Since the P8600 is a fairly common CPU on the XPS Studio 13/16 I thought this would be worth posting.
Stability was confirmed at each FID with Orthos for 60 minutes. I chose to disable SuperLFM and IDA. Under 100% load, at 2.4GHz the reported CPU temp dropped almost 12 degrees C. Obviously the difference was less at reduced speeds, but still worthwhile.
I do not notice any difference in the way the computer responds or performs, so as far as I am concerned this is free cooling. I was able to get 4hr 15min on the 9-cell battery with screen brightness @ 30% and low CPU demand (IE8 with wifi) so it probably modestly improves battery life too.
![]()
-
It's worth noting that if your CPU uses half multipliers your need to apply a reg hack otherwise you will lose 100MHz or so.
1. Run regedit
2. Navigate to HKEY_CURRENT_USER\Software\RightMark\RMClock
3. Find dword PStatesData parameter and increment the multipliers (circled in red) by 01 so for example on my P7350 CPU 06 becomes 07 and 07 becomes 08
-
For my 1340, the minimum voltage seems to be locked at 1.00V. Setting any value lower than that in rmclock is valid but will not allow the cpu to run at that voltage. Any ideas?
-
Did you try to use SETFSB, I heard that people managed to overlock their cpu up to 3 GHZ without any temperature rise!
-
I would love to try this undervolting but I am a bit scared to be honest. Can anyone provide a step by step instruction for the P8600? Has anyone also overclocked it? What consequences if any?
-
That's about as "step by step" as it gets. It's pretty much the same procedure for any CPU which can be undervolted. The long part is finding what voltages are good for yours(since it always varies per unit).
Oh and the purpose of undervolting usually is to reduce heat output. Overclocking would augment heat, effectively making undervolting pretty useless. Also note that the higher you clock the CPU, the more juice it needs and if you undervolted, it might cause the system to crash(when I overclocked my CPU, my computer crashed and i had to increase the juice running to it).
I wouldn't really consider doing both things at the same time to a same CPU. -
Wow you guys actually got RMClock running with a Montevina chipset with a Penryn processor? I remember having trouble even making it work on the XPS m1330 that I had for a couple of weeks! Didn't even bother trying on my Lenovo X200 with the P8600... I remember the good old days running this on my Inspiron 630m and undervolting my Dothan 1.86Ghz to 0.7 volts lol
They haven't updated RMClock since February 2008.. Too bad they didn't open-source it
You guys think I could run this on Windows 7 with my P8600? Any way to avoid the annoying UAC popups every time Windows boots up? -
how did u get ur p8600 to get to 0.8500 v? I'm running a p8600 too and the lowest it allows me to go is 0.8750v
-
Did anyone manage to actually overclock the P8600 using SET FSB?
Cheers -
- Voltage needed depends on the actual clock frequency. You can not compare voltages without defining at what clock frequency that voltage is used. -
I think he is talking about the drop down menu to select the voltage.
Like on my RMClock, .8750 is the lowest item on the menu. -
-
I tried it last night after reading this post. There is a tweak in the registry u can run.
But to summarize things so you don't waste ur time, it doesn't actually work.
The P8600 is hard capped at a min voltage of .8750 so even if you used the registry tweak to have RMClock "Show" the voltages, it will still only run at .8750 regardless of how low you go.
But if you want to try anyways and see for yourself:
RM Clock folder -> look for a reg file with tweak in the name
Run that reg file then open regedit (Start->Run->regedit)
Do a search for the word UnlockVid. When your comp finds that entry, open it and change the value 0 to a 1 (meaning true).
Restart RM Clock and it'll show the hidden voltages. But like I said, even if you go below .8750, say you chose .8500, ur min will still be at .8750 regardless. You can go under the CPU Info tab after you go below .8750 and see for yourself. It'll still stay at a min .8750. -
these are really pretty low volts, i tried .95 too for 9* multiplier, but shortly ended up with a BSOD. 1.0 volt was stable after 50 minutes testing.
-
it depends on the chip and its VID. These chips get binned with different vids thus why certain people can get lower and others can't.
P8600 undervolting results Dell XSPS16
Discussion in 'Dell XPS and Studio XPS' started by ekovalsky, Aug 15, 2009.