Noxinite

How bad was the CB in the end?

Sweet efficiency!

You need a Futuremark link!

Nice!

Come on people a good L3014 could beat this on LN2.

Full pot to -80C, outch...

Hmm, I see. I only just noticed the stage (nor the rules?) doesn't display what preset we should be using. I assumed performance... @mllrkllr88 @TAGG

Examples?

Did that actually become a rule in the end?

Sweet clocks, you need some better ram though. What kind of CB/CBB did you have?

Best ambient so far at 4GHz (+/- bclock movement): Not sure of effi or anything atm.

For reference I got 7200 with 4.5GHz on QX9650. x9.5 multi didn't want in work on phase, so had to drop FSB and lose effi.

I already told you, 2V for long periods of time is not good for hypers. I have 2 sticks that have become dead/degraded after something like 0.5-1 year. While you're all here, is it just me or does base clock jump around a ton for anyone else? Like +/-2.5MHz. About as bad as Z170/Z270 for trustworthy low clock results, if not worse.

Eh, I like having things to tweak and improve results. Yeah, I had a Thuban result from 2 years ago with 2100 7-9-7 32M stable, but couldn't remember if that was just a Thuban thing as I sold that chip. It could do 3.4GHz NB on 120mm AIO - I regret selling it now.

Now show us the sandbag with decent NB. :P

Bah, that's no fun. But good to know I guess,.

So far I've found that none of the voltages really moved the ram freq wall, but NB (not CPU NB) helped stabilise settings that you could already POST at. The wall was the same for 8-8-8 and 6-7-6, so definitely not the ram being an issue. I also didn't need them for stability at what I could POST at. Moving to the inner slots allows me to POST at higher freqs, but is then unstable at the tighter timings that the outer slots could run. When I put the CPU on phase quickly the ram freq I could POST at seemed to increase a bit, but not massively. More testing needed here as ram timings and voltages seemed to make a bit of difference. I also found that bioses 3027 and 3029 (Crosshair IV Formula) allowed higher ram freqs to boot, but they also had an issue where they would fail POST at any setting until you loaded defaults, rebooted and reset the same settings. So I moved to 2101 for initial testing.

Core voltage was probably 1.4-1.425V for this run. Doesn't scale with more volts as it gets too hot.

I've been playing with my hypers on a Crosshair IV using a 555 BE, and it refuses to POST at higher than 244MHz bus with the 1:4 RAM divider. The funny thing is that it is fully stable at 242MHz bus with 1:4, and the lower ram dividers work at the higher bus as well. I did find some stuff on this forum about enabling ECC on non-ECC ram and that was supposed to help mem OC, but when I quickly played with it, it didn't seem to help. So the question is, anyone have any tips for getting higher ram clocks working? Or do you just need a Thuban IMC to push higher? Edit: I forgot to add that I've tried PSC quickly as well as newer/older BIOSes, but no real difference there.

You have no prices for those CPUs?

So close yet so far from 1st place AM2. ?

That means it has been set as 6x 8350K chips, not 6x cores.

Screenshot. :(

Another mod/maintainer/developer quitting because the community attacked them.

There is a bios here somewhere I believe for the non-M board, but not the one in the OP. People normally buy used dewars off EBay/other members instead of buying new ones. I think the non-M board tops out at 3866 12-12-12 1T, it can only do 4000 at 2T IIRC.