Is this normal for the vcore?

[jpetermann]

I have my 8IRXP and P4 1.8a o'clocked to fsb 133. I uped the vcore to 1.575. When I ran a sandra burn in test, I left the gigabyte utility manager open so I could see the temps and voltages and such. As Iwas watching, the vcore went from 1.58 before the test to 1.53 during the test and stayed there till the test was done. Is it normal to have the vcore drop like that? This is my first oclock system, so I am sorry if this is a basic question.

Thanks for your help,

PJ

 

[Nate420]

What power supply are you running that system with? What other components are in the system? Could be that your PSU can't live up to the demand.

 

[Jwyatt]

The voltage your cpu is set at is the max voltage it should see. The MB will regulate and vary the voltages depending on the load you have. (Im not making it up i read it somewhere on intels site) I think it has to do with heat really.

 

[jpetermann]

I have an Antec 400 watt supplu (the 412 model)

 

[SteelCityFan]

Kindof a related thing here...

I have the Antec 635 case which comes with their 350 W PS. My VCore is 1.44-1.46 according to the software Abit included with the motherboard.

 

[Jwyatt]

As long as its stable I wouldnt worry to much about it.

 

[jpetermann]

Right now I have it nice and stable (except for a cold boot problem. I am expecting a new 8IRXP this week, so hopefully that will take care of it). I am running at fsb 133 (1.8a @ 2.4) mem at 356. Seems very stable. My cpu temp hovers around 40-42c so I think it is doing pretty well.

PJ

 

[Nate420]

Well, with that PSU, I guess you can count that out as a problem. The reason I mentioned the PSU is because I'm having severe V drops with my Antec 300W cause of all the fans I have. When I run only one of my case fans it's not a problem, but with all 4 running I see the 12V rail drop alot, like down to 10.8V. Not sure if my Vcore drops are related or not, but I too see the Vcore drop while stressing the CPU.

Does anyone know weather the P4 draws from the 12V rail or not? I was thinking that it did, with the extra PSU connection to the mobo and all. I'm using a regular molex(asus ezplug or whatever), since my PSU doesn't have the newer ATX12V connector. I have a new Antec True550W on the way, so I hope it fixes my V drops.



<< The voltage your cpu is set at is the max voltage it should see. The MB will regulate and vary the voltages depending on the load you have. (Im not making it up i read it somewhere on intels site) I think it has to do with heat really. >>



If that's the case, the CPU should get fed more Vcore during high use rather than less. I know more V=more heat, but heat isn't an issue with the P4.

 

[ElusivePeon]

The vcore should drop when the cpu is under load.

 

[Nate420]

Well...OK, maybe I'm wrong. Just seems odd to me that you would want to reduce Vcore under load, a time when the CPU requires the most power.

 

[Jwyatt]

Seems bass ackward to me as well nate, oh well I guess the intel team knows more than us anyways

 

[Nate420]

Yeah Jwyatt, I'm sure they do. I'm curious to hear the logic behind it though, I'll have to try and find that info on Intel's site sometime.

 

[WarCon]

Could it simply be a side effect of current draw? And poorly designed voltage feedback circuit from the voltage controller chips? My 12v rail and my vcore droop under full load. Maybe they based their feedback loop partly on the 12v rail instead of solely on vcore out and the added droop causes the feedback circuit to allow droop in the vcore.

Maybe someone can ring out the circuit on one of these Asus boards (specifically the P4B266-C as thats mine......) and see how the feedback loop is setup. Might be some easy to do mod to stabilize that droop.

Just a thought.

 

[Nate420]

Well, I'll know in a few days if a giant power supply helps...Should get my new PSU Wednesday.

My Vcore doesn't drop much, but my 12V rail really takes a dive. High 10's is pretty sad for a 12V feed, not to mention running lower voltage means increased current, which is not a good thing. Not only is it hard on components to handle higher current, but it also creates extra heat.

 

[Nate420]

Well, I got my new PSU today, an Antec True550. With my old Antec 300W I was seeing the 12V rail drop into the upper 10's, which is the main reason I bought a new supply.

Prime95 has been running about an hour now and is still running as I type. Here's what it looks like now:

Voltages in () are idle readings taken just before I started Prime95 torture test.

Vcore = 1.55V (1.57)
3.3V = 3.25V (3.25)
+5V = 4.80V (4.80)
-5V = 4.93V (4.96)
+12V = 11.25V (11.31)
-12V = 11.75V (11.81)

I shut down all but one case fan while the torture test ran for an hour to get my case temps as high as I could in order to see the temp controlled PSU fans at work. At idle, with a case temp of 25C, MBM5 shows the PSU fan at 1470rpm. When the case temp reached it's peak of 35C (CPU was 49C), the PSU fan was running at 1680rpm.

It seems that maybe MBM5 is showing a falsely low 12V reading, BIOS shows 11.92V. Though the voltages look very similar to the old PSU, one thing that has changed is when I turn on/off the 3 case fans I have wired to a relay&switch, sometimes my PC would reboot with the old PSU (sudden V drop I guess), no reboots with the new PSU. This thing is super quiet too.