AMD 9600 Black OC effort

Astro
Astro
Joined: 18 Jan 05
Posts: 257
Credit: 1000560
RAC: 0

to keep 2550 stable, I had to

to keep 2550 stable, I had to bump the cpu V back to 1.25. I see where tomshardware ran it at both 1.25 and 1.28, soooooo I hope it's OK

FSB freq - 204
PCIE freq - 100
Processor freq mult - 12.5X
cpu-nb link speed - auto
cpu volt - 1.25
cpu-nb volt - auto
dram volt - 2.1

Astro
Astro
Joined: 18 Jan 05
Posts: 257
Credit: 1000560
RAC: 0

Before I go much farther, I

Before I go much farther, I started reading other OC efforts. The guy at OCforums got his first one to 2.7 at 1.4 v cpu, but couldn't get it stable so he RMA'd it, but couldn't even get the second one to 2.5. I got that beat...lol. The Tomhardware says they got 2.6. Hmmmmm

The highest CPU temp I can find is 47C, but the caps are back into the 50's. NB at 37C and sb at 53C

Note: my AMD64 X2 5200 is now doing this same wu lot and coming in at 38 min and change. The Quad (at this level) is taking 41:06, 41:10, and the like.

Astro
Astro
Joined: 18 Jan 05
Posts: 257
Credit: 1000560
RAC: 0

OK, I did it. 2.6Ghz I put

OK, I did it. 2.6Ghz
I put the FSB back to 200, raised the multi to 13, and bumped up the CPU volts to 1.275

FSB freq - 200
PCIE freq - 100
Processor freq mult - 13X
cpu-nb link speed - auto
cpu volt - 1.275
cpu-nb volt - auto
dram volt - 2.1

Strangely enough CPU temp isn't rising as fast as I'd figured, still can't find a spot over 45C. still barely warm to the touch

Bikeman (Heinz-Bernd Eggenstein)
Bikeman (Heinz-...
Moderator
Joined: 28 Aug 06
Posts: 3522
Credit: 686219689
RAC: 550782

RE: OK, I did it. 2.6Ghz I

Message 79806 in response to message 79805

Quote:

OK, I did it. 2.6Ghz
I put the FSB back to 200, raised the multi to 13, and bumped up the CPU volts to 1.275

FSB freq - 200
PCIE freq - 100
Processor freq mult - 13X
cpu-nb link speed - auto
cpu volt - 1.275
cpu-nb volt - auto
dram volt - 2.1

But will it complete an E@H task ? ;-)

CU
Bikeman

Astro
Astro
Joined: 18 Jan 05
Posts: 257
Credit: 1000560
RAC: 0

RE: But will it complete

Message 79807 in response to message 79806

Quote:

But will it complete an E@H task ? ;-)

CU
Bikeman


I can guarantee it will try (after the well at Simap dries up..next day or so, it'll be dry, then run down the cache).. Oh heck, just wait and see....LOL If smoke isn't released, that is. Running my einstein total up got pushed back to I could run Malaria for Dogbytes in February, then I saw where the regular work at Simap was ending and they were going back to the "once a month" release of work. So, I figured I'd better "get some" while I could. Actually they have relatively stable run times so they're perfect for testing.

Bikeman, look fast....your credit is 903,903

Bikeman (Heinz-Bernd Eggenstein)
Bikeman (Heinz-...
Moderator
Joined: 28 Aug 06
Posts: 3522
Credit: 686219689
RAC: 550782

RE: Bikeman, look

Message 79808 in response to message 79807

Quote:

Bikeman, look fast....your credit is 903,903

Cheers

Astro
Astro
Joined: 18 Jan 05
Posts: 257
Credit: 1000560
RAC: 0

The stable run times I was

The stable run times I was talking about for the last 3 wus were 40:38, 40:35, and 4025, so they're pretty constant.

Changed FSB to 204 for 2.652 Ghz

FSB freq - 204
PCIE freq - 100
Processor freq mult - 13X
cpu-nb link speed - auto
cpu volt - 1.275
cpu-nb volt - auto
dram volt - 2.1

I just have to assume I'm about to hit a wall.... I just wonder what wall, and what'll happen.

[edit] It froze, rebooted and upped CPU V to 1.2875 Linux is now "recovering journal"
FSB freq - 204
PCIE freq - 100
Processor freq mult - 13X
cpu-nb link speed - auto
cpu volt - 1.2875
cpu-nb volt - auto
dram volt - 2.1

Looked at results and haven't found an invalid yet.

Astro
Astro
Joined: 18 Jan 05
Posts: 257
Credit: 1000560
RAC: 0

I'm finally there. Equal to

I'm finally there. Equal to the highest clock I have found by any OC review so far on air cooling (still looking). It's a 17.39% OC.

The last batch of wus came in between 39:37 and 40:01 .

I had to run out for some water softener salt, that's why I was gone so long.

I'm at 2.7 Ghz.

FSB freq - 200
PCIE freq - 100
Processor freq mult - 13.5X
cpu-nb link speed - auto
cpu volt - 1.3125
cpu-nb volt - auto
dram volt - 2.1
Mem freq 533/1066
2T, 5,5,5,15
SB at 51C
NB at 34.1
Ram at 34.1
checked each cap, ranges from 44-59C
max measured cpu temp 53C (using fluke 65 IR thermometer, which I can point directly at the cpu and heatsink pad.)

Now to watch the temps closely. But by "Gosh" I'm there. The others were unable to get a stable clock at 2.7, but they were running Vista. Not sure if that matters???

[edit] oops, got a freeze, bump CPU Volts to 1.3125

Brian Silvers
Brian Silvers
Joined: 26 Aug 05
Posts: 772
Credit: 282700
RAC: 0

RE: I'm finally there.

Message 79811 in response to message 79810

Quote:

I'm finally there. Equal to the highest clock I have found by any OC review so far on air cooling (still looking).

[edit] oops, got a freeze, bump CPU Volts to 1.3125

So far what I see are people fixated on increasing the FSB. Why don't you try decreasing FSB to 193 and have a CPU Multi of 14 (193x14=2702)? If the FSB plays a part in the HTT clock, then at least for Athlon64 you can decrease HTT from the nominal 1000 (200x5) and there is still plenty of bandwidth... I admit that I have no idea about the NB linkage and if you'll then have to change that...or what effect FSB has on the memory... If it reduces the memclk like what it would do with K8, you could perhaps make up for some of the loss by going to slightly faster timings...

Astro
Astro
Joined: 18 Jan 05
Posts: 257
Credit: 1000560
RAC: 0

The lowest fsb option is 200.

The lowest fsb option is 200. I've read many reviews where it craps out over 210. the Phenom has seperate NB, SB, and HT settings, and isn't dependent on cpu clock any longer.

Hmm, Timing?? maybe that'll be next.

thanks
tony

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.