AAPL Stock: 117.81 ( -0.22 )

Printed from

Browser showdown shows OS X browsers catching up

updated 04:55 am EDT, Wed August 31, 2011

Browsers vary wildly on different platforms

Noted Windows review site Tom's Hardware recently reviewed the latest versions of some of leading web browsers running on Mac OS X and Windows 7 and found that, overall, Mac browsers -- particularly the one judged the best overall, Safari (v5.1) -- were catching up and in a few cases exceeding the Windows browsers, particularly with page load times, Flash, HTML5 and WebGL. Google's Chrome was judged the best overall for Windows and a stiff competitor on OS X as well.

Windows browsers overall were still judged to be somewhat faster than Mac browsers overall, but the differences that were once very pronounced are getting narrower. Windows programs emerged as faster in 19 of the 29 categories listed -- which covered many aspects from startup times to hardware acceleration stress tests -- but often with only minor differences.

The tests showed that the same browser on different platforms can in some cases perform dramatically differently. For example, using the Kraken test for Javascript, Safari was 70 percent slower on Windows than it was on the Mac. In another example, Firefox on Windows cleared the field by a huge margin on the Facebook JSGameBench test, whereas the Mac version was next to last in speed.

On the Windows side, Chrome and Firefox were judged to be very evenly matched, with only the latter's lackluster CSS performance holding it back. The new and more standards-compliant Internet Explorer 9 was determined to be best with Flash, HTML5 and memory management, while Safari on Windows racked up the lowest scores for that platform, trailing even the Opera browser. Apple has much more work to do with the program beyond relying on the WebKit engine, particularly in the areas of HTML5 hardware acceleration and WebGL support.

For Macs, the main race was between Safari and Chrome, with Opera again taking third place but with Firefox being judged the worst of the four on the Mac. Chrome was held back only by poor memory management, lack of HTML5 hardware acceleration and spotty reliability. The substantive differences in performances between browsers on different platforms means that Windows users cannot rely on their own experiences to recommend a possible Mac browser, and vice-versa.

The website, known for its expertise on Windows-based systems, had high praise for the overall Mac OS X 10.7 ("Lion") experience, but did note that the tests for Mac were conducted on a "Hackintosh" rather than genuine Apple hardware -- a factor that could, it said, actually improve the already impressive performance of the Mac in catching up to Windows browsing speed. [via Tom's Hardware]

by MacNN Staff



  1. Arne_Saknussemm

    Joined: Dec 1969


    PC users just don't get it.

    Apple iOS & OS X products are not better or worse because of their speed.

    Apple products are better because of their hardware integration, user experience, and of course, the "ecosystem"...

    Comment buried. Show
  1. TomSawyer

    Joined: Dec 1969



    A windows-centric shop tested Mac browsers on Hackintosh hardware and found them lacking. I'm stunned. How about ponying up for a Mac Mini and at least working with a true baseline. These results are kinda like doing all the benchmarking on a MacPro and using Parallels/Win7 to qualify the Windows specs.

  1. wrenchy

    Joined: Dec 1969


    Do people...

    still use Safari??

    Comment buried. Show
  1. _Rick_V_

    Joined: Dec 1969


    Yes, people still

    @ wrenchy: Yes, I know your comment was made in jest. However, for a long time I didn't really ever use Safari... until Lion when Safari introduced Webkit2 and sandboxing, etc.

    Other random observations:

    Truth is, while benchmarks are always fun and interesting; in regular day to day usage, they all seem about the same in terms of speed. There's not a single page (that I go to) that is so javascript-heavy that I can tell a difference between any of them.

    Safari on Windows is about the worst thing I've ever seen. As excellent as Safari is on Mac, the thing is glitchy and almost unusable on Windows. I wish Apple would polish it up some, so people can see how nice it can be. Otherwise, if that was my only impression of Mac software, I might never switch.

    On principle, I refuse to use Internet Explorer even on Windows. (first thing I do when rebuilding Windows is download and install Firefox). However, that said, I do recognize that MS has finally done a pretty excellent job with 9. Yet, the first chart above doesn't even place IE anywhere. For example, launching IE 9? a page is open the second you click the icon (yes, that's because Windows "pre-loads" IE so it's always running, etc.). But still it should be included.

