I know man! It's temping! But, I'd still rather try and find that golden 170 ya know?Quote:
Originally Posted by itznfb
Printable View
I know man! It's temping! But, I'd still rather try and find that golden 170 ya know?Quote:
Originally Posted by itznfb
i wish you the best of luck man! and if i spot a good stepping you'll be the first to knowQuote:
Originally Posted by Reinvented
zoomx, if you really want dual core, then go for it, i'd confirm the prices with someone else though, as i'm not too familiar with single core value.
Sounds like a great deal to me. $130 cash is $130 cache. A 0610 should be good for at least 2.8-2.9GHz with enough voltage.Quote:
Originally Posted by z00mX
temperature is all. it also depends on weather your mobo is ok, any hot spots on that, ram, and this can kick errors out which look cpu related.
mine, i prefere the , if it runs, its stable aproach.
i can prive on core 1 for 24/7 if i wanted on passive water at 3ghz. and 3.15ghz on chilled, but core 0 craps out at 2.9 no matter what temp or voltage, it will allways crap out after so long,, some times 12 seconds, other times 2 hours.
but as there isnt a game or benchmark that uses the cpu to 100% for more than a few mins, why should it be 100% stable.
80% stable is usually good enough for months of usage without problems.
i guess ill do the trade :woot:
Well it looks like something is holding me at 2.9ghz, I will burn this sucker in after the move.
*EDIT* It was bios corruption that was holding me back. Repo's suggestion of trying to boot at 3Ghz did me in, LMAO. This CFX3200DR is a sensitive board and likes to get corrupted when pushed too hard. 2.9Ghz takes 1.52V so it looks like some burn in will be worth while.
I just run prime. If it can do 5 mins, I clock higher by 100mhz. With my CCBBE 0610. I started high at 2.8Ghz at stock voltage. It worked! So I did 2.9. 5 mins later I let it prime for 12hrs when I went to school.Quote:
Originally Posted by itznfb
:lol: Sorry man, doesn't work for everyone :lol:Quote:
Originally Posted by babalouj
lol, i've corrupted my bios trying to boot too high too soon many times.
nice!
I got my new build up and running on my desk now...and I got a RMA refund from Egg, so i'm gonna grab another processor for sure now...although stepping I am unsure of. ;)
IMO, and from what I've noticed as a pattern with prime, and maybe just it's behavior interacting with my 165 CCBBE 0610, if it primes past 2Hrs, it's rock stable. Anywhere between the first min to thirty mins is the critical pass. If your setup passes that, it is 3D stable, and could probably handle whatever you throw at it. Next pass is between 30mins to around 2Hrs. If it craps out just before 2hrs, it's more than likely stable as hell, but if it bugs you, bump up the voltage ever so slightly.
So far I've owned an Opty CCBBE 0610, and an Opty 170 CCBWE 0609FPAW(which I still have...any takers?) and during my experience priming with both, if ever the two prime for over two hours, THEY DON'T FAIL AFTER. I guess I don't wanna say it's a waste of time to keep priming after, but...it is. Pretty much everything else over two hours is lame bragging rights.
It seems that the most intensive program anyone runs here is rosetta. If a thirty minute primer setup can handle that, then a general, run o' the mill OC'er doesn't need to worry about anything else. If you're using your machine for very valuable calculations, or as server, or etc. then I can understand why being so "prime stable" is so important, that is if you are super paranoid. But who the F$#@ buys expensive components to OC so they can do these type of task? The answer: very wealthy enthusiasts that could really give a damn.
I agree with itznfb, and repoman---->anymore than two hours is a waste of time. Yet these are opinions, and it seems that some here are enlightened enough to know that whatever is in a readme must be the certain truth...
With my board, it has been a b#$ch to configure the DRAM settings to work well with my clocks, but the CCBBE has been very kind to me. Once you get the hang of it's taste in settings, you will never get a BSOD, or a lock up, or a random reboot. Ever since I reinstalled windows, and decided to take a slower route in OCing this badboy, I've been flying.Quote:
Originally Posted by itznfb
My most stable setting so far is 2.86Ghz@1.39V, with :banana::banana::banana::banana:ty ambients. I'm at 2.9 1.42V right now, but I think things might go better if I switch my machine over to my other room where it can stay nice and chilly by the AC.
I run prime while I'm out for 12 hours. I don't sit there and stare at the screen. I leave it unattended.Quote:
Originally Posted by phi|os
I've had it fail between 2 and 4 hours, and that isn't very stable for running games, etc. I certainly wouldn't run Rosetta on it after it fails. Usually it's just a temp issue. But it needed to be fixed. In fact, I just had prime fail on me. I came home today and saw that it stopped after 12 hours. I ran blend, so I will run it again tomorrow while I'm in class, with better ram cooling.
^
^
^
If my setup passes 30mins of prime, it runs smooth as butta. But it is most certainly a temp issue. The best to test is to prime and then run something very 3D intensive. Half Life 2 is a crappy test, because it doesn't heat up the GPU. FEAR is a good one, and should get your GFX card nice and toasty, so you can see how your processor reacts to other variables in your machine. I would assume high DRAM voltage could be a factor too. You can always just throw a fan in the case to blow down on the RAM and CPU.
for a golden 170, what you want is a CCB1E 0550VPMW.Quote:
Originally Posted by Reinvented
let see what time was my lost post 10:22 my time, its now 11:10, i just spent almost an hour because i couldn't reset my corrupted bios :slap:Quote:
Originally Posted by phi|os
i hate msi
wow, itznfb. your 0615 did better then the 0610? that worries me, lol.
Wow, that sux. Have you tried killbios on Tmod's bios cd?Quote:
Originally Posted by itznfb
Well 2.9Ghz @ 1.52V fails dual prime in 5 seconds (core 0). I guess I will just leave dual toasts on at that speed and voltage all night and see what happens. Is that the suggested way to burnin? Oh and please no negative comments on burn in, it is a well known phenomenon in the integrated circuits world, sensors actually go through this process as part of their calibration.
Can anyone tell me witch prime test to use to test stability? I belive small FFTS is the most stressing on the cpu, but I see there are some that runs Large FFTS. If I run large FFTs it could run for hours more than small FFTS.
Any good going 170 Steppings known since the 0550VPMW?
Run two instances of Prime, one small fft's, the other large fft's. Your choice if you want to give them affinity to each of your cores, lets you know which one is weaker when one craps out. Good luck!Quote:
Originally Posted by ekjr
Regarding what's stable or not, I can agree with both arguments, but from my experience (2 SC and 4 DCs) with DCs when prime fails is 99% due to heat. What does this mean? Well while you don't stress the CPU as much as prime, you can tell stable anything that fits your needs.
I use my comp for benching, gaming, web surfing, email, movie watching and music listening. I don't care if I get some corrupt download from time to time if I can run a DC at 3ghz instead 2.8ghz. Of course other's opinion is also 100% valid. Do we need a definition for stable? Well at dfi-street it's 8hr large FFTs, here we can diverge because it's XS, and XS stands for XtremeSystems not XtremeStable :D
So each one feels comfortable with its own definition. Case closed? ;)
P.D. to those that say here in XS are a lot of 3ghz chips, well maybe they're not 24hr prime stable, but I'm sure many of them are used 24hr at those speeds (like mine before I killed her :( )
8hrs of one instance? or two?Quote:
Originally Posted by mursaat