add arrow-down arrow-left arrow-right arrow-up authorcheckmark clipboard combo comment delete discord dots drag-handle dropdown-arrow errorfacebook history inbox instagram issuelink lock markup-bbcode markup-html markup-pcpp markup-cyclingbuilder markup-plain-text markup-reddit menu pin radio-button save search settings share star-empty star-full star-half switch successtag twitch twitter user warningwattage weight youtube

980 ti Strange Anomalies

G_rot777
  • 21 months ago

First off, no artifacts or lag. Temps are fine. My little brother's 980 ti was switched out with a gtx 960 for many months because we thought it was an issue with the power supply, now that he has a new one it's still doing it. When he plays games the computer would just shut down. I loaded up firestrike and it would crash during the 1st scene. I loaded up and left afterburner running in the background with a custom fan profile and it wouldn't shut down randomly, the score is quite impressive too (system overall 12,117 and a graphics score of 18,203) I told him when he games just to leave afterburner running in the background until I find a fix for it. Well turns out it shut down during a Counterstrike competitive match with it in the background. He told me the temps were at 52c.

Useful information: -EVGA GTX 980 ti SC+ -Was an openbox from microcenter -No overclock -GTX 960 ran fine in his system for months -Tried different powersupplies (CX600m and new one is a CX650m) -Tried different cables -Reset bios -I have not tried the card in my own system yet -No lag -No artifacts -Temps are fine -Computer is a Lenovo k450 with the i5-4440 and 8gb ram

I'm lost, someone help me out, I don't know what else to try. The card used to run without doing this, it started doing this one day while he was playing PUBG with me almost 4 months ago. Before he got into a match the splash screen would come up and the computer would just shut down.

Comments

  • 21 months ago
  • 2 points

Hi, G_rot777,

do I get it right, that:

  • So the GTX 980 Ti acts up with both the Corsair CX600M and Corsair CX650M.

  • The GTX 960 does work fine with both the Corsair CX600M and Corsair CX650M.

  • The GTX 980 Ti works fine under normal circumstances, it only makes the PC crash is specific games (PUGB, CS:GO).


From your post, I gathered that you have a computer as well. Have you tried running the suspicious GTX 980 Ti in your system? Does it work there? If it's not an easy one (e.g. you have a full custom loop with a different video card water block), do you have any other system (that is not your brother's) and test it there?

I told him when he games just to leave afterburner running in the background until I find a fix for it. Well turns out it shut down during a Counterstrike competitive match with it in the background. He told me the temps were at 52c

What did shut down? Afterburner/CS:GO/the GTX 980 Ti/the whole PC?

I had some issues with my GPU as well, and only in certain games. The temperatures of the graphics chips were fine, however, the VRMs (Voltage Regulator Modules) shot to the sky in a certain game. The game was GPU bound, so the GPU was stressed to the max. I used AIDA64 to monitor the GPU status VRM as well as processor temperatures.

For me, a solution was to increase the airflow around the VRMs. For this, I cranked up the GPU's fan quite a lot and I could see that the VRMs temperatures did not creep above 97 C (I think that's where they crashed for me) anymore, rather stayed somewhere around 60 - 70 C. Now the whole GPU is under liquid cooling. This is a GTX 970 STRIX, for the record.

Check the VRMs and if they do not overheat when the crash happens (useful to log temperatures to file or see them on a 2nd display), then try it in a different PC in similar circumstances (gaming the same games in as close hardware as possible).

Another good idea could be trying the exact same model of video card replacing the one that acts up. Would it produce the same error? (Does anyone you know have the same/similar GPU?)

The problem could be many things:

  • The graphics card overheats (processor or VRMs, or memory chips, etc -- double check the chip temperatures as well before it dies ==> logging, monitoring)
  • Other GPU issue.
  • PSU (not necessarily bad, but somehow not supporting enough power, if the PC has high power draw -- 600 W to 650 W is not much different, though you'd need extreme power draws from lots of other components to this become a limiting factor).
  • Wrong settings in the motherboard (though I don't suppose it supports overclocking -- try loading the BIOS defaults)
  • Other motherboard issue

The only way to figure out what's going on is to keep testing the components individually and replacing them one by one (not buying new ones, just replacing them for a test period, by borrowing from friends, family, etc).

  • 21 months ago
  • 1 point

I also have a 980 ti in my own system. I will try switching them tomorrow. What settings could be wrong in the bios? The whole computer turns off. I forgot about the vrm temps, I'll have to enable monitoring for that. The computer seems to shut down during higher load on the gpu.

  • 21 months ago
  • 1 point

What settings could be wrong in the bios?

There are timings that involve bus speeds in the BIOS for instance. Also, voltages that affect the PCIe busses. So loading a BIOS defaults make sure, that everything works according to the factory settings (which had been tested thoroughly).

The whole computer turns off.

Yeah, that's a bit strange. When I had problems with my GPU (VRMs overheating) only the running game crashed. Not even the operating system. Do you have a Blue Screen of Death (BSOD) or it just turns off immediately as if someone cut the power?

Let's see how it works if you guys swap GPUs. Do you have a higher capacity PSU in your system? If you do, you can leave your brothers GPU in his system and swap the PSU only to see if his PC works fine with your PSU (though 650 W power draw is rather extreme...)

  • 21 months ago
  • 1 point

Ok so bios is out. The whole computer does just shut off as if the cord was unplugged and then restarts on it's own, no BSOD. I did notice his core clock is higher than my ftw 980 ti, since the card was an openbox (meaning pre-owned) did someone flash a ftw or kingpin bios on the card? My computer has a 750 watt gold rated evga powersupply.

  • 21 months ago
  • 1 point

Update: put it in my own system, instead of shutting down the display goes out and the fans in the computer ramp up it seems. My asus monitor says "out of range" instead of no signal.

  • 21 months ago
  • 1 point

LOL. So his card may have been tweaked, and that may cause the acting up...

I'd research how to put the original (or latest approved) firmware back on the card. Contact the manufacturer if you must, they should provide you with details.

Sort

add arrow-down arrow-left arrow-right arrow-up authorcheckmark clipboard combo comment delete discord dots drag-handle dropdown-arrow errorfacebook history inbox instagram issuelink lock markup-bbcode markup-html markup-pcpp markup-cyclingbuilder markup-plain-text markup-reddit menu pin radio-button save search settings share star-empty star-full star-half switch successtag twitch twitter user warningwattage weight youtube