Opinion on Voltage

[Amrathe]

Okay, I want everbodies opinion on max safe Voltage for my Opteron 165. (The system named Order in my sig).

I can do 2.925Ghz with 1.425v and a load temp of 42C. (325x9)
When I try 333x9 I can't get more than 2 hours of Orthos before a failure even with 1.45v.

Should I push the voltage higher to try to hit 3Ghz on air? If so what would be the highest voltage everybody would feel comfortable with?

 

[inkfx]

I never owned a 165 but I think most people stopped around 1.55v. Watch your temps, though. With air cooling I'm not sure if you will be able to keep the temperature under control.

 

[Maviryk]

I'm running mine at 1.625 volts on water. Note, however, that I'm not worried about killing it, since it will be an excuse to upgrade to C2D.

I'm seeing temps of 47load for both cores on water.

I think you can go to about 1.5 on air safely. Just watch your temps.

 

[ziggo0]

Here is what I've always went by for A64.

Air = 1.55vcore max
Water = 1.65 ~ 1.7vcore max (pending load temp)
Phase = well...I don't use phase...but if you can afford it...why not go real hiiiigh

 

[hitbyaprkedcar7]

Here is what I've always went by for A64.

Air = 1.55vcore max
Water = 1.65 ~ 1.7vcore max (pending load temp)
Phase = well...I don't use phase...but if you can afford it...why not go real hiiiigh

my standards as well

 

[PerlAddict]

That chip deserves a good water cooling! I just pushed my chip up to 1.65v to try and prime at 330 x 9. Would love to see what that one in your rig could do! Lucky dog.

 

[Moto7451]

The max I've run though a 90nm core without feeling uneasy was 1.85v. Ran great for nearly 6 months before I gave the system to my brother. He put in his 3400+ Newcastle which he didn't overclock at all.

 

[Opteron 1100]

to be safe i would say 1.55 max

 

[QuietIce]

I'm too chicken to go over 1.55v ... yet ...

 

[PerlAddict]

I maxed out my DFI the other day on vcore trying to hit 3Ghz, and the CPU got up to 52C, but I went back my lower voltage and clock when the PWMIC alarm went off in MBM5 after it hit 71C. I'm gonna give it another try at max volts once I get a heatsink on that little IC.

Btw, am I missing something max voltage wise? My CPU volts only go up to 1.55 plus 13%, so ~ 1.75v. I haven't messed with this DFI SLI-DR board in a while, so I didn't know if there was a setting somewhere that enabled it to go higher, or if it was just a differences in the board I've got and the board you were running 1.85v in.

 

[AlabamaCajun]

I'm with the 1.5 crowd air or water. I may go a little higher (1.6) if buried in LN or I got the chip for under 50 bucks!

 

[SuperDave1685]

Nah... push it! This overclocker's forums ain't it? You can see in my sig that I've been pushing my 3200 Venice at 1.678v 24/7 since August... under air for most of that time too... I just upgraded to water after Christmas... Max load I saw with the stock HSF was ~ 47C while priming. I'm running it at 260 x 10 at 1.525v + 110%.... come on...Besides, I'll be a good excuse to upgrade!!

As a side note... I read somewhere a while back that the E3 cores actually are more stable at a slightly higher vcore due to the way the current leaks through the memory controller and the transistors switching on and off. If you run them too low I heard, you can potentially "wear out" the IMC because when its switching, theres not enough current to keep it operating properly and that can cause frequency disruptions within the chip, leading to a premature death... I read that about 8 months ago somewhere. I think it might have been on here or over at XS. Of course that could have been just hearsay, but ya never know... my chip is still going strong so...

 

[Socket eh?]

My advice would be to listen to the first hard wall your cpu puts up for a 24/7 setup. Mine did 2.6 on 1.4, but it took 1.525 to get to 2.87 and now 's it's dead after just about 5 months. Unless money isn't a huge deal to you, don't bother with alot more vcore for just 80-100MHz more.