A64 3000+ Venice Overclocking Help? (DFI Ultra-D)

im quite frustrated as well. for some odd reason, my once stable 280HTT overclock has been giving errors like crazy. i had to drop down the HTT to 275 but at least now, i have my ram rock stable, 31 passes error-free @ 247mhz 2-2-2-6 3.3v.

im going to the amd tech tour coming up and im gonna pick up the 3200+ bundle and selling off the unneeded mb and os. just wanna see what the 3200+ venice can do in my board.
 
Hello everyone.

Don't know if you guys have read this thread in a while, but here is my update.

My 300x9 was never stable. Even at 1.71V. It would last about 10 hours in Prime95 then the entire computer would reboot. So i changed to 295x9 @ 1.69V and although Prime ran fine I just didn't like the "feel" I know this may sound insnae but I just swore that windows was running a little "slow" and things would pause for half a second. This could all have nothing to do with my overclocking, but the first time Wow crashed, I backed it off to 290x9. Perfection.

It's been rock stable. The ONLY concern that I have had is heat. Still @ 1.69V after 8 hours (yes 8) of constant world of warcraft my cpu was at 52C. Also, one time, I closed wow, and noticed that mbm5 said 60C, I panicked and wanted to unplug the computer but i sat there and watched and with nothing running it stayed @ 60 and didnt budge. I rebooted normally to the bio and it said 38C so i think the 60c was a fluke.

Anyway, 290x9 @ 1.69v was fine. I have been running it like that for almost a week now.

I decided to lower the Vcore as far as I could, to reduce the heat as far as a could (the cooler the better ya know) so I set it to 1.57V. That is quiet a large jump down from 1.69, but I'll tell you what it has been running Prime95 for 2 hours now!

Will report back...
 
now that you've found the top end of your cpu, please post in my revE oc database? (in sig, and stickied in the amd forum)

:D
 
I'm currently still looking for the highend for my CPU on air cooling....needs a fair bit more voltage, but 300x9 is Prime stable...just needs 1.72v. I'm actually seeing how far I can push it atm for enough stability to run SuperPi 1M for fun.....

I'll hopefully be putting my comp in the deepfreeze tonight or tomorrow night, which will simulate Peltier temps on the CPU, to help me decide if I wanna drop the cash on a 226W Pelt/h2O setup for my baby. I probly will, cus I'll be doing 2.9GHz for sure.
 
http://i20.photobucket.com/albums/b236/wiz486/21e17723.jpg

.

Its rather funny. but if you go back in my posting you can see that the o/c that I have above simply wouldnt happen at the voltage that I am crrently at. I just dropped it even lower and I am running Prime95 as i type.

2 weeks ago.... 1.7V to be stable
now................. 1.54 and falling

Is it possible that the "burn-in" myth might have some truth to it?
 
joker_927 said:
Is it possible that the "burn-in" myth might have some truth to it?

I keep telling people, yes.....it IS real.....that's exactly what happened to my chip awhile ago...and I'm still working at burning it in whenever I can, slowly stepping up the speeds....in the hopes that I can lower my voltages a touch more, this stepping sux on my Venice. :(
 
As I had mentioned before, I was getting quite frustrated as I could not get my machine prime stable at ANYTHING but stock (or minimally over stock) yet the ram was running fine through memtest.

Cornelius informed me of the new 615 bios this morning, so I've just flashed it and am going to give it a try. I'm going to be generous on the ram at leave it at 133 resulting in 192mhz @ 300HTT. At the very least I'd like to get this thing proven stable at 300 HTT (Which should not be an unreasonable expectation after this long) as see how it goes.

I'll keep all y'all updated.
 
Lukano said:
I'll keep all y'all updated.

Even though this is still your thread, I'll keep you updated too...I sold my bunk overclocking 0516EPKW 3000+.....stupid thing was so horrible 300x9 @ 1.72v is unacceptable. I'm getting my 0517APJW today and I have very high hopes...should be doing 2.7GHz - 2.8GHz on less then 1.6v with this puppy.

