- Apr 1, 2013
- 517
- 3
- 19,365
Recently upgraded my gpu from GTX970 to GTX1080 (few months ago). It was all working fine until now I have updated my Gigabyte Aorus Engine which helps me to select oc mode. After few days of this upgrade I started to get artifacts in game and the game crashes whith this wired message "The GPU has physically removed". When I checked the clocks I saw now the aorus engine engaging wrong clocks in oc mode. Previously it was 1936 core, 10400 mem. Now 1936 core and 10203 mem but during game play my memory clock even touching 10590. Manually tried to reduce it to 10400 but now observed still it is crushing. Many times this crash is so root level that it is leading to BSOD and total system freeze.
Now I have seen oc failure with older gpu as well but never faced this type of BSOD, gpu removed msg.
One more imp point is, some times I have faced during cold boot my system starts for a fraction of second then stops and not booting at all. Like a short circuit however if I turn off ac power and try once or few times more it turns on. No short circuit or anything in my case. All are quality components even the psu cables which are custom made.
Tried to test with Kombuster but stable for 15min at least but in game it crashes sometimes in less than 5 min.
Any suggestions to trouble shoot will be very helpful to me. thanks for reading.
Update 1: My gpu not crashing at much higher clock in Kombuster memory as well as core burner. Like for memory runs smooth at 10.59Ghz (effective) and 2.04-2.06Ghz core. but even i down clock it in games system crashes at 1.98Ghz and 10.29 mem most of the times within 10 mins. For some reason I am thinking it is the psu causing this issue. I think it is not possible to get a driver crash due to poor gpu overclock that will freeze my system so hard that it is showing BSOD and restarts (whole system restart on it'es own).
Update 2: Very random and seems less related to clock. Sometimes stress test running for long but sometimes crushing. Down clocking not solving this issue.
Update 3: As per gigabyte clcok should be
Boost: 1936 MHz / Base: 1784 MHz in OC mode
Boost: 1898 MHz / Base: 1759 MHz in Gaming mode
but in oc mode actual clock is hitting 2.06Mhz (2.03-2.04) and memory is crossing 10.59Mhz. Is it normal or for some reason it is pushing too much (gpu boost)?
Now I have seen oc failure with older gpu as well but never faced this type of BSOD, gpu removed msg.
One more imp point is, some times I have faced during cold boot my system starts for a fraction of second then stops and not booting at all. Like a short circuit however if I turn off ac power and try once or few times more it turns on. No short circuit or anything in my case. All are quality components even the psu cables which are custom made.
Tried to test with Kombuster but stable for 15min at least but in game it crashes sometimes in less than 5 min.
Any suggestions to trouble shoot will be very helpful to me. thanks for reading.
Update 1: My gpu not crashing at much higher clock in Kombuster memory as well as core burner. Like for memory runs smooth at 10.59Ghz (effective) and 2.04-2.06Ghz core. but even i down clock it in games system crashes at 1.98Ghz and 10.29 mem most of the times within 10 mins. For some reason I am thinking it is the psu causing this issue. I think it is not possible to get a driver crash due to poor gpu overclock that will freeze my system so hard that it is showing BSOD and restarts (whole system restart on it'es own).
Update 2: Very random and seems less related to clock. Sometimes stress test running for long but sometimes crushing. Down clocking not solving this issue.
Update 3: As per gigabyte clcok should be
Boost: 1936 MHz / Base: 1784 MHz in OC mode
Boost: 1898 MHz / Base: 1759 MHz in Gaming mode
but in oc mode actual clock is hitting 2.06Mhz (2.03-2.04) and memory is crossing 10.59Mhz. Is it normal or for some reason it is pushing too much (gpu boost)?
Facebook
Twitter
Instagram