i7-4770k, bad binning/below average (does 4.6ghz @ ~1.35 for stable, 4.6@1.2 is lucky to post)
Delid CLU/CLU
H110 4x140mm YL-H blues
Z87X-UD3H v1.0
F6 BIOS
2x4GB Hynix CFR
Temps stay below 80C as max
Uncore 4.5ghz@1.3VRING unless otherwise noted
Using ~40minute H264 benchmark as stability test (way more intense than AIDA, seems like it's more intense than p95 according to a few people, non-AVX) for stability testing.
So I verified my chip stable in the first few days as stable at 4.8ghz@1.45vcore/1.95VRIN. The next day, suddenly, it was unstable. Chip degradation, break-in, whatever.
So, I have to settle for 4.7ghz@1.43vcore/2VRIN.
Then, I find out, that using 125 BCLK Strap, I can do 4.75ghz@1.44vcore/2.03VRIN. Awesome, right? I thought BCLK straps were worse or something, but it was pretty easy, quick 50mhz bump up.
I try to make 4.8ghz work many, many times - I try 1.475vcore (which is 1.506v on DMM), I try to do way more VRIN, etc, etc, no dice.
Then, I try 47 x 102.14, and it works! 4.8ghz, 4.49uncore, pretty sweet, this was with some 2600mhz+ ram speeds too (haswell's IMC is beast, I can do 3ghz easily on these $55@newegg ram). Going to raise BCLK a bit so I can get uncore to 4.5ghz, but I did not have these kinds of results on Ivy.
I also tried 125BCLK strap for 4.8ghz, it was fail post but I'm pretty sure that's because I had uncore at 4.55ghz, I know it's just right on the limit of stable at 4.5@1.3vring, but if you guys want I can test it (just set uncore to 4.45ghz, i just personally dont care ill just use bclk only because i dont want to sacrifice uncore if i dont have to).
I mean the difference is night and day. 47x101 is just stable, no change in voltage from 47x100=4.7ghz. 48x100=4.8ghz, it crashes within 5 minutes.
TLDR I've had a lot of success increasing my core overclock significantly using both BCLK and BCLK straps, this is part advice and part anecdotal/discussion. I thought this finding was worth it's own thread rather than a post in a club.
Delid CLU/CLU
H110 4x140mm YL-H blues
Z87X-UD3H v1.0
F6 BIOS
2x4GB Hynix CFR
Temps stay below 80C as max
Uncore 4.5ghz@1.3VRING unless otherwise noted
Using ~40minute H264 benchmark as stability test (way more intense than AIDA, seems like it's more intense than p95 according to a few people, non-AVX) for stability testing.
So I verified my chip stable in the first few days as stable at 4.8ghz@1.45vcore/1.95VRIN. The next day, suddenly, it was unstable. Chip degradation, break-in, whatever.
So, I have to settle for 4.7ghz@1.43vcore/2VRIN.
Then, I find out, that using 125 BCLK Strap, I can do 4.75ghz@1.44vcore/2.03VRIN. Awesome, right? I thought BCLK straps were worse or something, but it was pretty easy, quick 50mhz bump up.
I try to make 4.8ghz work many, many times - I try 1.475vcore (which is 1.506v on DMM), I try to do way more VRIN, etc, etc, no dice.
Then, I try 47 x 102.14, and it works! 4.8ghz, 4.49uncore, pretty sweet, this was with some 2600mhz+ ram speeds too (haswell's IMC is beast, I can do 3ghz easily on these $55@newegg ram). Going to raise BCLK a bit so I can get uncore to 4.5ghz, but I did not have these kinds of results on Ivy.
I also tried 125BCLK strap for 4.8ghz, it was fail post but I'm pretty sure that's because I had uncore at 4.55ghz, I know it's just right on the limit of stable at 4.5@1.3vring, but if you guys want I can test it (just set uncore to 4.45ghz, i just personally dont care ill just use bclk only because i dont want to sacrifice uncore if i dont have to).
I mean the difference is night and day. 47x101 is just stable, no change in voltage from 47x100=4.7ghz. 48x100=4.8ghz, it crashes within 5 minutes.
TLDR I've had a lot of success increasing my core overclock significantly using both BCLK and BCLK straps, this is part advice and part anecdotal/discussion. I thought this finding was worth it's own thread rather than a post in a club.