Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
7800x3d
7900xtx
64gb ddr5
1200w psu
b650e taichi
Has happened 3x so far, completely random. Although it does seemed to be tied to how long the game has been open.
I can't duplicate the issue with other games or software. System is very stable.
Is my cpu fine or something?
Corrected Machine Check
Translation Lookaside Buffer Error
screen:
https://i.postimg.cc/5f7c2XTg/whh.jpg
This is riciciolous all should be stable from start to end. I am so mad now and have headache what to rma now. Thanks
Either raise your CPU Voltage (vcore) or reduce the P or E core clocks.
This error has nothing to do with the video card. It's from an unstable overclock.
Shader compilation in Unreal Engine 5 uses AVX instructions and absolutely HAMMERS the CPU. The reason why you get a 'video' error is because the compilation failed, and under a 'fully' working stable CPU, this is usually video memory related--but the system doesn't know your overclock failed.
You get the same performance stress by running Cinebench R15--which oddly enough, is harder to run on overclocked systems than R23 is (Not sure about the new version 2024).
Check hwinfo64 (NOT hwmonitor), you will probably see WHEA errors at the bottom.
If you want to check this, you can delete the shader cache.
its stored in appdata / local / nvidia / dxcache.
Delete the files there then re-run the game, and it will recompile shaders. If you're unstable you'll get out of memory errors, machine check or BSOD's (clock watchdog timeout--error 0x101).
*Edit* if your CPU is COMPLETELY STOCK (besides XMP memory settings), blame the gigabyte BIOS for probably doing auto overclock for you. I don't have a gigabyte board (the last one I had was their Z490 Aorus Master) so I can't help you too much. But I can give you some things you can try.
Try:
Advanced Mode:
Fixed cpu P-core ratio: 55
fixed cpu E-core ratio: 43
Min/max cache (this may be called "Ring") ratio: 45
CPU Vcore : Fixed mode (NOT override mode--on Z390 and older, override mode is what fixed mode is now, but now, override mode changes the CPU VID, fixed mode changes the VCORE):
Fixed mode: Vcore: 1.30v
CPU Vcore Loadline Calibration: Turbo (this should be in your VRM settings).
XMP Memory: XMP 1.
This should be stable on 100% of CPU's regardless of their silicon quality.
This should be guaranteed stable.
if it is, you can try reducing the cpu voltage to 1.275v and test that (fixed mode, remember).
I buy new pc and problem will be the same. I bet that. I mean the same combo. Thx Falkentyne for answer. But why many peoples on reddit had the same issue during that shader compilation during first boot.............? on 13900k and 4090 combos. Thats why i ask.
Not with a BSOD. A CTD is not a BSOD.
On second try and again its fine. Just on first try after updated drivers. I dont know if this is faulty cpu or just what?
I dont have any crashing in games etc.
Buying 14900K is like throwing money on in without solving the problem, which is what 13900K is to begin with (the problem).
You get it precisely when your CPU is stressed to its full potential, for example by compiling shaders.