Lame X2-3800 and X2-4200 OCs

[Diogenes2]

Originally posted by: Rythan
Hmm... the True 430 is rated at 17A/17A on the 12V1/12V2 lines, while something like the OCZ PowerStream 600 is rated for 20A/18A, with nearly identical line and load regulation. Doesn't seem very significant. I will look around for something else to test with, though. I think I've got a nice server power supply at work I might borrow.

The PSU might not be the problem.. But here is some info I picked up.. ( here, but I don't remember the thread )


The true 480 has 1 12v rail @ 22A ..

I have heard the dual 12V ( no particular brand ), which don't usually have more than 20A on one rail ) are more problematic when pushing vcore past 1.5 on X2's...


I grabbed the Rosewill 550 from Newegg with 36A 12v ( stated, prob not rated ), and haven't had the reboot problem with 1.6 volts any more.. Still not Prime stable over 2.68, but I just get errors and not black screens ...

 

[Zebo]

Originally posted by: Diogenes2

Originally posted by: Rythan
Hmm... the True 430 is rated at 17A/17A on the 12V1/12V2 lines, while something like the OCZ PowerStream 600 is rated for 20A/18A, with nearly identical line and load regulation. Doesn't seem very significant. I will look around for something else to test with, though. I think I've got a nice server power supply at work I might borrow.

The PSU might not be the problem.. But here is some info I picked up.. ( here, but I don't remember the thread )


The true 480 has 1 12v rail @ 22A ..

I have heard the dual 12V ( no particular brand ), which don't usually have more than 20A on one rail ) are more problematic when pushing vcore past 1.5 on X2's...


I grabbed the Rosewill 550 from Newegg with 36A 12v ( stated, prob not rated ), and haven't had the reboot problem with 1.6 volts any more.. Still not Prime stable over 2.68, but I just get errors and not black screens ...

Rosewills buying a clue how good thier stuff is. I just looked at our Rosewill it's $68 now.. I got it for $45 from newegg when it first came out!.:thumbsup:

 

[Rythan]

I swapped the True380 out of the X2 4200 system for a single 12V rail Enermax 465 EPS system, with no change.. still can't hit 2.4 at 1.55V core, so it would seem to be either the MB or the CPU - perhaps OCing dual cores is just too much for whatever voltage regulators Epox used on the 9NPA+Ultra? I may have to look for a cheap DFI just for the sake of scientific inquiry

 

[Diogenes2]

Originally posted by: Zebo

Rosewills buying a clue how good thier stuff is. I just looked at our Rosewill it's $68 now.. I got it for $45 from newegg when it first came out!.:thumbsup:

Hey Zebo, even the $68 price tag made me hesitate, but when I saw it in your sig, I figured it was good enough for me ...

 

[mikester]

I would definitely try a better power supply before you give up. Beg borrow or - well, don't steal - a better one to test with just to make sure. I have a hard time believing there is THAT much of a difference between Manchester cores. I've got my 3800+ rock solid at 2.6 GHz at 1.55v vcore and my load temps are around 52C.

I tried to drop the vcore to 1.525v last night (to reduce the temps a bit) and one core failed Prime after 2 hours while the other one went all night. So you definitely need the volts to get it up there, and this also shows that you are at the mercy of the weakest core.

BTW - This is with a Seasonic S12-600w power supply. A good power supply makes all the difference - I had an Antec Smartpower 450w PSU on my previous machine and it couldn't overclock for crap. Just watching the voltages with a monitoring program and they were always fluctuating with the Antec. Same program with the Seasonic and the rails are rock solid, little to no fluctuation at all.

 

[Zebo]

Seasonic is the best PSU from things I've seen but that Enermax he just tried is bas ass too.. like 30 or 33 amps on 12..obviously it's not PSU at this point.He shoudl try what I said eariler.. underlcoking.. testing board and mem.

 

[mikester]

Oops - missed the post about trying the Enermax PSU.

Well, if you do try a DFI it will be interesting to see the results. I'm using a DFI NF4 Ultra-D for my system, but I almost bought that Epox that you have. Maybe I dodged a bullet. It's weird, because in AT's review the Epox overclocked as well as the DFI, but that wasn't with dual core, so maybe that makes a difference.

 

[sixth]

Well i just put my buddies X2 (toledo core in my epox board) hit 2650 prime stable, so its my chip, manchester core is either not a good OC'er or i just have a chip that doesnt wanna go over 2300 w/o insane power and insane cooling...o well..

 

[mikester]

Originally posted by: sixth
Well i just put my buddies X2 (toledo core in my epox board) hit 2650 prime stable, so its my chip, manchester core is either not a good OC'er or i just have a chip that doesnt wanna go over 2300 w/o insane power and insane cooling...o well..

Toledo's usually do overclock better, but not all manchester core's are bad OCers (mine is a manchester at 2600). It does sound more and more like you got a dud.

Can you RMA it?

 

[MDE]

The PSU on the second rig is NOT the problem. I have a similar system and my X2 3800+ is at 2.6GHz with no issues at all.

 

[biostud]

mine is a Toledo E6, the limit seems to be 2650Mhz @ 1.5V, I tried to run it at 2700 @ 1.55V, but no go. (memory and HTT below stock speed)

 

[videoclone]

I had the same problem and it ended up being the HT link in the middle of the mother board The heatsink was so hot it would burn your finger if you touch it . i put a fan over it and it was still so hot it would burn sooo.... ummm i gave up .... ...the end

I was able to get to 2.4 unstable and 2.2 rock solid but i am fine on 2.0 200FSB 5HT 200Mhz ram

Maybe later i'll play around with it more. But it's probably my 300w Generic micro atx power supply holding it back cant be helped i spose! ( Shakes fist )

 

[jEnus]

Started o/cing yesterday, I am able to take my X2 4200+ to 250 (10 x 250FSB) so far.

 

[ksherman]

I had some problems OCing my X2 3800 too... I could get it to around 2.4 Prime95 stable, but the system never quite ran right. The RAM was not the problem, as i can take it all the way to 245 before failing any Memtests... even a modest 200MHz overclock made the system a little wacko... went back to stock and evrything is happy!