Opty 165 Overclocking database

[greatscott68]

I just used A64info and tried your suggestion.So far so good.Priming now to test for instability.

 

Sweet! I would list my current RAM timings, but we have different RAM, so that wouldn't really help. Just use the RAM guide on our site for help with the choices to tweak + the fine tuning. I'm curently pulling 128,000 on AQM3, 10165 on '06, and am prime stable for 14 hours at 9x335. Keep bringing in those results - and good luck!

[Prospector]

Prime crapped out after 38 minutes so I went back to the other timings for now.From the OCDB entries that have the same RAM as mine my settings are almost the same.I'll look a little closer at them and see if I can tweak a few.

[greatscott68]

Prime crapped out after 38 minutes so I went back to the other timings for now.From the OCDB entries that have the same RAM as mine my settings are almost the same.I'll look a little closer at them and see if I can tweak a few.

Well that sucks, but oh well, it was worth trying. I'm very sure that you can improve your performance by tweaking even just a couple of those stock settings - so hang in there and keep experimenting

[elessar777]

I've backed off HTT a bit,to 323*9 Ran Prime last night for 7hours @ 325*9. One core was still going with no errors but the other core gave an error in Prime95 @ 7hours and 2 minutes.

this is most probably the CPU right? Not RAM?

 

Just wondering what the fix might be, whether to back off voltage a little or tu increase it?

 

Haven't found any definitive advice on what to do when you get prime errors after 6 hours or so.

 

any tips gents?

[elessar777]

something else I wanted to ask.

 

I have noticed quite a difference between the two cores - in terms of times in SuperPi - sometimes up to 5 secs per iteration, in Prime95 the time between iterations lags more and more the longer I run it for.

 

Is this normal?

[greatscott68]

I've backed off HTT a bit,to 323*9 Ran Prime last night for 7hours @ 325*9. One core was still going with no errors but the other core gave an error in Prime95 @ 7hours and 2 minutes.

this is most probably the CPU right? Not RAM?

 

Just wondering what the fix might be, whether to back off voltage a little or tu increase it?

 

Haven't found any definitive advice on what to do when you get prime errors after 6 hours or so.

 

any tips gents?

 

That's a great question. When I dual-prime at 9x340, core0 errors at about the 5 hour mark, while core1 just keeps going...and going. I've tried voltage up to 1.50, but nothing beyond that.

[elessar777]

Yeah I'm trying to up the voltage. Since I got my new shiny PSU I also installed an extra intake fan in the front of the case - scythe drive bay fan - takes up 3 drive bays and creates great through-flow of air over the CPU. Definately an improvement as core temps have decreased by a few degrees.

 

Currently sitting on 1.5 @ 328 . Not satisfied with backing off, I adjusted a few settings. Super PI was spitting out non convergency errors on one core within a few secs of running 32M test. After a bit of fiddling, managed to get it to run through without errors. Prime...bah....later.....lol

[elessar777]

No luck so far going higher than 328FSB. SuperPI passes 32M test dual instance (one per core) no worries but Prime95 gives errors within 5minutes. I'm now at a loss as what to do as I just don't know at this stage what settings I should change, memory timings, mem voltage, CPU voltage....???

 

Perhaps I've hit the limit for my CPU..shame....so close to the 3Ghz...yet so far eh :-(

 

Anyone with suggestions?

 

I also am currently trying out Merlins Bios (10/7/2006) and it seems to be a little more stable with lower memory timings and will actually allow me to set my MAL to 8, which I previously was unable to do as setting MAL any lower than 9 resulted in a no boot situation

[RoBoZeO]

just got my Opty 165 CCBBE 0615EPWM. upgraded from a 3500+ @ 2.6ghz

 

most of the info should be in my sig.

 

http://home.insight.rr.com/robozeo/untitled.PNG