Page 1 of 2 12 LastLast
Results 1 to 10 of 18

Thread: Mobile Users Beware: Linux Has Major Power Regression

  1. #1

    Mobile Users Beware: Linux Has Major Power Regression

    For those that follow my personal Twitter feed will know that for the past week I've been closely testing Ubuntu 11.04 and all Ubuntu releases going back to Ubuntu 8.04 on many mobile devices in the office. The overall system performance, power consumption, and boot performance have been the principal targets. However, late this week I discovered a glaring regression: Ubuntu 11.04 is viciously going through power. Compared to Ubuntu 10.10, the power consumption on Ubuntu 11.04 for mobile devices is up about 10% on average but under some workloads, I am seeing the power consumption up by nearly 30%. This is happening on many mobile systems spanning multiple generations of Intel CPUs and with Intel / ATI / NVIDIA graphics. This issue has been tracked down to a frightening kernel regression in the mainline tree that is still not addressed.
    Source

    I have been very skeptical about things coming from phoronix (Still waiting for native Steam support in Linux)... however, if this is true it is a bit of a concern, especially if Unity is targeting tablet devices and such as battery life is of paramount importance... (and on a personal note my aging laptop already doesn't have brilliant battery life, this will just make it so much worse )


    404
    404

    Come shoot the breeze with us on the Ubuntu Forums IRC channel - #ubuntuforums

  2. #2
    Join Date
    Jan 2008
    Beans
    4,324

    Re: Mobile Users Beware: Linux Has Major Power Regression

    I have been using 2.6.38 and if anything my battery life is better than ever. Even using 1000hz.

  3. #3

    Re: Mobile Users Beware: Linux Has Major Power Regression

    Quote Originally Posted by NightwishFan View Post
    I have been using 2.6.38 and if anything my battery life is better than ever. Even using 1000hz.
    Well clearly your delusional, but using Debian for an extended period of time will do that to you

    OR

    You are the exception that proves the rule


    404

    PS -
    404

    Come shoot the breeze with us on the Ubuntu Forums IRC channel - #ubuntuforums

  4. #4
    Join Date
    Jan 2008
    Beans
    4,324

    Re: Mobile Users Beware: Linux Has Major Power Regression

    There has been a lot of device power saving code since 2.6.32 I believe.

  5. #5

    Re: Mobile Users Beware: Linux Has Major Power Regression

    Quote Originally Posted by NightwishFan View Post
    There has been a lot of device power saving code since 2.6.32 I believe.
    I know... but an average regression of 10% is pretty bad since .38 ...
    404

    Come shoot the breeze with us on the Ubuntu Forums IRC channel - #ubuntuforums

  6. #6
    Join Date
    Jul 2007
    Beans
    13

    Re: Mobile Users Beware: Linux Has Major Power Regression

    Quote Originally Posted by not found View Post
    I have been very skeptical about things coming from phoronix (Still waiting for native Steam support in Linux)...
    That is wise. The methodology commonly seen at Phoronix makes it a somewhat interesting site to read, but not a reliable source of news. Stick with the more journalistic sites (such as arstechnica.com) for your tech news.

  7. #7

    Re: Mobile Users Beware: Linux Has Major Power Regression

    I tend to stay pretty close to kernel changes that alter battery life (because of Jupiter, and my other netbook related efforts), and I haven't seen this. It could be a device specific regression.
    [ Fuduntu.org ] - [ Fuduntu Forum ] - [ Fuduntu Blog ] - [ Fuduntu Wiki ]

    For a classic desktop with the latest apps, get Fuduntu!

  8. #8
    Join Date
    Jul 2007
    Location
    Reykjavík, Iceland
    Beans
    1,555
    Distro
    Ubuntu 10.04 Lucid Lynx

    Re: Mobile Users Beware: Linux Has Major Power Regression

    Actually even if there was a major problem with battery life, having many devices on it would make things better, as one of the makers of such a device might patch it. The more linux will be used the more it will be fixed.

  9. #9
    Join Date
    Feb 2009
    Beans
    50
    Distro
    Ubuntu 10.10 Maverick Meerkat

    Re: Mobile Users Beware: Linux Has Major Power Regression

    Quote Originally Posted by Bölvağur View Post
    Actually even if there was a major problem with battery life, having many devices on it would make things better, as one of the makers of such a device might patch it. The more linux will be used the more it will be fixed.
    I've never had a laptop battery that could hold a charge longer than 45 minutes after a few months' use, so I'm not sure I trust this Fauxronix benchmark. But even so.

    ???? How is a kernel regression ever a good thing? Unless something has changed with the actual hardware, it's a problem that didn't exist and now needs to be fixed thereby taking time away from other problems. Especially something crucial like power mgmt.

    IOW, if you're busying hacking the kernel to repair the issue, that's time you aren't running folding@home... that 3. GHz quad-core could be curing cancer... Or it could be doing empty quatraterabagoogle-flops while you stare at vi.

    And if you're burning through batteries, that's more toxic waste someone in Africa ends up burning to extract the gold and silver from.

    Like I said I don't trust the 'benchmark' here, but I already feel like I need to get a tiny SSD/Atom netbook to make the dedicated Facebook + youtube machine and keep all the bloated megahardware (seriously, the oldest laptop I have still running has almost two orders of magnitude more processing power than the one I used to dial the library catalog BBS in the '90s) running something folding@home or turned off and sitting in a refrigerator.

    Apologies for the moral outrage, after a month of "OMG GNOME Shell sucks" this seems a little more real than some ugly-*** css colors.
    Last edited by nathan28; April 23rd, 2011 at 05:13 PM.

  10. #10

    Re: Mobile Users Beware: Linux Has Major Power Regression

    Quote Originally Posted by nathan28 View Post
    I've never had a laptop battery that could hold a charge longer than 45 minutes after a few months' use, so I'm not sure I trust this Fauxronix benchmark. But even so.
    Interesting, I have a couple of machines some of which are over 3 years old that still hold 90% or more of their charge.

    My old 1000HE still lasts over 8 hours on a charge, and it's almost 3 itself.

    Quote Originally Posted by nathan28 View Post
    Apologies for the moral outrage, after a month of "OMG GNOME Shell sucks" this seems a little more real than some ugly-*** css colors.
    I still believe that it's only that one model that's impacted. I just tested 2.6.38.4 on my 1015PEM, and it drops to 7 watts on battery (dual core Atom N550) as expected.

    A fair study would have tested the effect across several models, all using fresh installations with no added configuration or software.
    [ Fuduntu.org ] - [ Fuduntu Forum ] - [ Fuduntu Blog ] - [ Fuduntu Wiki ]

    For a classic desktop with the latest apps, get Fuduntu!

Page 1 of 2 12 LastLast

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •