![]() ![]() They just have something not right at the moment, hard to pin down exactly what though. Switching to any newer drivers always ends up with crashes and considerably lower frames again. Haven't had a single instance of a locked GPU since in 3 days, or so of heavy testing. Rolling back to a Radeon driver build maybe four, or so revisions behind has really improved stability for me, and the frames have increased dramatically (talking 32fps back to +55fps again.) (something else I'm thinking isn't being managed correctly because even with provided guidelines the variance in configs is too high to be accurate, which makes me think there is some sort of overvoltage spike being caused by who knows what while we're running tarkov). Seems to be keeping the temps within manageable zones much more efficiently once I lock in a dedicated fan speed, and also ensures I can see what voltage the card is receiving. I've been utilizing MSI afterburner to handle core clock and fan speeds. The main issues similar to yours also started roughly a week or so ago for me I've fiddled with an rx580 and discovered the recent AMD software with Wattman doesn't manage the fan throttling as well as one would like. Thanks for reading, I just wanted to throw this out there in the hopes that it might help someone experiencing the same issues. Ofc a downclock would cause less power consumption, which in turn might negate this too. Not sure about CPU, and I would assume a PSU error might be hard to negate. Keeping your setup to one screen might also help, as pointed out by u/NoSmallTask. So if you're experiencing a hard GPU crash, a simple downclock might negate this issue for you too. So for me it was a GPU crash, I simply used AMD control panel and did a 10% downclock, along with a slightly higher fan speed, and I have been playing all day with no crashes, still getting the same performance.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |