This is a great program! It completely allows the most from the efficiency of the P-M.

I have a 1.8 Dothan in my I9200.
Minimal voltage already was .700 and the max was set to 1.34 V. (I assume the program chnages these maximum/minimum voltages according to the CPU model. That would be the safest way to keep folks from cooking them.)

I now have my max set to 1.068 V. I'm still testing with Prime (and playing NFSU at the same time ). It failed at the step lower.

[(EDIT...) Posted too soon. I've had to go up one more step to 1.084V.)

I run SpeedSwitch XP as well as I8kFanGui. It was really something to watch my temperatures go from 62* C with both fans at high speed (while running the stress test) down to it's current 49* C with both fans on low speed under stress.

I can now confidently reset my fan switching thermal thresholds so that it may be possible that the fans are hardly ever on slow speed, and never at all needing to be on full speed.

This is really great!


I do have concerns about freezing, etcetra, in the future. However, I never once had the screen freeze up even for a second while playing NFSU with Prime running! (Game settings at fullest detail, default Omega settings for Radeon 9700).

I look forward to more testemonials from other Dothan users.

can you run rmclock, speedswitch and I8kfangui at the same time?

INSPIRON 8600c here.

1.116v MAX
.700v MIN.


Just a question, does everyone set .700v for their MIN? I just went all the way to .700v and never looked back.

I am seeing very nice results. With regular browsing, I've seen a 6* drop. CPU (see below) now hovers around 39*C with an Evercool NP-101

Under Prime95's "torture test", which one should I be using?
I only have 256MB RAM. After reading the help text for Prime95, I realize that it could be the RAM itself causing me to crap out.

I thought last night I was nearly on track with 1.086V as the high, but when I started stressing again this morning, it puked.
I was using the second choice. Now I'm testing on the first choice.

Use the first choice. That stresses the FPU the most (which is what you want to do).Since you're not messing with the RAM voltage, they don't need to be tested.

It seems I no longer will have a need for SpeedSwitch XP, correct?
RightMark seems to be overriding SpeedSwitch's allowances for CPU throttle maximum.
The other thing I'm seeing, and not too happy with, is that SpeedSwitch is NOT giving an accurate account of the CPU load, and I suspect not the speed, either. I can understand the speed not being the same as the RightMark because RightMark is showing it in realtime, or close to it, whereas SpeedSwitch is calculating it (or asking Windows- depending on which you select) at one second intervals; the lowest time frame it will set to.
But the load can remain fairly constant through the course of two or more seconds, depending on apps, and SpeedSwitch is NOT telling me the correct load, according to RightMark and cpuz.exe.

I would like to know what you guys are using as a utility to show CPU load and speed in the taskbar that loads @ startup, and is calcing as close to real time as possible.

Thanks

Yes.

I disabled speedswitchxp when I loaded rmclock since I thought they might conflict. What is the benefit of using both? should I use both?

There really isn't any need for SpeedSwitch when you're running RMClock.

Hm, my 2GHz Pentium started at a max of 1.340, could only get it down to 1.148V for a stable run. Minimal .700V no problem. Also looks like undervolting had no effect at all on my SuperPi results, so right, no performance changes.

Is there a results thread from those who have tried this? How much does undervolting affect battery life, in general?

I'm keen to give it a shot, and I probably will, but I'd like to see some numbers as well.

My feeling regarding battery life increase from any mod is that you'll have to test this for yourself. I have yet to read about two people with identical computers getting identical battery life running identical apps.
There are just too many variables involved.
Same thing with CPU Temps.

You can always change it back.
If this were overclocking, I'd have been too much a scardy-cat to try it. But I don't think this can hurt anything at all.

I've gained about 20 minutes of battery life on average.

Same for me, 1.148v is the lowest I can go. I bluescreen hard at 1.1v and am unstable at the next level up. 1.132v might work, but I went one more step up just to be sure.

I get around 20 minutes more battery life as well. More importantly, my cpu temp maxes out around 69-71C playing WoW compared to 77-79C prior to undervolting.

What is interesting is to fire up mobilemeter, switch to battery and watch the discharge rate change when you really work the system. My lappy discharges around 18-20w on battery when doing normal stuff. Fire up prime95 and that leaps to 38w. Very interesting to watch.