[Kal-EL]

Thank you Neuromance, its much appreciated.

[hellcamino]

I actually have never seen a software voltage measurement be accurate on any of the systems I have built but you believe what you like! As for instability...maye you are right and I don't know what I am doing here? That must be why I game at 4.3ghz daily as well as anything else I do on my PC as it's triple tested stable there ?

For the record my first pc was a TRS-80 from Radio Shack as well, I don't remember what brand of tape deck I used for it's hard drive though.


The reason I keep on about psu voltage readings is due to the fact that motherboards don't have sensors for it, measure actual live voltage against OCCT if you like with an actual meter.



Dig those volt readings! If those readings were even close to the truth this pc wouldn't even turn on much less overclock or do antything...

[Neuromancer]

heh heh

HC I was not doubting your ability nor competence in the OC arts, quite the opposite really, kind of why I was so annoyed (although really it was the losing power thing that just flipped my switches)

When I first posted up in this thread I was not even considering PSU sw voltage measurements, although I have seen numerous time its dmm'ed pretty accurately.

I have seen software PSU voltages like yours, reading completely wrong. This is usually caused by inconsistencies between the computation formulas and the motherboard. Everest probably reads the 12v rail much more accurately I take it? What does your BIOS report? I have not seen a BIOS report off like that in years, so its just a matter of finding the right software. When you run the latest hardware software also needs a chance to catch up, if the manufacturer does not send them a prerelease sample, than the software author needs to get the board on his own to find the proper measurements to read from. Everest is from a large corporation and offers paid versions of their software so their accuracy is probably going to be a lot higher.

Others (without a firm grasp of Celsius and Fahrenheit) have no problem quoting Coretemp readings as their "idle/load" temps. Though they are notorious for being 10-15C off "real" temps. Yet I see coretemp listed in useful tools.


Couple of Q's...

Does it read this far off when running stock?

I am getting a DMM soon I hope, going to use it to test vcore and such though. Would be surprised if it read that incorrectly, since I use the BIOS software to adjust voltages...

With multi rail PSUs you measure from the 24pin connector or the CPU connector for accurate 12v readings?




To anyone else reading this thread.


The point I am trying to make is this, if you use software to monitor your temps, you can use software to monitor your voltages. Find the one that is most accurate of course (different software uses different formulas for calculating these figures, for instance, Coretemp does not read a temp sensor it reads delta to tjmax IIRC, which is why it became so unreliable when Intel went 45nm (Intel was very hush hush about what the tjmax is) whether this has changed or not, I am not sure. They had to love all the people posting their coretemp screenies with these ridiculously low idle temps)

The idea of using any temp or voltage sensor is not to report the actual data, but to have a quick reference guide for you to notice changes that occur. As I mentioned previously about 3d clock mode on my card. Obviously if your 12v reading software is reading 12v, and all of sudden is reading 11.2v you should bust out a multimeter and recreate the situation. This is indicative of a problem. If however like HC is saying it is reading 2v, then it is completely useless to you, much like AOD temps sensors are useless to me on my CPU when I unlock the third and fourth cores.

HWmonitor however reads much closer to my "real" temps.

And sorry about the typing. I am on my aunts laptop with some busted keys. Should be home in another day or so though.

[hellcamino]

The voltages read all over the place with various monitors on every board I've used I think but on this one in particular the PSU volts read no differently at stock speeds. Multimeters are the only way to go when putting your system under maximum strain (benchmarking with high volts) when accuracy counts the most, I even use an IR thermometer to check all my component temps while stability testing or benching with untested settings.

Last time I checked there weren't any monitors that were anywhere near accurate for this board or any of the others in my house currently or previously.

[Neuromancer]

Damn shame that.

I will get a mulit soon and post upresults of my current collection with the next couple of weeks

[Chuchnit]

Frank, two things. First can you please resize your pics for those who don't have 26 inch monitors , and second could it have something to do with the fact that you run vista 64?

[hellcamino]

No it has nothing to do with os, I have similar screenshots dating back for several years on multiple pc's and platforms using XP32, XP64, vista 32, vista 64 and now Windows 7 64 bit.

[dinos22]

nice list of tools
i use SetLOD a bit these days

also there is a new tool on XS for GPU voltage increase called GPU voltage tuner which is pretty nice to use

[Chuchnit]

Originally Posted by dinos22

nice list of tools
i use SetLOD a bit these days

also there is a new tool on XS for GPU voltage increase called GPU voltage tuner which is pretty nice to use

Thanks Dinos. I actually found setLOD after you mentioned it in your thread during the last LLC. Nice app. I'll add that there. Can you link me to the voltage tuner?

[dinos22]

http://www.xtremesystems.org/forums/...d.php?t=227628