Overclocking apps broken after Meltdown/Spectre Windows Patch


Limp Gawd
Jul 7, 2015
Hi folks,

I've come to find that my system has become unstable after enjoying nearly a year of light overclocking after installing the Windows update for the Meltdown patch.

Firstly my OC (may not reflect outdated specs in sig sorry):
7700k, 4.8Ghz, 1.24v (4.5 Ring + 4.5 AVX @ 1.21v)
MSI Xpower Titanium Z270, LLC mode 2 (dead flat)
GSkill 32GB 3200Mhz C15, 1.36v, SA 1.15V, IO 1.15V (both 1.16-1.17 on physical probe)

This test previously passed OCCT Linpack for 10 hours, passes many loops of 3D Mark etc and a good few hours of Prime95 mixed. Linpack in OCCT is completely broken as soon as it hits 100% CPU usage, PRime95 may last a few minutes, even when BIOS is cleared and reverted to default. Either reboots with no warning, or occasionally produced WHEA 0x124. I have even attempted 1.3v and it fails at everything else stock (no xmp or wehatever)

I've been noticing general instability in some games, and other heavy apps I use, usually the type of thing I attribute to slight Vcore starvation or IO/SA starvation. But after various tests I can't deem anything stable, prime95 and OCCT Linpack are unusable to do any stress testing, the will crash the PC at 100% usage. However the Intel processor diagnostic passes every single one of them. For shits and giggles I setup a non-XMP 4.8Ghz @1.2v (LLC 2) and it passed the Intel PDT. That should lead to Vcore related BSODs in most good torture tests, so this is straight up BS.

Anybody out there sporting an overclock, and maybe even just those with decent enough hardware or whatever, can you please check your systems for any instability and confirm this is the case. I have found a couple of reddits with similar posts but the info is few and far between at this point.

I believe I had similar but have been busy lately so haven't tested much. Since then I did update to the newest bios and all all my old settings wouldn't work, even gave reboot loop. Had to work my way back to 4.8ghz but still had slight instability, increased all voltage s and couldn't get it stable. So lowered to 4.7ghz and it's solid
I've got some folks testing on my Discord server, so far no dice. I've sent my supplier a query regarding if there's been an rma trend since the patch, will wait to hear from them.
Bios overclocking best choice anyway. Im still running my 5820k at 4,4 1,275 v stable as ever. didnt notice any performance drops in any games ive played.
I'd sit out the OS and firmware patches awhile tbh, update your browsers and the only random home user threat basically goes away - javascript. (opening email attachments or downloading files from dodgy/compromised sites is not really random) Enterprise has bigger problems but you'll have to pay me for advice there, get in line too ;)

This shit isn't even known stable on servers now, and there are still no public details about why or what configurations matter:


As part of this, we have determined that similar behavior occurs on other products in some configurations [DETAILS? NOPE!], including Ivy Bridge-, Sandy Bridge-, Skylake-, and Kaby Lake-based platforms. [Haswell and Broadwell mentioned in previous release] We have reproduced these issues internally and are making progress toward identifying the root cause. In parallel, we will be providing beta microcode to vendors for validation by next week.

Even Intel itself is willing to admit every 'core' cpu arch since 2009 could have issues with the patch lol!
I work as an assistant sysadmin myself, I understand how the attack is deployed, but I'd hate to think that shits gonna start crashing at a dime because of these updates. They've had since like mid last year from what I understand when they first found out, I'd hate to think they pushing broken patches in a sudden rush o_O

Ive got some folks reporting stable and fine, I've got others seeing some issues, just can't pin point what it is. Hell I can't even default my Bios and get my system to run atm, constant cache bug checks from CPU. A friend of mine suggested it could be some drivers going to shit with the patch, but other wise no concrete evidence.