Opteron at 1.6v?

J

Justintoxicated

[H]F Junkie
Joined
Apr 10, 2002
Messages
14,522
So years ago I was afraid to stress my $300 processor. No1 was really sure what the "safe" voltage was.

is 1.6 safe? I'd like to bump my 165 up to 2800 mhz if it is safe. I'm thinking to stick with my DDR1 and 939 for a bit longer. Not sure there is a need to upgraded yet in all honesty, might pickup the next generation vid card when they are released though.
 
I have my Opty 175 running at 1.675v in order to break the 2.8Ghz mark

Just make sure you have good cooling.

Mine is de-capped and using a lapped Thermaltake Big Typhoon with a Big Typhoon VX fan on it.
 
v1.6 is getting up there, but I would do it as long as you have good cooling.
It will likely be obsolete before it dies.
 
I am running 2.9Ghz stable, and I can ALMOST hit 3ghz stable, but it usually gives and error in prime95 after 3 hours or so.....still, 2.9 at 1.45 isnt bad, 1.6 is high for an opty, but it should handle it alright so long as you can keep temps down...

you might look into water at that voltage
 
I already have water, when I tested it took a while to get my temps above like 42 if I remember correctly.

But what I read two years ago was that temp didn't matter, too much voltage might quickly kill the chip, even if you keep it cool.

Mine showed 48 hours stable in prime at 1.4v 2700mhz, but it would sometimes hard freez in WoW so I bumped it and the issue is gone.

Not planning to remove the IHS if that is what it takes.
 
I haven't looked deep into any SDS problems with K8 (more specifically 90nm), but I think the 1.55v limit is a bit overstated. Though I have never heard of a report of 15% causing problems (which would be 1.55v for 90nm Opty), even with Northwood which had an infamous SDS at 1.7v. More importantly, in my years of clocking, I have never had a chip fry at 15% or below, and I usually run my chips around 20% over stock. Even when the consensus was saying TBred limits at 1.85v, I had no problems for years at over 2v with my DLT3C, and my Palomino at 2.15v. Ran fairly warm, but were stable. Although you are right about voltage being more adept at frying chips than temps, I think a chip kept within reasonable voltages that is stable will out live its usefulness even at temps most would consider "high." I've heard the SOI process is more fragile regarding voltage, but the only major issue with voltage I can see is with high VDimm and low VCore (memory controller diode). I believe at high voltage, the quality of PSU and motherboard can have a tangible effect on life of components. Clean power is important, which is more difficult to maintain under high stress and heat for PSU's. Your motherboard may require a fan on its VRegs.

That said, my topless 0551 UPMW requires about 1.6v to run 2800 as well, and that is where it will run until it either craps out, or decent DDR3 memory costs around $100/GB. I also like this speed because it nearly maxes out my 2Gigs of BH-5 memory (255MHz 2-2-2-5).
 
PliotronX said:
I haven't looked deep into any SDS problems with K8 (more specifically 90nm), but I think the 1.55v limit is a bit overstated. Though I have never heard of a report of 15% causing problems (which would be 1.55v for 90nm Opty), even with Northwood which had an infamous SDS at 1.7v. More importantly, in my years of clocking, I have never had a chip fry at 15% or below, and I usually run my chips around 20% over stock. Even when the consensus was saying TBred limits at 1.85v, I had no problems for years at over 2v with my DLT3C, and my Palomino at 2.15v. Ran fairly warm, but were stable. Although you are right about voltage being more adept at frying chips than temps, I think a chip kept within reasonable voltages that is stable will out live its usefulness even at temps most would consider "high." I've heard the SOI process is more fragile regarding voltage, but the only major issue with voltage I can see is with high VDimm and low VCore (memory controller diode). I believe at high voltage, the quality of PSU and motherboard can have a tangible effect on life of components. Clean power is important, which is more difficult to maintain under high stress and heat for PSU's. Your motherboard may require a fan on its VRegs.

That said, my topless 0551 UPMW requires about 1.6v to run 2800 as well, and that is where it will run until it either craps out, or decent DDR3 memory costs around $100/GB. I also like this speed because it nearly maxes out my 2Gigs of BH-5 memory (255MHz 2-2-2-5).
Click to expand...

Holy crap thats a good memory OC, I never got my C2 memory above 200MHZ not even 205 was stable...Thats why I have the C3 G-skillz.

I have to be careful because I don't have the good stuff I read bookmarket but I remember that if I add too much voltage to one component and too little to another that one of the regulators can overheat on the MB but I think this was the xtreme guys that were doing phase etc... I think I'm safe at 1.6v then. 1.55v does not work for me because I'm still unstable at 2800mhz, in fact I need to try games at 1.6v as I only tested to see if it was prime stable.

Oh yea your sig is great, my 300a pwnz your C2D haha. My 300 is still stable at 464, i just retired it last year so now its in a box somewhere.
 
Hehe.. I have a couple sticks of Corsair Value Select DDR400... running at DDR566 :)
 
Id love to know how you got 323*9 stable on your Opty 165. I cant get really more than 9*301 stable for a 2709 oc on my 165 CCBBE model. Im running Gskill memory on a 4:3 ration I think (ddr333) and I got it at 245.5 with my 2709 mhz chip. Any ideas on that would be awesome.


I dont know if I got a 1.6 volt option on my Abit AT8 32x mobo. But I would think cooling it suitably wouldnt pose an issue.
 
You must log in or register to reply here.
Back
Top