Printable View
My buddy just finished overclocking his new rig. He has his P4 2.6 up to 3.3 ghz on air (SLK-800U). He has the voltage in the bios set to 1.625, but abit's hardware doctor only reads it as 1.55 volts. Which one is accurate? He is using an abit IC7-Max 3. Thanks for the help gents. :D
Anybody?
1st, if he has a crap psu the voltage the cpu gets will be lower.
traces in mobos can get v long with high resistance becasue they are thin, which can make the sensors on the board not accurate - though the cpu voltage also has to go thru lots of traces, regulators, caps, more traces so might be lowish aswell.
it *should* be within 0.05v of what its set at ...
Yeah I have a mobo that reports strange voltages when I mod it - comes with the o/cing territory - shame you can't stick a multimeter on a CPU really, lol.
although i dont own one, i've heard lots of reports that the IC7 Max3 undervolts like a bitch, so i would be more inclined to believe that it probably is running at ~1.55v
His powersupply is not the problem as he's using an antec 430 watt truepower
bump
I've got the Abit IC7 p4c board and i've found a-bits hardware doctor to be less than useless, its appaling buggy piece of software, i'd suggest your mate gets something else installed, say MBM5 and try that.
Hope that helps
TiG
I have an IC7, and as TiG said, that program sucks, but my IC7 also undervolts my CPU quite alot, even thought i have a 460watt enermax, also it reads the Cpu temp higher than it is :( Bit of a pain when your overclocking, have to be sure about it and go against the software lol
I would stick my neck out and say it is.Quote:
Originally posted by |SilentDeath|
it *should* be within 0.05v of what its set at ...
Its been documentated that the sensor is notoriously inaccurate, but the only way to really prove or disprove undervolting / fluctuations is to measure the said voltage via a voltmeter.
Having said that, CPUs are designed to work within certain voltage tolerances, (I seem to remember it was something between 3% to 5% but don't quote me on this) and anything exceeding these would cause instability, BSOD etc.
In the case of the OP's what he is observing would fall inside of that margin if we take the high side of the tolerance and outside of we take the low, and since nothing has been mentioned regarding experiencing stability issues I would hazzard that his Vcore was infact alot closer to 1.625.
I would guess that as soon as you start to overclock the CPU, the tolerances (mentioned above) will become more critical and if the voltages weren't 'on the button' so to speak, then stability would deifnately be an issue?
ok, cool. Another question, after running it o/c for about a week, he booted up windows the other day and got a boot disk error. The o/c was prime 95 stable. He had to reinstall windows XP 4 times before it would load correctly. He's using a raptor. Do you know why this would happen and is there a way to set a SATA hard drive as the boot drive with this board? Thanks
running stuff o/c will always have a higher risk of corruption - sounds like he was unlucky. saying that i vaguely seeing a thread on the abit boards recently about 'boot disk error' or something similar. Also, when installing any OS, make sure your PC isn't overclocked, he may of been getting that same corruption that caused it to break in the first place.
and yes, you can boot off the raptor - which controller is he plugged into?
Nox
Sata 1, which means that it's the intel controller I believe.
Is he running a Raptor or Raptors in RAID 0 ?Quote:
Originally posted by Spritzup
Sata 1, which means that it's the intel controller I believe.
I seem to remember issues with Raptor firmware on older versions and also driver and bios issues with the SATA controllers.
All known issues can be found on abit-usa forums.