Thinkpads Forum

I have looked at the undervolting guide and so far have gotten my max multiplier down .2volts completely stable... orthos 10hrs (and dropped about 17-19*C). This puts my highest multiplier at 1.1v and I am still moving further down with it in .025 increments to see if I can gain even more. Rightmark's lowest voltage selection for my system is 1.0125. Is there any way to get rightmark to go lower? (I can't lower my SuperLFM voltage anymore as it's already at 1.0125 and feel as if it would still be stable at a bit lower voltages and thus saving more battery life.)

Also I didn't see anything in the guide online, can rightmark conflict with other power settings on my laptop? If so, how should I set the battery manager and IBM's power manager to get the most out of rightmark?

Thanks ahead of time. I am enjoying tweaking this guy!

[edit]: Almost forgot, it's a T7300 cpu.

You didn't say but I take it you have a T61 as that's an 800MHz FSB CPU.

I haven't used Rightmark, instead I use Notebook Hardware Control. It allows me to set vcore for each 'speedstep', I don't know if Rightmark does this.

Anyway, I have a T60 with a T7400, a 667FSB part and the vcores for each step are as below:

6 x 0.9500V
Same for 7x, 8x, 9x and 10x. 0.9500V is as low as NHC will set
11 x 0.9875V
12 x 1.0373V
13 x 1.1000V

Voltages confirmed with the latest release of CPU-Z.
All stress-tested and stable with Prime95 ver. 25.5 which fully tests both cores in a few different ways. The next lowest step on the higher multipliers gives errors. This is as low as my particular CPU will go stably. If I could go lower than 0.95V then I'm sure my CPU could handle it fine at the lower speeds. However I read somewhere that the 0.95V minimum thing was an Intel 'ruling' and part of their chipset design?

I'm an ex-desktop hard-core overclocker who's not adverse to spending hours over a few days to get the best from a CPU.