Skip to main content

Futuremark Unveils Web Browser Benchmark

It's not often you see people benchmarking browsers. Futuremark wants to change that with the public beta release of Peacekeeper.

Futuremark's aim with this benchmark is to put an end to the so-called "browser wars" by removing subjective opinions or analysis of performance with too much room for variability such as loading times.

Typically, a newly released browser will be subdued to the well known Acid3 web standards test but there are few universally used benchmarks which measure the performance of common web page functions. Peacekeeper is a purely JavaScript-driven benchmark and therefore only tests JavaScript functions. It does not test actual web page loading times and is therefore not affected by connection speeds and the variation that can occur there, nor does it test things like Flash.

In order to have more basis on real world performance, each of the tests are weighted differently depending on how heavily they are used on commonly accessed websites such as YouTube and Facebook. A few tests, such as those using arrays, are not based on these website profiles.

Of course, these websites do not represent all website designs, and are built with heavy load in mind and optimized to work on most computers. Performance differences between browsers as determined by Peacekeeper may not reflect the performance you see on the websites you visit.

Peacekeeper also has a simplified Online Result Browser which shows average browser scores for commonly used processors.

  • curnel_D
    This app also takes into account the machine running the browser, which means that it will never be taken seriously as a browser benchmarking standard. Just another bragging right for those running an i7.
  • Greatwalrus
    I think the main point of this is not to compare browser scores between other systems, but to compare browser scores with your own system. With that, you can find which browser runs the best on your system.
  • Edex
    It's pretty relative...
  • Greatwalrus
    For me, Firefox 3.07 scored 197, Firefox 3.1b3 scored 282 and Midori scored 480.
  • rooket
    I don't have a problem at all with web browser speed, what I have a problem with is -> spyware & malware. I'm not sure if this is alleviated in MSIE7 but I have my doubts. I just use firefox but I read somewhere that it has its own security issues. I haven't ever got a trojan from firefox. MSIE6 is a malware magnet and as such it seems like a microsoft engineered trojan horse.

    I would imagine that if you want the fastest browser you'd be trying out chrome or some other stripped down browser. MSIE is fast enough, firefox feels no different.
  • @greatWalrus pardon my ignorance, but isn't Midori a drink not a web browser? If it is a browser where can I find it? Sounds interesting...
  • IzzyCraft
    I would never judge browers just on javascipt speed it's rarely just that how well a brower handles the web.

  • deepcover
    Just tried this and got:

    Internet Explorer 8.0 224
    Firefox 3.0.7 292
    Chrome 675
    Safari 4.0 968

    These benchmarks seem to be much more comprehensive than other browser benchmarks I have tried.

  • lucuis
    I also tried this out for kicks.

    IE7 220
    Firefox3.0.7 617
    Safari4.0BETA 1571 (Didn't really seem any faster then firefox the short time i used it)

  • IzzyCraft
    LucuisI also tried this out for kicks.IE7 220Firefox3.0.7 617Safari4.0BETA 1571 (Didn't really seem any faster then firefox the short time i used it)MehWell script speed is fast enough now that other factors just make about all browers relatively the same for when they render a webpage for you. Really the limiting factor is that you still have to download the webpage so who cares if new FF is a ba-gillion times faster your internet isn't any speeder so you get more or less the same result rending css etc.