How to bypass speed limit w/ K10stat when on battery

Discussion in 'Hewlett Packard' started by whazzup, Jul 5, 2009.

Thread Status:
Not open for further replies.
  1. whazzup

    whazzup Scribbler - Standard Member

    Likes Received:
    Trophy Points:
    Guide to running at full speed while on battery:

    On the tx2500, and probably on the tx2 as well, it's been noted by various users that the max clockspeed is halved when running on battery. This seems to be a bios setting, so it's not tweakable by normal means.

    However, I found out that, by utilising K10stat, one can actually get close to the max clock speed of the cpu even while on battery.

    This is the default setting:

    What i realised is, the FID doesn't really go higher than 12, so trying to overclock is useless, and quite a number of users have reported test scores lower than the original clock when benchmarking.

    The trick, thus, is to:
    1) change P1's FID to a number lower than '12', for example, '11'.
    2) change P1's CPU Voltage to a higher value, since we're pushing the clockspeed up. Just match it to P0's CPU Voltage for now.
    3) change P1's DID to '0' (from the default '1').
    4) Tada! Now P1's clock will be almost the same as P0. I've tested with SuperPI, and the results scale linearly with the increase in clocks.
    5) You can now undervolt your CPU Voltage if you want to.

    It'll look like this (FID is 5 instead of 11 in this case):

    And finally, to configure K10stat to start up together with the notebook, see here:, and search for 'Setting K10STAT for StartUp'.

    My thanks to all the guys here for experimenting with the various settings:

    Hope this works for all those who have been frustrated by the limiting of clocks when on battery! And juz to protect myself: PLEASE PROCEED AT YOUR OWN RISK! :D

    *juz wondering: is the limiting of clocks only applicable to the tx series? Or does it apply to all puma notebooks?
    Last edited by a moderator: May 18, 2015
Similar Threads - bypass speed limit
  1. DanielPalmieri
Thread Status:
Not open for further replies.

Share This Page