• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

CPU-Z Voltage Reading Error?

Joined
Oct 30, 2007
Messages
512
I have my CPU OCed for about a year now (see sig) and I just recently downloaded to the newest CPU-Z just so I can have the latest application. Now, there is a discrepancy between version 1.49 (left) and 1.46 (right). Which should I believe? The one on the left shows my cpu being overvolted according to Intel. I believe I should be fine, but I may reduce it down to a safer level <1.5V.




Occasionally, the numbers go to 1.52V and low to 1.488.

Under load, Vdrop is bad. Going from 1.440 to 1.4556
 
What did you set your voltage to in the BIOS?

Well, I had it on auto ever since b/c it was within range of 1.5V. So I left it at that. Now, I am going to set it manually within range.

Just wanted to know whether other people were surprised when moving to the newer version. I checked around with some other programs and some show the 1.5V range that I reported earlier while others report the lower voltage (one seen in the right). I dont know what to believe to I'm going to set it manually and stress test.
 
What you are experiencing is likely vDroop or lack of load line calibration.

To fix this, you can enable load line calibration, if you have that option, or do a pencil mod to fix the vDroop.

Also by changing the voltage to a specific number value instead of setting it on auto.
The auto option tends to overvolt the CPU more than it needs to be to become stable and jumps voltages around from idle and load.

That is what I know to resolve this issue.
 
Well, I had it on auto ever since b/c it was within range of 1.5V. So I left it at that. Now, I am going to set it manually within range.

Just wanted to know whether other people were surprised when moving to the newer version. I checked around with some other programs and some show the 1.5V range that I reported earlier while others report the lower voltage (one seen in the right). I dont know what to believe to I'm going to set it manually and stress test.

Unfortunately, since you set it to Auto, there really is no way to know what the real voltage is. Setting it manually is definitely a good idea.
 
with it at auto your board most likely changes voltages everytime it restarts.. i had a board do that when it was set to auto.. one time it would boot at 1.4v next time 1.45 then 1.4 again and would just keep going back and forth everytime it was restarted/booted up..


edit: just downloaded 1.49 seems to be running fine comparing it to 1.48
 
Notice the titles for the voltage readings, the software on the left is reading the VID, the software on the right is reading the actual supplied voltage to the core. Two completely different things. For an an understanding of VID all the Intel CPU data-sheets have fairly non-technical explanation of exactly what it is and are available on Intel.com. The datasheet also will have the Absolute Max Vcc (cpu core voltage) and an explaination of "load line leveling" (Vdrop) that explains why on most boards the voltage will be a bit lower than what you set in the bios. The fluctations you see as the load changes is called Vdroop and a google will provide more details but basically as the cpu works harder it draws more current which means the voltage drop (losses) in the cpu power supply components increases slightly robbing some of the set voltage from the CPU. It is not generally economical to build a cpu voltage regulation circuit that exactly compensates for all this kind of stuff so "real world" values are slightly off. Nothing to get excited about.
 
Notice the titles for the voltage readings, the software on the left is reading the VID, the software on the right is reading the actual supplied voltage to the core. Two completely different things. For an an understanding of VID all the Intel CPU data-sheets have fairly non-technical explanation of exactly what it is and are available on Intel.com. The datasheet also will have the Absolute Max Vcc (cpu core voltage) and an explaination of "load line leveling" (Vdrop) that explains why on most boards the voltage will be a bit lower than what you set in the bios. The fluctations you see as the load changes is called Vdroop and a google will provide more details but basically as the cpu works harder it draws more current which means the voltage drop (losses) in the cpu power supply components increases slightly robbing some of the set voltage from the CPU. It is not generally economical to build a cpu voltage regulation circuit that exactly compensates for all this kind of stuff so "real world" values are slightly off. Nothing to get excited about.

the picture actually has it switched around, right - VID, left - vcore, but the fact still remains that the one on the right seems to be mislabeled, my cpu-z v.1.46 has vcore listed, not VID, why does his have VID...? I have the same cpu (E6600)...

cpuz.jpg
 
yeah i noticed that too billparrish and wol-va-rine.. looked odd to me as well.. but i just couldnt place why it would be showing that for him.. maybe a issue with his motherboard bios and what cpu-z reads so its just automatically showing the cpu vid instead of the voltage because it cant see it..
 
haha, I cant tell left from right, I ment my other left - doh !

Interesting observation simonkey1985, I put it down to differences hard coded in the software and did not think about the possiblity the software might attempt to read both real and VID voltages and if unable to provide real voltages then it would default to VID. Its possible...

hmm got curious and took a look at the home web site for CPU-z and on the instruction page I saw this which might explain the difference depending on if an option was set.

http://www.cpuid.com/cpuz.php#history

Sensor - Set to OFF (or 0) disables sensor chip detection and voltages measurement.

Sooo using version 1.46 which was what I had laying around, I found that with sensor=OFF the software gives the VID and with sensor=ON it gives the actual core voltage on my board (and here is where I bet you are correct in your thought that if the software cannot determine/read the sensor it would defaut back to the VID. I have no way to check that atm but it seems very very likey to be true).

So I think you nailed it!

LOL as a last resort I RTFM FTW.
 
yay i finally guessed something right for once... woot for me.. only took 500+ posts.. maybe 500 more ill guess something else correct :-P
 
Back
Top