even looked for it.

It sounds like a dying card or PSU.

Can you put your PC power settings to high performance... have you installed any new software or updates?

EDIT: ALSO- Geforce experience is known to do this sometimes - can you uninstall that? or install new drivers without it active?
No new software. When I'm at my PC tomorrow I'll check the power settings but I'm pretty sure that is set accordingly but you never know
 
No new software. When I'm at my PC tomorrow I'll check the power settings but I'm pretty sure that is set accordingly but you never know


The good news is, it's unlikely that it's a dying card, since ive never seen a card die that way (ever) - they usually artifact and then kick the bucket. So it's most likely a software issue.
 
I think the current lack of might be the problem :).
:laugh:

Try to disable geforce experience. It has a built in framelimiter that uses memory clock like you're showing.
good suggestions I'll check that

The good news is, it's unlikely that it's a dying card, since ive never seen a card die that way (ever) - they usually artifact and then kick the bucket. So it's most likely a software issue.
My thoughts exactly. This is why I am hesitating to mutilate my card or take my PC apart just to diagnose this problem
 
@Wolvyreen …… Welcome to TPU. Just to set the record straight..... replacing thermal paste is not mutilating a graphics card, over time thermal paste degrades and often, as a result of that temps increase, this coupled with the fact that some Card manufacturers use middle of the road quality paste (at best) the suggestion would be on anyone's list (certainly any enthusiast's) as a goto potential solution or simply to rule out anything relating to heat/throttling/instability etc.

Whilst I respect completely that this may be new to you and therefore possibly daunting, in itself a good reason not to do it at least until other solutions have been tried, this site has many enthusiasts so I would consider the suggestion to be legit...… either way I hope you solve your problem.
 
@Tatty_One thank you but I consider removing any part of a card such as a heat sink or changing the thermal paste to my own preferred brand or modifying anything that the manufacturer did to the card out of the box, to be mutilation. We can agree to disagree.

While I never said that this was the wrong thing to do, changing the thermal paste might very well be the solution, im simply saying that there are a lot of other things to try before taking my card apart and this should never be the first approach while there are safer things to try first
 
Back
Top