Massman

Gamer is the intelligent one?

HAHAHA! ALL of them are beyond awesome!

Alright, done testing the 3rd one of the day. L152B586, so an old batch again. Pretty shite CPU, only 6270 32M with 1c. THRASHBIN! One left ... luckily this one is from batch el magico again.

Muahahaa. Gamer attracted spambots!

Quickly tested another one, L202B763. - 6100 4c8t - cold issues starting around -140°C or so Shite! Next!

National holiday here, which means time for more CPU binning action. This is a L152B586 chip, does 4900 pi32M on air. Only 6380 PI-32M, so sucky-suck-trash-can. Seems to be a very hot chip too. I think the 3204 chips are just better because the production process of IB is getting more and more optimised. Looking forward to test chips that are even newer! Next!

As you might know, I have this 180L dewar at home to do easy (read: I'm lazy) LN2 refills when needed. Once in a while, my teammates swing by the house with their dewar and fill 'r up. Normally, they call me before they come by, but recently I noticed that the big tank was emptying quicker than before. So I installed a security-cam. And this is what happened today ... Here's how the situation normally looks Look at that ... we have an uninvited guest named Gamer Well, well, that sure looks like a really heavy dewar you have there. Might be filled? In conclusion: Gamer is a big big LN2 thief!!! CAUGHT YOU GAMER!! :banana::nana:

Yes, HD4000 is allowed IF on an MSI mainboard.

I'm not really checking for voltage scaling specifically, but I don't really see any big difference going from 1.4 to 1.5V. I assume it's related to the CPU temperature but, as said, I don't really check for scaling. For what it's worth, I've also tested Ivy with a ROG Z68 before and I wasn't really too happy with the system. Too much crashes and cold issues. The Gene V was doing everything perfect, as well as the other six or seven Z77 boards I've tested under LN2 now. Imho, for Ivy it's best to just move on to Z77.

2c2t to keep temps down. Seriously, I'm not saying it's a perfect rule like Sandy Bridge had one, but just mentioning it because it seems consistent with the few chips I tested .

I'm not even mounting the CPU cooler, really. Just some thermal paste, twitch'n'turn the heatsink on it so it creates a vaccuum and then one low speed 120mm fan. The key of this 'air process' is probably keeping all the variables like thermal paste, mounting 'pressure', heatsink etc the same. Only that way you can compare chips and see which are doing better under air. I don't think it's possible to compare online with other folks as the environment parameters are just different. 1.4V.

I'm just testing, man. I don't want to spend time finetuning ... that's something for later. Tested 2nd chip of the night, this one did 5100 pi32m on air. A solid 6.6G+ one, but no 6.7G. Oh well, good enough! FYI, we've now tested about 7 of this one retail batch and got 1x6700, 2x6600 and 1x6500. Good good.

Quickly testing another one ... this seems better

No

6.1 4c8t is trash

Tested another one on LN2. Did about 5G on air 32M, but total shit on LN2. Batch: 3204B824 (like the 6.7G one) - 6.1G 4c8t with full pot - 6.5G pi with full pot Tbh, if the chip you have doesn't do 6G wprime at -100°C just ditch it. Still have to find a retail Ivy with cold issues though . Onto the next one!

You're also the only man touching another naked man.

Mainly because of this:

I might be doing something wrong; still about 9s too slow. No waza or subtiming tweaking, though ... so maybe it's just laziness.

Finally got a board that will match the power of the 6.7G chip, lol

As long as they give it to Uncle Fester last, I don't think there will be an issue.

No, the new ones I'm testing. First one is safely stored in the box until a better board comes along.

Alright mates, your complaints have been heard. Mainboards can be shared, passed on. All performance altering components (CPU, VGA, memory) can not be shared. With these benchmarks, the mainboard serves more as a competition compliance requirement than anything else. Or as Vivi said it: So, now get benching already! EMEA, yes.

So this new chip doesn't do 4800. Looks like I struck negative gold again!

Oh well ... let's check the video footage again!