Announcement

Collapse
No announcement yet.

Ubuntu 10.04 Is More Power Hungry Than Windows 7

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • movieman
    replied
    I see the number of wakeups per second when idle seems to have dropped by about 1/3 on my laptop with the latest updates; I don't know whether that's a coincidence or whether they've fixed something in the kernel.

    I'll have to upgrade the netbook and see if the power usage drops there.

    Leave a comment:


  • movieman
    replied
    Originally posted by ioannis View Post
    although an old discussion now, may I add that there is a bug in Ubuntu 10.04 kernel that prevents the CPU from staying in lower states. People report half the battery life compared to the previous Ubuntu release.
    Interesting: I noticed at the weekend that my netbook seems to have about 2/3 of the battery life with 10.04 that it used to have with 9.10.

    Leave a comment:


  • ioannis
    replied
    although an old discussion now, may I add that there is a bug in Ubuntu 10.04 kernel that prevents the CPU from staying in lower states. People report half the battery life compared to the previous Ubuntu release.

    powertop reports many wakes per second (quantity depending on system) in "[kernel scheduler] Load balancing tick" task, rising with little load, on many kinds of multi-core (?) systems (original report was on a Core 2 Duo processor (T6500) with a single core enabled (multicore disabled in BIOS)). Cause of the problem: With kernel 2.6.32, there came a patch to the scheduler that introduced this problem (that was backported to some other versions as well). Even though this problem occurred fi...

    Leave a comment:


  • Yfrwlf
    replied
    Originally posted by ua=42 View Post
    Ubuntu by default disables power saving for Intel wireless 5100 cards. Turning the powersaving changes my battery life from 1 hour to 3 hours on my netbook.
    That's a pretty huge difference.

    So, this is the point: while comparing stock OSes with all the services/daemons running that they will, is the power savings difference REALLY in the drivers, or is it in the configuration? If you do tweak the settings and it does make them equal power-savings-wise, how will they perform? In other words, tweaking one to just be basically clocked really low would be cheating, since it would perform far worse when running programs.

    Doing fair comparisons is very hard work, but regardless doing a "default OS" comparison is fair on that level. I would like to know about where ATI is at as far as power savings over Nvidia cards. AMD has often had better power savings and cooler chips, but again unless you measure it against performance too it's not really fair to try to say who is ultimately better at that point.

    Leave a comment:


  • ua=42
    replied
    Ubuntu by default disables power saving for Intel wireless 5100 cards. Turning the powersaving changes my battery life from 1 hour to 3 hours on my netbook.

    Leave a comment:


  • yotambien
    replied
    Originally posted by kraftman
    Good to hear this. Those were only examples, but I've got your point. I don't have any objections to what you wrote later and I very appreciate you bothered to do so. I usually have objections to the article titles and sometimes to some parts of the article itself. I also believe there are better ways to do the benchmarks, but as far as the defaults are tested it's understandable, but can be misleading.
    Excellent ; )

    The issue with the wording of the articles and their titles depends on personal sensibilities. I don't have a problem with it partly because I'm convinced that Phoronix is a friendly site towards Linux, Open Solaris and so on, so I never think it's trying to undermine their image when pointing out faults or comparing these systems or distributions to, say, Windows or Mac. It all depends on what you think the intention is. I know some of us here are constantly scratching the itches about this or about that, but the same thinking applies really.

    As for the benchmarks, absolutely, there will always be better ways to handle everything. But I'm convinced this is a two-way road, meaning that Michael and company surely improve and learn from past mistakes and comments in the forum. I don't think anybody sees the benchmarks as coming from above and set to stone--the criticism in the forum proves it, as well as the calls for suggestions about how to improve the test suite (which you can even do yourself if you want to). Actually, if I got it right, the idea is to bring as much as possible exposure to the automatic testing system to the projects that could be interested, with as much as possible involvement from their developers, who could point out how and what to run or why the numbers are what they are. How cool would it be if Arjan, Garrett or whoever popped in and said "look, guys, this is rubbish, you have to take into account x, y, z", as the people from the OSS drivers do?

    Leave a comment:


  • Xheyther
    replied
    Just in case Michael isn't already aware of that, tom's hardware France (formerly presence-pc) had published a new from Phoronix yesterday : http://www.presence-pc.com/actualite...mmation-39263/

    Leave a comment:


  • kraftman
    replied
    powertop reports many wakes per second (quantity depending on system) in "[kernel scheduler] Load balancing tick" task, rising with little load, on many kinds of multi-core (?) systems (original report was on a Core 2 Duo processor (T6500) with a single core enabled (multicore disabled in BIOS)). Cause of the problem: With kernel 2.6.32, there came a patch to the scheduler that introduced this problem (that was backported to some other versions as well). Even though this problem occurred fi...


    Reported on 2010-02-19

    I always considered the release schedule deadline is stupid. Now, you're paying, Ubuntu

    Leave a comment:


  • caramerdo
    replied
    These number don't make any sense. How can a T61 consume that much energy out of the box? And the netbook consumes almost as much?

    I don't think you can rely on that power meter you are using.

    Leave a comment:


  • kraftman
    replied
    Originally posted by fewt View Post
    If 10.04 uses that much power, then it is VERY broken or misconfigured out of the box.
    I just upgraded the pm-utils package in Kubuntu, so who knows?

    Leave a comment:

Working...
X