04-02-2019 03:54 AM - last edited on 03-06-2024 08:28 PM by ROGBot
04-02-2019 05:24 AM
04-02-2019 11:47 PM
jab383 wrote:
It's been a while since I've used my R5E10, so I can't get too specific. That overclock that ran for a long time was pretty good for X99. It doesn't appear that there was much stress on anything. Recent start of random crashes may indicate a slight degradation that makes voltages that worked before be not quite enough now. Degradation happens - its inevitable. My long time 4790K has gone through two stages of voltage increase to keep it running at a mild overclock.
BSOD 124 usually indicates Vcore too low for the clock. The cure is usually to increase Vcore by a step or two of .02 volts.
Freezes without a blue screen are often caused by memory errors. The combination of core and RAM errors/crashes could also be due to Vccsa and/or Vccio too low (or too high, but that's unlikely) due to slight degradation of the System Agent part of the CPU. If Vccsa and Vccio are under 1.25volts, try increasing by one or two steps of .02 volts.
Prime95 and memtest 86 are not very good. Prime 95 is all stress and no test - its high temperatures can cause degradation and it doesn't check it's results for errors. The Realbench test is better. MemTest 86 hasn't worked well at all for DDR4. Try HCIMemtest.
04-03-2019 12:27 AM
04-03-2019 03:16 AM
Arne Saknussemm wrote:
Try raising VCCSA to about the 1.0v level...
Also worth noting that it doesn't take many BSODs to FUBAR the OS...so a fresh install might help
04-05-2019 03:23 AM
04-05-2019 03:33 AM