It’s A Bug In That Other OS, Not The Browser

It turns out that Google’s Chrome browser has been telling that other OS not to save power by napping, for years.“Instead of waking up the processor every 15.625ms, Chrome tells Windows to have it wake up every 1.000ms. So while your PC normally wakes up the processor 64 times per second when it’s idle, as long as you have Chrome running, the processor wakes up 1,000 times per second.
Chrome doesn’t have to be running in the foreground to have this effect, either. There’s only one platform timer, so when one application changes its resolution, the new value becomes a system-wide setting.”
This has been a huge drain on the batteries of notebooks.

Now, some claim this is a bug in Chrome, but it’s not. It’s a bug in that other OS that lets a user-space application mess with a system-wide setting. This is yet another example of a single-user OS designed in the 1980s still being fragile decades later. This is yet another example of M$ making an OS with too many vulnerable edges for malware to interfere with our use of the hardware we own. This is another example of what happens when you let salesmen design an OS.

I recommend Debian GNU/Linux. It’s a real OS that works for you, not some distant team of salesmen. In GNU/Linux, introduction of the “tickless” kernel and other such features combine to give serious reduction in notebooks’ drain. Intel wrote a whitepaper which shows many watts saved for notebooks and servers by tickless idle and several other measures. Of course, an application could set up interrupts to defeat that but it’s not a system-wide problem. The operating system responds automatically and does not take a new setting from one errant application.

See Chrome browser has been DRAINING PC batteries for YEARS.

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology and tagged , , , , , . Bookmark the permalink.

2 Responses to It’s A Bug In That Other OS, Not The Browser

  1. oiaohm says:

    thr this is a OS difference. OS X, Linux, BSD and most Unix systems the OS kernel is in fact charge of the tick. Applications on these posix OS are in fact only giving suggestions.

    Why is this the case. Application may require one wake up speed when active and another when the system is sleeping.

    thr now remember most of the developers working on Chromium are OS X and Linux users. Yes 2/3+ of google staff run either OS X or Linux and never touch Windows. Windows developers are a minority in the Chromium development team.

    Chrome is not the only application build cross platform and not being consistent with other OS conventions bring issues to Windows.

    The posix world include OS X found the same issue. Microsoft wrote a blog entry everyone else altered their kernels. Chrome developer may have thought Microsoft might have followed every else lead so there was nothing to fix.

    thr the intel document that robert is referencing refers to what becomes the tickless kernel for BSD, OS X, Linux…. everything bar windows. Windows having a tick every 15.625ms is in fact still draining battery. Chrome is just making existing issue worse. Now if Windows kernel was tickless and chrome was making it tick maybe your arguement would have legs.

    Remember its the implementation of tickless in the other OS kernels that split applications away from playing with the direct clock.

    Both Microsoft and Google have a possible bug here. Microsoft 100 percent has a bug and fixing it could prevent the google issue if it fixed the same way as other OS kernels have.

  2. thr says:

    “Devices are not the only area that require efforts for great energy efficiency. Application and service software can also have a big impact on power consumption. Take for example an application that increases the platform timer resolution using the timeBeginPeriod API. The platform timer tick resolution will be increased and the processor will not be able to efficiently use low power idle modes. We have observed a single application that keeps the timer resolution increased to 1ms can have up to a 10% impact on battery life on a typical notebook PC.”

    Get out! Does Google’s Chromium engineering team not read blogs? Apparently not. They’re too busy sabotaging a competitor.

Leave a Reply