Question XMP on Gigabyte UD Z690 DDR4 causes constant bluescreen when opening demanding software

AlexanderDK

Distinguished
May 11, 2016
31
5
18,545
It's with great grief that I have arrived at this point, a lot of trial and errors trying to figure out why my PC just kept bluescreening when I enabled XMP, but now even without XMP enabled I occasionally get the same BSOD "clock watchdog timeout".
It's pretty known already that the gigabyte z690's and a lot of others z690 boards are notorious for being unstable, so I was already pretty sure that it was my motherboard that is causing the issues.
This post sums up pretty much my issues: View: https://www.reddit.com/r/gigabyte/comments/sc1wzh/gskill_ddr4_ram_totally_unstable_with_gigabyte/


My current setup is a 12900K on the z690 Gigabyte Ultra Durable DDR4 MB, running G.SKill TridentZ 2x16gb CL16 which can run at 3600 on my old system fine but on my current I cannot run it at XMP so I am stuck at 2133.

The question is, should I try and spend money on the RAM kits that have been confirmed by gigabyte to work on the motherboard, or shell out for a 'better' motherboard that is more likely to run using my tridentZ's? I'm really hoping you can lead me in the right direction, because this is really affecting the already limited time I am on the PC.

The funny thing is that the reddit post mentions the F6 BIOS update, now we're on F22 and it really hasn't fixed my issue at all. I am also unable to get an RMA due to my location, and I'm sadly past the date to return it.
 
Solution
When I was on F22 on gigabyte z690 ud ddr4 rev 1, I got bsod, "Clock_watchdog_timeout", every time I stressed the cpu (12600k) in multicore load. Either by cinebench or by other cpu intensive load. I could stress test in single core with no issue.

I have g.skill ripjaw 2x16 (F4-3600C16D-32GVKC)

I tested 'everything': Ram (ramtest86) , psu, gpu, etc. and ran extensive testing on each core, turning off and on individual p-cores. Results was inconclusive even if dmp-files reported cpu error on cpu number 6, it probably wasn't, based on my testing.

I could most of the time get it to complete stress test (no bsod) with one p core turned off, or running ram in single channel or non-xmp.

Since it seemed like it was not cpu's fault, but...

kanewolf

Titan
Moderator
It's with great grief that I have arrived at this point, a lot of trial and errors trying to figure out why my PC just kept bluescreening when I enabled XMP, but now even without XMP enabled I occasionally get the same BSOD "clock watchdog timeout".
It's pretty known already that the gigabyte z690's and a lot of others z690 boards are notorious for being unstable, so I was already pretty sure that it was my motherboard that is causing the issues.
This post sums up pretty much my issues: View: https://www.reddit.com/r/gigabyte/comments/sc1wzh/gskill_ddr4_ram_totally_unstable_with_gigabyte/


My current setup is a 12900K on the z690 Gigabyte Ultra Durable DDR4 MB, running G.SKill TridentZ 2x16gb CL16 which can run at 3600 on my old system fine but on my current I cannot run it at XMP so I am stuck at 2133.

The question is, should I try and spend money on the RAM kits that have been confirmed by gigabyte to work on the motherboard, or shell out for a 'better' motherboard that is more likely to run using my tridentZ's? I'm really hoping you can lead me in the right direction, because this is really affecting the already limited time I am on the PC.

The funny thing is that the reddit post mentions the F6 BIOS update, now we're on F22 and it really hasn't fixed my issue at all. I am also unable to get an RMA due to my location, and I'm sadly past the date to return it.
Can you run MEMTEST for multiple hours at 2133 ?
 
