980ti fluctuate pwr / core clock

[Y.sam]

Dears
I have 980ti g1
Today while playing ,fps flucating 40 to 120
Gpu z shows that core clock & pwr also going wrong ..

I try below :
Format and clean install
Win10 1090 & 1903
Clean instsll many driver ver
Flash card bios with same ver of orginal
Another psu
Another games ..
Disconect all hdd and use only ssd
.... and same issue .....


I did Screen record and upload as link below :

As we can see .... high and low ...

 

[RJARRRPCGP]

Reminds me of a GTX 660 Ti that had a wrong BIOS flashed.

 

[Y.sam]

I had backup of orginal one i will reflash it
Also i flashed it because of this probem ...

Reflash orginal bios didnt fix .
Up and down .....
May its cards capacitor ? Gpuz show perfCap reason PWR ..

 

[oobymach]

Can you take a pic of the front of the card with heatsink and the back with stickers?

Looks like a throttling issue but idk the cause (is the card dirty? have you cleaned it lately?).

Have you tried using MSI afterburner? Download afterburner with link below and install it.

Afterburner

MSI Afterburner is the world’s most recognized and widely used graphics card overclocking utility. It provides detailed overview of your hardware and comes with some additional features like customizing fan profiles, benchmarking and video recording.

www.msi.com

Once it's running hit settings and change user interface to the Default v3 skin and apply (may require a restart of the app), with the app in v3 mode click the down arrow next to power limit and uncheck the link box that appears and apply (you may also need to set general settings as I have). That should fix non dust related thermal throttling.

 

[Y.sam]

‍

 

[RJARRRPCGP]

I had something like that happen back in 2016, with an eVGA GeForce GTX 660 Ti and with Nvidia Inspector, found an Asus VBIOS on it! So to TPU, was exactly where I went! Then I found an eVGA VBIOS image with the 915 Mhz GPU clock, IIRC then I flashed that VBIOS on to it. Right after, it was like brand new, saved a GTX 660 Ti from the landfill! Before, the clock would keep bobbing in Heaven, FFS!

 

[Y.sam]

First its clean and i temp never reach 65c

Msi ab , i try it today and even 135% pwr limit , i will try v3 skin and replay to you .

Thanks for replay

Can you take a pic of the front of the card with heatsink and the back with stickers?

Looks like a throttling issue but idk the cause (is the card dirty? have you cleaned it lately?).

Have you tried using MSI afterburner? Download afterburner with link below and install it.

Afterburner

MSI Afterburner is the world’s most recognized and widely used graphics card overclocking utility. It provides detailed overview of your hardware and comes with some additional features like customizing fan profiles, benchmarking and video recording.

www.msi.com

Once it's running hit settings and change user interface to the Default v3 skin and apply (may require a restart of the app), with the app in v3 mode click the down arrow next to power limit and uncheck the link box that appears and apply (you may also need to set general settings as I have). That should fix non dust related thermal throttling.

View attachment 189105

e

 

[oobymach]

If that doesn't fix it you need to track down a different bios, taking a picture of the stickers on the back of the card will help us to track down the right one or at least one that works better.

 

[Y.sam]

If that doesn't fix it you need to track down a different bios, taking a picture of the stickers on the back of the card will help us to track down the right one or at least one that works better.

Ok .
Msi ab dosen't fix ...
Last month i did backup for bios (to show a friend how to backup bios ) even i have an old screenshot for gpu z and benchmark for the same game (farcry primal ) .
Name of bios , GM200.rom
Ver 84.00.36.00.6E
and i reflash it today (old file )

And same bios was running great ..

Pix of stickers

When last benchmark finished . Game stop working , and problem reported image attached , but i cant find the log file

 

[johnspack]

Video gurus here, doesn't that sound like vrm failure? God I hope my 980Ti doesn't do this.

