My x2 555 OCing & unlocking experience

SithSolo1

Diamond Member
Mar 19, 2001
7,740
11
81
x2 555 BE air cooled w/ thermalright SI-128
MSI 785GM-P45
4gb G.skill Ripjaw ddr3 1600
Win 7 Pro 64bit

Chip doesn't unlock on this board at least as far as I can tell
Tried upping the vcore to 1.45 just to boot it and see if it worked, nothin'
Didn't mess with the other voltages


Did some basic overclocking over lunch, nothin special just upping the multi:

Default voltage: 1.344v in cpuZ under 100% load
3.8ghz seems to be the max at the default voltage using only the multi, ran prime95 for an hour w/o errors and temps hovered @ 39c
4.0ghz blue screens on the login, upped the voltage a lil 1.365v in the bios but didn't help
backed it down to 3.9ghz @ stock vcore, boots into windows but blue screens when prime is started


I'm going to leave Prime running while I head back to work
 

evolucion8

Platinum Member
Jun 17, 2005
2,867
3
81
Well, it isn't totally guaranteed that you can unlock the remaining two cores without problems, some SKU's may unlock, some other may not.
 

jvroig

Platinum Member
Nov 4, 2009
2,394
1
81
3.8 - 4.0 seems to be a common non-extreme max OC of C3 Denebs.

My X4 C3 gets to 3.813 @1.45V. If I disable 2 cores and run it as a dualie, it tops out at 4GHz @ 1.45V. In both cases, CPU-NB is at, or near, 2.7GHz (with CPU-NB adjusted as necessary)


1.365V is still pretty low. C3 chips are rated up to 1.4V officially, so up to 1.4V should be pretty harmless unless your temps are out of control (but an X2 should be pretty easy to cool). You can still increase vcore by quite a bit, and you may hit 4.0GHz with a little more increase. It took mine 1.45V though, but all chips are different.

A lot of people are more comfortable setting vcore up to 1.5 - 1.55V, but I'd rather go a little safer and top out at 1.45V. It doesn't matter for me anyway, as even going to 1.5V (just to test the OC) didn't give me any better frequencies, so I determined that 1.45V was the start of my chip's diminishing returns. 3.8GHz X4 C3 is not exactly a shining example, so my chip wasn't particularly blessed by the OC gods it seems :(
 

SithSolo1

Diamond Member
Mar 19, 2001
7,740
11
81
Just got home

So Prime has been running for about 4.5hrs now, no issues. CoreTemp peaked at 40c but its sitting at 35c atm. I do notice that the cores are computing the same iterations about 1-1.5 mins apart from each other. I could have sworn at the start they were in sync.


Unlocking would have been icing on the cake, I went in assuming it wouldn't unlock. Its a huge upgrade from my Opty 165. :D

As far as I can remember the chip is a CACAC AC 1011bpmw, I got it from MicroCenter
 

bryanW1995

Lifer
May 22, 2007
11,144
32
91
my q6600 did that on prime 95, allegedly it means you have a problem. try linx 0.6.4, ruby told me to try it recently, it typically finds errors much faster than prime 95.
 

jvroig

Platinum Member
Nov 4, 2009
2,394
1
81
Yeah, it also actually depends on what Prime test you use. Small FFT detects different errors from Large FFT, and different errors from Blend. Being prime stable for 4.5hrs in one of the tests is no guarantee you'll pass the others. But if you used Small FFT and only OCed by upping the CPU multiplier (and didn't mess around with FSB, RAM timings, and CPU-NB frequency), then you probably are ok.

Testing with a variety of programs is still the best way to go. OCCT Linpack, Prime95 (Small FFT + Large FFT), and Memtest HCI are the tools I most often use - 1-3 hours of each (just enough to take up a whole day, 12 hours max), and I call it good.