Apr 30, 2023
1
0
10
It's with great grief that I have arrived at this point, a lot of trial and errors trying to figure out why my PC just kept bluescreening when I enabled XMP, but now even without XMP enabled I occasionally get the same BSOD "clock watchdog timeout".
It's pretty known already that the gigabyte z690's and a lot of others z690 boards are notorious for being unstable, so I was already pretty sure that it was my motherboard that is causing the issues.
This post sums up pretty much my issues: View: https://www.reddit.com/r/gigabyte/comments/sc1wzh/gskill_ddr4_ram_totally_unstable_with_gigabyte/


My current setup is a 12900K on the z690 Gigabyte Ultra Durable DDR4 MB, running G.SKill TridentZ 2x16gb CL16 which can run at 3600 on my old system fine but on my current I cannot run it at XMP so I am stuck at 2133.

The question is, should I try and spend money on the RAM kits that have been confirmed by gigabyte to work on the motherboard, or shell out for a 'better' motherboard that is more likely to run using my tridentZ's? I'm really hoping you can lead me in the right direction, because this is really affecting the already limited time I am on the PC.

The funny thing is that the reddit post mentions the F6 BIOS update, now we're on F22 and it really hasn't fixed my issue at all. I am also unable to get an RMA due to my location, and I'm sadly past the date to return it.
When I was on F22 on gigabyte z690 ud ddr4 rev 1, I got bsod, "Clock_watchdog_timeout", every time I stressed the cpu (12600k) in multicore load. Either by cinebench or by other cpu intensive load. I could stress test in single core with no issue.

I have g.skill ripjaw 2x16 (F4-3600C16D-32GVKC)

I tested 'everything': Ram (ramtest86) , psu, gpu, etc. and ran extensive testing on each core, turning off and on individual p-cores. Results was inconclusive even if dmp-files reported cpu error on cpu number 6, it probably wasn't, based on my testing.

I could most of the time get it to complete stress test (no bsod) with one p core turned off, or running ram in single channel or non-xmp.

Since it seemed like it was not cpu's fault, but still cpu related, I tried one last ting before i went to bed 4 in the morning.. .. I though it might be some controller/instruction issue.. Idk..

I updated bios From F22 to F23. First test with Intel xtu (5 min test) with ram (non xmp) @ 2133mhz completed with no issue. Loaded xmp @3600mhz, set Intel xtu to stress test for 3 hours and went to bed.
Test passed and consecutive cinebench tests also passed. Seemed like it worked and I haven't had a bsod since.


TL /DR: Updating bios to f23 fixed it for me.
 

AlexanderDK

Distinguished
May 11, 2016
31
5
18,545
When I was on F22 on gigabyte z690 ud ddr4 rev 1, I got bsod, "Clock_watchdog_timeout", every time I stressed the cpu (12600k) in multicore load. Either by cinebench or by other cpu intensive load. I could stress test in single core with no issue.

I have g.skill ripjaw 2x16 (F4-3600C16D-32GVKC)

I tested 'everything': Ram (ramtest86) , psu, gpu, etc. and ran extensive testing on each core, turning off and on individual p-cores. Results was inconclusive even if dmp-files reported cpu error on cpu number 6, it probably wasn't, based on my testing.

I could most of the time get it to complete stress test (no bsod) with one p core turned off, or running ram in single channel or non-xmp.

Since it seemed like it was not cpu's fault, but still cpu related, I tried one last ting before i went to bed 4 in the morning.. .. I though it might be some controller/instruction issue.. Idk..

I updated bios From F22 to F23. First test with Intel xtu (5 min test) with ram (non xmp) @ 2133mhz completed with no issue. Loaded xmp @3600mhz, set Intel xtu to stress test for 3 hours and went to bed.
Test passed and consecutive cinebench tests also passed. Seemed like it worked and I haven't had a bsod since.


TL /DR: Updating bios to f23 fixed it for me.
I tried updating BIOS to F24 now and it didn't initially work, I then just tried to reinstall windows and it fixed everything, have been able to even run my system overclocked so it was probably something rooted in windows itself and the fix from the BIOS update!

Thanks for the help :)
 
  • Like
Reactions: daABBA
Solution