 

[Y.sam]

Video gurus here, doesn't that sound like vrm failure? God I hope my 980Ti doesn't do Lq

I think its pwr or bios . If vram failure i dont think that can reach 120fps for while without crush or black screen and normal temp

 

[agent_x007]

VRM = Voltage Regulator Module
VRAM = Video Memory
^They are NOT the same thing.

First of : Use latest GPU-z when checking what limit you hit (it shows WAY more current sensors than the old version).
PWR limit may mean one or more shunt resistors are going awol on ammount of current that goes through them (making card throttle under 3D load). Liquid metal on them perhaps ?
You could use a modified vBIOS (with raised power limits), to see if this will fixes the problem, BUT if there really is a problem with shunts - you are risking shorting them out and making the card unusable until they are replaced.

Lastly : This may also mean card is close to giving up

 

[freeagent]

This used to happen with my old 970 FTW, the dual 6 pin model.. I used to remove it from the system for about 10 minutes, as well as remove the battery on the mobo. It’s like a weird uefi bug. It hasn’t happened on my 980 though.

 

[Y.sam]

VRM = Voltage Regulator Module
VRAM = Video Memory
^They are NOT the same thing.

First of : Use latest GPU-z when checking what limit you hit (it shows WAY more current sensors than the old version).
PWR limit may mean one or more shunt resistors are going awol on ammount of current that goes through them (making card throttle under 3D load). Liquid metal on them perhaps ?
You could use a modified vBIOS (with raised power limits), to see if this will fixes the problem, BUT if there really is a problem with shunts - you are risking shorting them out and making the card unusable until they are replaced.

Lastly : This may also mean card is close to giving up

Vrm & vram
Sorry for this mistakes .
I just update gpu.z and recoreded benchmark

..also log file at the same time (attached ) .
First Pci slot down from 12 to 11.6v
Both 8pin reach 70 -75 watt but its fluctating...
Im confused .......

This used to happen with my old 970 FTW, the dual 6 pin model.. I used to remove it from the system for about 10 minutes, as well as remove the battery on the mobo. It’s like a weird uefi bug. It hasn’t happened on my 980 though.

dosnt fix it ... thx for replay

 

[agent_x007]

Your VRM controller may be even more confused than you.
It just randomly throttles card based on weird reading it gets.
Three options :
Your shunt resistors are "shut" and need to be replaced (they are responsible for measuring how many amps goes through each power plug/PCIe conntector), when they go bad, you can see this type of stuff (example : PWR limit reached on 2D).
Second option is a fault in PCB (less likely than first and last option), a trace or more is damaged and prevents VRM controller from correctly monitoring power usage on your card.
Lastly, your VRM controler is busted, and needs to be replaced

 

[Y.sam]

Your VRM controller may be even more confused than you.
It just randomly throttles card based on weird reading it gets.
Two options :
Your shunt resistors are "shut" and need to be replaced (they are responsible for measuring how many amps goes through each power plug/PCIe conntector), when they go bad, you can see this type of stuff (example : PWR limit reached on 2D).
Second option is a fault in PCB (less likely than first and last option), a trace or more is damaged and prevents VRM controller from correctly monitoring power usage on your card.
Lastly, your VRM controler is busted, and needs to be replaced

Yes . First i will replace shunt resistors , then test and check if problem fix or not . Will back and replay here . After few days ...

 

[Pokkel1982]

Hi there. Any luck on the card? Mine is doing the exact same thing. Tested in other pc's and psu's.
This thread sums up exactly what happened to me.

Fluctuating frame rate and not sure what to do

1. If you look at other posters messages, under their avatar you will see "system specs", please put you full specs in there; also consider that some of us old folks, after many years of PC use, (and teenagers for other reasons) have RSI to one level or another and if the "main info" isn't out...

www.techpowerup.com

I have tried upping the voltage limit, seeing that it is not at all related to temps...

To me it seems like a voltage limit that reverts the card back to a failsafe voltage to protect the card, which causes the memory and clocks values to drop.

It's either a software issue or a hardware issue. I'm leaning towards a capacitor that cannot handle the load anymore...

I have also updated the bios with stock and custom, but same results each time.