^ That's insane but is that really 24/7 stable? I remember you saying how stupid people are for testing for 24 hours so... I mean you still got at least a very good chip (so let's say it takes .2v more than that to be 24 hours prime stable with no whea errors) but I'd love to know how stable that is. I mean I can do gaming and such on less than 1.4v.
my i5-2500 is weak, needs 1.30 for 4.2, and LLC level takes that to 1.32, otherwise it fails P95 when it drops under 1.30
Hmmm that doesn't really sound right. Can you do a more 'normal' overclock like see what voltage you need for 4.5? 1.32v for 4.2 (which is probably more like 1.35 max voltage since software is inaccurate and your true
voltage is going to be a bit higher) just sounds like something is wrong.
My 3570K needs 1.27v for 4.4GHz, 1.325+ for 4.5. I don't have an exact voltage for 4.5 as I've never really stability tested, just bumping the voltage up when I encounter instability.
Whilst I can run at 4.5 with alright temps, I think I'm just going to keep it at 4.4GHz for the lower voltage and lower temps. It's not like 100MHz will make much difference
Then you are probably nowhere close to stable. I can game under a good .1v less than what I really need to be stable. It all depends on what kind of user you are though. I'm a streamer, so it's very very important for me to test for 30+ hours on an overclock, as it's just unacceptable to crash when streaming with a live audience that'll switch streams just because you paused the game. And, I use some of the slower presets, which will make a system that passed 24 hours of prime95 but isn't quite rock stable, crash in seconds. So stability is important to me. Plus, overclocking every little bit helps in a heavily memory dependent, multi-threaded application like streaming which will bring an i7 to it's knees (but any modern quadcore can stream, i streamed HD just fine on my athlon ii x4, but an i7 is obviously lightyears ahead of it especially with a good overclock).