IP35 Pro Voltage Problem

[katank]

Hi everyone,

I have a Q6600 with an Abit IP35 Pro. Things were going fine until a few days ago when I noticed my temps were really high.

I checked the voltage with CPU-Z and found that the voltage reported was over 1.5V! I checked in the BIOS and it said 1.3125V which is the lowest it can go as it is the VID of the chip. As far as I can tell, there's no way of selecting a lower voltage. According to Abit's utility, the voltage is 1.26V. Thus, I think the mobo's voltage detection is wonky.

I'm fairly certain that CPU-Z is right as that number is backed up by SpeedFan and also the temps are waaay up.

After resetting the CMOS, things returned to normal. Yesterday, the voltages shot back up again. This time, a few CMOS resets did the trick. However, I have a feeling it might revert.

Anyone experienced similar things in the past and have an idea for a fix?

Thanks in advance.

 

[M1A]

I have not seen that with my Pro but heard some issues with this board none though that the voltages changed if set user defined. It will adjust with speed changes or speed stepping from 1 multiplier to the other. My E8400 changes when its at 6x and goes to 9x. Are you using user defined voltages or auto? Are you overclocking? Also that seems high for the VID of that chip. What are other Q6600 VID's usually? The E8xxx chips are usually a lot lower than that but then they are 45nm based.

 

[katank]

The voltages are user defined. I also turned off EIST and C1E. That should eliminate speedstep as a source of changing voltages, right?

The VID is a bit on the high side but not too bad for a Q6600. It's also a B3 if it matters. Everything is running at stock (I even turned the FSB from 272 down to 266).

 

[Heidfirst]

on an uGuru equipped mobo I would always take abitEQ ahead of CPU-Z.

 

[VirtualLarry]

I noticed some wierdness with the voltage on an IP35-E. Clearing CMOS and re-selecting the voltages fixed it. Also, updating the BIOS to BIOS 18 seemed to help too, much less vdroop.

 

[katank]

I see. I have BIOS 17 for IP35Pro which seems to be the equivalent of BIOS 18 for IP35-E (released around same time with similar patch notes).

I don't have a problem with vdroop. It's more vdrop. Furthermore, it seems negative.

I've also found that clearing CMOS seems to do the trick, at least temporarily. The most recent time, the voltages are now wandering all over the place according to CPU-Z, anywhere from 1.15V to 1.38V or so according to CPU-Z. Abit's utility thinks it's 1.26V throughout. At least it doesn't seem to be overheating much any more.