Cross your fingers for me, I'll see what I can do, cus I'll only have a few hours to play with it before I go to work tonight. :(
 
cornelious0_0 said:
Even though this is still your thread, I'll keep you updated too...I sold my bunk overclocking 0516EPKW 3000+.....stupid thing was so horrible 300x9 @ 1.72v is unacceptable. I'm getting my 0517APJW today and I have very high hopes...should be doing 2.7GHz - 2.8GHz on less then 1.6v with this puppy.

Cross your fingers for me, I'll see what I can do, cus I'll only have a few hours to play with it before I go to work tonight. :(

I don't mind in the least if people post all of their updates in here, or even if they derail the thread into oblivion. I'm confident that the step my step learning progress for me has helped at least a few, so if it can help more all the power to it.
 
Well the first thing I notice is I'm still having no luck at 300+. No matter the voltage (unless I'm actually giving it too much at 1.7-1.76 and causin electron migration) I still error on Prime within a minute or so.

Next step is to clock down the HTT a bit and see what it's like back around 290-297.

edit - I had actually given up for a while there, not in a permanent sense but in a "until something new like a new bios or techniques come along" sense. Even at 293x9 1.6-1.8v I was still not prime stable no matter how hard I tried.

I wish ClockGen would actually be able to set VID's from within windows. That would save me a step for rebooting.
 
Lukano said:
I don't mind in the least if people post all of their updates in here, or even if they derail the thread into oblivion. I'm confident that the step my step learning progress for me has helped at least a few, so if it can help more all the power to it.

Good call bro...that's how I am most of the time...after being here this long I kinda have to end up helping people indirectly a lot of the time....ah well. :p
 
Lukano said:
I wish ClockGen would actually be able to set VID's from within windows. That would save me a step for rebooting.

Yeah, it's kind of a tease having the VID options there and then not having them fully functional/working at all. :p
 
HOLY $HIT!!!

I love this new chip...took the IHS off the 0517APJW in about 60 seconds this time, it's been running less then 10 minutes and first thing I tried was 300x9 (2.7Ghz) at 1.5v in BIOS.....and it works. All I've tried so far is SuperPi 1M....but at the very least, it's 1M stable at 1.45v for 300x9.....gonna see if I can do 300x9 at 1.4v...got me a good one. :D :D :D
 
Well, all I'm running for now is SuperPi 1M...but man am I having a good time.

I am able to do 300x9 for 2.7GHz @ 1.36v.....and so far all I'm doing is raising the voltage and htt one bit at a time and I'm already at 317x9 for 2.85GHz @ 1.45v with the ram at 285MHz. This chip is fricken killer....if I can pull 2.9GHz on this Zalman 7700ALCU I'll flip. :D :cool:
 
cornelious0_0 said:
Well, all I'm running for now is SuperPi 1M...but man am I having a good time.

I am able to do 300x9 for 2.7GHz @ 1.36v.....and so far all I'm doing is raising the voltage and htt one bit at a time and I'm already at 317x9 for 2.85GHz @ 1.45v with the ram at 285MHz. This chip is fricken killer....if I can pull 2.9GHz on this Zalman 7700ALCU I'll flip. :D :cool:

Your new chip is a 3000+ as well isnt' it? Where'd you get it from, and how much? And how much did you sell your old chip for, and know anyone else who wants to buy one? I'd like to ditch mine (also appears to be a craptastic stepping) and replace it, but I'm a bit tapped out :p
 
Bought it at a local shop that only had 1 Venice left...

LBBLE 0517APJW
Cooling: Zalman 7700ALCU /w AS5

CPU Validation @ 1.600v

apjw1m4ib.jpg


I'm gonna start working on 32M now and see where I top out for that, which will also help to burn this thing in so that I can hopefully get some 1M runs done at 330x9 or 331x9...oh, and she does 300x9 at 1.36v. :cool:
 
Back
Top