    Comment buried. Show
  1. marcstreeter

    Joined: Dec 1969


    Yes it matters

    I would probably say that in some ways this is the best selling point Windows has - native driver support. That's why just about any component of home-brewed computer will run well (notice I didn't say configuration since some components just don't get along) and run the speed its supposed to run because every ability that *that* particular component has is supported(or recognized) under the driver being used. Fast forward to OS X and the drivers are pretty sparse, and when you do have a driver (for hardware not made by Apple) it is rarely of the same quality of its Windows counterpart. Heck its more of an afterthought when compared. Case in point : Logitech. Blech. Native support doesn't always mean you are getting the same thing. It just shows where the company's priorities are (usually expressed by the number of experienced programming personnel they are willing to put on payroll to support your platform). Apple is no saint here either (probably noticeable in this article's comparison of Windows and OS X performance for Safari even on a Hackintosh).

    Comment buried. Show
  1. marcstreeter

    Joined: Dec 1969


    Yes it matters(continued 1)

    Now fast forward *a lot* to the Hackintosh community that has sprung up around Apple's use of the intel architecture. You have less people working on their own pet projects in even less organization than those employed in the companies that produce substandard drivers that "work" already. Result : drivers that "work *sometimes*". This is not to say that their efforts are fruitless or useless. In truth, they are doing an amazing job of leaping over restrictions that Apple has purposely placed to prevent such a community. But there is a realization that many of us, me included, come to: native drivers on slower hardware will run circles around any PC hotrod that has been hacked to "work". I repeat, drivers that "work" to allow the use of a component on your particular machine's OS (in this case OS X) don't necessarily mean that they are using *all* of the capability of the component, just enough so that it fulfills the word "works".

    Comment buried. Show
  1. marcstreeter

    Joined: Dec 1969


    Yes it matters(continued 2)

    Most of the experience I have had with shoddy driver performance on Hackintosh's in general is the graphics card. The usual litmus test for graphics cards within the Hackintosh community is whether or not you get QE/CL working (usually visible if the ripple effect is present when you add a widget ). I have found that is not sufficient. For more info on why that is I invite you to read up on tonymacx86's forums. But if you have used a Hackintosh and used a (spec's wise) slower Macintosh you will be astonished to find that expected performance can be deceiving. Especially with web browsing. One example is when I "upgraded" from an iMac Core 2 (dual core) to a Hackintosh i5(quad core) you'd think I could reap the benefits of a drastically (spec's wise) faster computer. Not so. True *some* things are drastically faster. And some things are slower, browsing included. Why? My own lack of expertise in making a Hackintosh? Probably. The effect of being hacked in the first place. YES. One should not expect full performance in a hacked environment.

    Comment buried. Show
  1. marcstreeter

    Joined: Dec 1969


    Yes it matters(continued 3)

    Don't fool yourself and don't do tests on hacked hardware and publish the results like anything but an article for the Hackintosh community. If anything the title of this article should say "Browser showdown on *my* Hackintosh show that OS X browsers are catching up" and then at the end put the disclaimer that "Not all Hackintosh computers (even with the EXACT same configuration) are the same. Duh" instead of putting the most important piece of information at the end. Yes, the scope of your audience has been limited to just 1 computer, but at least you keep your credibility for the future.

    I thought I would just through in the definition of hack (from a computing perspective, source is ):
    Hack: 1) Originally, a quick job that produces what is needed, but not well.

  1. bdmarsh

    Joined: Dec 1969


    Performance hit on hackintosh

    There is a 3 to 10% performance hit from the patches necessary to run MacOS X on non-Mac hardware. (depends on how different the hardware is from a stock Mac)

Login Here

Not a member of the MacNN forums? Register now for free.


Network Headlines

Follow us on Facebook


Most Popular


Recent Reviews

Ultimate Ears Megaboom Bluetooth Speaker

Ultimate Ears (now owned by Logitech) has found great success in the marketplace with its "Boom" series of Bluetooth speakers, a mod ...

Kinivo URBN Premium Bluetooth Headphones

We love music, and we're willing to bet that you do, too. If you're like us, you probably spend a good portion of your time wearing ...

Jamstik+ MIDI Controller

For a long time the MIDI world has been dominated by keyboard-inspired controllers. Times are changing however, and we are slowly star ...


Most Commented