The Official Opteron 939 Overclocking Thread

[dinos22]

Major is sitting at #2 on the 2gig stickied thread at 300HTT 1:1 with his FX-57. Long odds on his Opteron having a better memory controller, but it would be great to see a side by side comparison with both the CPUs running 300x10.

i know but MM didn't give it any chilly temps....i think he'll squeeze a few more MHz out of the memory controller under colder conditions.....too bad MM is selling his stuff.....i guess we'll see what happens on Major's rig

[Ferry82]

what was the problem?

Just burning in i guess

[bokis]

Anyone tried four sticks of (bh-5) 512mb

[HermS]

I have one suggestion that someone could make a new thread for collecting steppings of those Opterons?

Opteron 148 CABNE 0528 GPMW here, results to come later.

[groovetek]

funny that there are only two steppings that i've come across so far

144, 146-----0530APMW...........looks like us Aussies are getting these as well
150------0534SPMW
175------0534SPMW

dinos man... what source of information tells you this???

i really would like to hope we get the 0530APMW stepping as well, but i'd hate to get my hopes up unless there's some solid evidence we'll be getting some decent chips...

i'd actually be disappointed with a 2.9ghz chip... i'm looking to get at least 3ghz stable for 24/7 use...

[Perc]

i know but MM didn't give it any chilly temps....i think he'll squeeze a few more MHz out of the memory controller under colder conditions.....too bad MM is selling his stuff.....i guess we'll see what happens on Major's rig

no major sold me the cpu he wont even get to use it. but dont worry ill run it under the phasechange and post all my results

peace perc.

[Ferry82]

Please make some pics.

[Pt1t]

pic of 50 opteron is nice

[187(V)URD@]

Pictures from computersmsa

[Pt1t]

lol , too big

[Waxman]

This is what I got with my 146 0530APMW for 24/7:




That's watercooled.

It's not that bad

[dinos22]

dinos man... what source of information tells you this???

i really would like to hope we get the 0530APMW stepping as well, but i'd hate to get my hopes up unless there's some solid evidence we'll be getting some decent chips...

i'd actually be disappointed with a 2.9ghz chip... i'm looking to get at least 3ghz stable for 24/7 use...

the first 146 CPUs delivered are 0530APMW...fingers crossed the rest of them would be the same

groove my man you've orders how many chips again 2,3,4....surely one of them will be a 3GHz 24/7 one......but you and I know that 1MB cache gives a boost so even if it's a 2.9GHz stable you are still going to have a quicker chip than your Venice........once you pop the IHS you are looking even better for a 3Giga

[dinos22]

no major sold me the cpu he wont even get to use it. but dont worry ill run it under the phasechange and post all my results

peace perc.

very much looking forward to this post

[WesM63]

w00t my 144 is on its way. Should be here tomorrow

Just found out my ETA has been pushed back till 10/10. (monday)

[Justinkoko]

where can i order it officially/unofficially?

links please!

US buyer

[koosjebram]

Any results for the Opteron 144 CABGE 0536 VPAW ????????

[foofreaknfighte]

I am hunting for a 144, 146, or 148 if anyone has a lead....

Need one!

[koosjebram]

Read previous page

[HermS]

Today I finally got to play around with my 148.

I didn't mess around with the stock cooling and stuck it straight under my Prommie. I had some major headaches as this is the first time I've needed to use dividers to achieve my maximum overclock, but after checking up on one of eva's divider guides I got it running nicely. Changing Max Async Latency to '9' made a world of difference.

I've only been checking Super Pi 1m so far but here's a brief rundown of what I've managed:-

1.450v - 298 x 11 (5:6) 3280MHz
1.475v - 302 x 11 (5:6) 3322MHz
1.500v - 305 x 11 (5:6) 3355MHz
1.525v - 308 x 11 (5:6) 3396MHz
1.550v - 311 x 11 (5:6) 3421Mhz

Thats as far as I've pushed it and still haven't tightened up the timings yet.

Edit : fixed link - thanks wes

[WesM63]

Nice Clocks!

btw, ur last pic doesn't work. (3421mhz)

[socrilles]

If we really want to know what the silicon being used with these chips is, I think the first five letters of the second line indicate it. When people were buying the mobiles these first five letters were the same as the FX53's and people were getting similar clocks (I love my 3000+ DTR) both were CAA2C. I have seen a pic of an FX57 w/ CABCE, and the opterons have CABGE so pretty close and might mean same assembly line/silicon

Edit:

It appears as those the 3700+ and 4000+ are CABGE, while the FX's are CABCE, so it seems more likely these are the same as 3700+ and 4000+ and not the FX

[ozzimark]

SILICON! no e

but anway, i agree with you. the stepping (CABNE, CABGE, etc etc) probably has a lot to do with the mask used, which production line, transistor type, maybe even wafer quality?
this probably has a pretty large effect on how the chip is going to perform, excluding random defects and such.

[Naasar]

i just ordered my 146 ......should be here sometime by the end of the month

whats the best s939 cooler on the market? xp90 or 120...or the big typhoon

[computersmsa]

Opteron 144 CABGE 0536 VPAW
DFI nF4 Ultra-D 702-2BTA
Puit Cooling -> Full = ~ 35°C
Max Stable -> 2729 MHz @ 1.49V

[Vapor]

The first 146s and 144s were CABNE 0530APMWs and the newer ones (148 included) seem to be CABGE 0536VPAWs, so we really have no way to tell what is responsible for any differences in OCability as there isn't any similarity in the steppings (i.e., CABGE vs CABNE comparo would need all else equal while the converse is the same).

EDIT: computersmsa....that's a slightly dissappointing result
What exactly is puit cooling? Stock? Have you tried loosened preamble?