Max voltage level for 9800gts

Default is 1.12V
I O/C it to 710/900/1800 and raised voltage to 1.16. Am I going to crash the system?

 

How much more of an OC did you get with that .04V? I can only OC my shaders to 1600MHz. 1650MHz causes artifacts. I remember someone saying that he raised the voltage to 1.20, saw a big increase in temps, but wasn't able to OC that much more.

To answer your question, increasing the voltage will shorten the life of the card, as will increasing the temperature, and if you raised the voltage you could fry the card but I think you are safe with .04V.

 

I overvolted to 1.25 and maybe even higher (BC2 times)... it made almost NO difference... I gained ~5mhz core speed at the cost of an overheated computer. It will most likely reduce your computer's lifespan.

 

Crazysoccerman what OCing software do use? My shaders round to the nearest 50MHz using the OCing from the Nvidia Control Panel.

 

I don't game anymore, but I do not remember the previous version of control panel rounding. AFAIK, there is no option to adjust clock speed in the latest control panel version without loading a profile made with a different program. I keep my shader at 1825, which I adjusted through the bios. The version of control panel I have is 3.3.532.01

 

I've done 1.23volts with no improvement in overclockability.

I actually run at 700/900/1800, all +/- 25 depending on how tolerant a game is to OCing at a stageringly low 0.81v

 

Hey guys. If you are using the voltage table editor of Nibitor to try and change the running voltage of the 9800m GTS, note that all you are doing is changing a label and it will do nothing to change the actual voltage set to the voltage regulator of the the GPU, hence you will find practically no difference in overclockability or temperature.

 

What is more. I noticed that if you have higher clocks and lower voltage you may suffer from screen glitches and artifacts.

 

Yeah kravis I think you are right. .81V would generate like half the heat of 1.12V (anyone know the formula?) which would be very apparent. Not to mention that that voltage for an OC is staggeringly low. However, how do you explain the "Definite solution to the nvlddmkm video crash on the P-7805u!"?

 

Whoa this is true, putting the gpu on Vid 01 or 0.9v makes it impossible to use even stock Extra clocks.

You wouldn't happen to know how to change the actual voltage set on the voltage regulator would you?

 

Yes indeed, I noticed the same behaviour when I tried this too.
I wish I did know how to change the actual voltage of the extra clocks (apart from using the 0.9v VID) but at the moment I don't know.

 

Hey UD, the formula to use would be P = C * F * V^2. Now if you can overclock to the same clocks then the power used would be very roughly:

0.81v: P = C * F * V^2 = C * F * 0.81^2 = C * F * 0.6561
1.12v: P = C * F * V^2 = C * F * 1.12^2 = C * F * 1.2544

Power ratio: (0.6561 * C * F) / (1.2544 * C * F) ~ 0.523 (assuming C & F constant)

So yeah, the power usage at 0.81v compared to 1.12v would be close to half and I'd imagine the heat dissipated too. Any way, it would be very noticeable.

Hmm, about the thread about the nvlddmkm video crash solution, it is possible that they changed the extra clocks to use the 0.9v VID but in that case they would most likely not be able to achieve the overclocks that they had using the 1.12v VID. The main thing is you won't be able to choose voltages that haven't been predefined by the existing VIDs (1.12v, 0.9v).