Safari retains speed crown over newcomer Chrome in OS X

Posted:
in General Discussion edited January 2014
While besting Firefox by nearly 50 percent, Google's newly released Chrome browser was 12 percent slower than Apple's Safari in benchmark tests.



In benchmark tests run by Computerworld, Apple's Safari browser claimed the top spot over Firefox, Opera, and Chrome. Google released a beta build of Chrome to the public on Tuesday.



The results showed that Safari slightly edged out Chrome, was nearly twice as fast as Firefox, and over ten times faster than Opera.



Computerworld's method of testing used the SunSpider JavaScript benchmark suite. The test was run three times for each browser in OS X 10.6 and the numbers were averaged.



According to November numbers by Net Applications, Internet Explorer commands 63.62 percent of the total browser market, followed by Firefox with 24.72, Safari with 4.36 percent, 3.93 percent for Chrome, and 2.31 percent for Opera.



"As you might expect, the speed of Google Chrome for Mac is something we're very proud of. If you have a Mac, try installing the beta and see how fast it launches - there's hardly even time for the icon in the dock to bounce!" the Chrome development team said on its official blog.



Google's Chrome Mac beta can be downloaded from Google's website, and weighs in at 17.6MB.The program requires Mac OS X 10.5 (Leopard) or later and only works on Intel-based Macs.



Graph courtesy of Computerworld
«134

Comments

  • Reply 1 of 69
    Oops! Back to the drawing board!
  • Reply 2 of 69
    quadra 610quadra 610 Posts: 6,757member
    Still very impressive for a Beta.



    I tested Safari, the latest WebKit nightly, and Chrome on Sunspider as well.



    Safari beat Chrome easily. Wasn't all that close. And the latest WebKit nightly blew both right out of the water.



    Which is also testament to how well Apple is doing in this area. Which also means that if your browser isn't running WebKit you're doing it wrong.
  • Reply 3 of 69
    vvidvvid Posts: 9member
    May not be faster, but doesn't hog memory like Safari does.
  • Reply 4 of 69
    Quote:
    Originally Posted by vvid View Post


    May not be faster, but doesn't hog memory like Safari does.



    Memory usage != slowness, when you consider that a lot of stuff can be and IS cached for quick access. It's only worth complaining when a program doesn't USE the RAM you have to make the program faster.



    Unused RAM is wasted RAM.
  • Reply 5 of 69
    Quote:
    Originally Posted by Quadra 610 View Post


    Still very impressive for a Beta.



    I tested Safari, the latest WebKit nightly, and Chrome on Sunspider as well.



    Safari beat Chrome easily. Wasn't all that close. And the latest WebKit nightly blew both right out of the water.



    Which is also testament to how well Apple is doing in this area. Which also means that if your browser isn't running WebKit you're doing it wrong.



    I ran the same tests, but also added a nightly build of Chromium to the mix. And as you said, Webkit does blow the other out of the water.



    \t

    Chromet 542.4ms +/- 4.4%

    Chromium:549.2ms +/- 3.6%

    Safari\t 537.6ms +/- 1.2%

    Webkit: 450.8ms +/- 0.5%
  • Reply 6 of 69
    quadra 610quadra 610 Posts: 6,757member
    What's funny is that NetNewsWire's built-in browser is about equal to Chrome (or extremely close) on the Sunspider tests. I'm not even sure how current NNW's engine is, either. Interesting . . .



    I use mostly NNW for posting in forums. it's a wonderful app.
  • Reply 7 of 69
    djrumpydjrumpy Posts: 1,116member
    Quote:
    Originally Posted by Sgt Zeppelin View Post


    I ran the same tests, but also added a nightly build of Chromium to the mix. And as you said, Webkit does blow the other out of the water.



    Chromet 542.4ms +/- 4.4%

    Chromium:549.2ms +/- 3.6%

    Safari\t 537.6ms +/- 1.2%

    Webkit: 450.8ms +/- 0.5%



    Still no adblock and extensions are still disabled in Chrome, and although enabled in Chromium, they don't seem to work at all in the latest build (or at least the Adblock extension doesn't). I find all the ads far to distracting to use. Get the extensions working, and I'll consider it.



    As to the speed tests, simply testing Java speed isn't all that interesting to me. In addition, as to the numbers above, a tenth of a second isn't all that significant to the end user. Web pages are not 100% java. I find that Safari, Firefox, and Chrome all perform well enough that any would work from a speed perspective. I do like the light weight of Chrome though. Something to watch.
  • Reply 8 of 69
    leafyleafy Posts: 34member
    But, to look at other areas where people "feel" fast, I think Chrome has a clear edge over Safari.



    For example, startup time is extremely fast with Chrome. Also, switching among tabs is almost instantaneous. Safari tab suffers when one of the tabs is loading a complex page/javascript. I don't mind a 12% slow-down on page rendering itself, UI overall responsiveness is more important IMO.



    My 2 cents
  • Reply 9 of 69
    djrumpydjrumpy Posts: 1,116member
    Quote:
    Originally Posted by leafy View Post


    But, to look at other areas where people "feel" fast, I think Chrome has a clear edge over Safari.



    For example, startup time is extremely fast with Chrome. Also, switching among tabs is almost instantaneous. Safari tab suffers when one of the tabs is loading a complex page/javascript. I don't mind a 12% slow-down on page rendering itself, UI overall responsiveness is more important IMO.



    My 2 cents



    You forget. In OS X, you only 'start' the browser once. Unless you manually go into the menus and close it, it will open instantaneously after that as it stays running with you 'close' the window. I don't see quick startup time as much of a benefit. I agree on the tab switching as that is something that is perceptually faster for the user.
  • Reply 10 of 69
    bdkennedy1bdkennedy1 Posts: 1,459member
    Remember when Safari was released in 2003 and Opera cried? Now their up to tendy ten.
  • Reply 11 of 69
    While not exactly misleading, the report does cross reference both Mac and Windows platforms. The market share clearly includes Windows in the mix as IE (which is not even available on the Mac) is at 64% and Safari at 4%. Yet the speed tests exclude IE as a comparison. All 5 browsers are available on Windows and I'm sure the speed results would change somewhat if tested there.



    A more balanced report would only include market share on the Mac platform exclusively so long as the speed tests are relative to the Mac testing.
  • Reply 12 of 69
    Despite the fact that I'm sure the benchmark is technically correct. Apart from starting up new pages Chrome is really fast bringing back Google requests which is about half the new pages we get. So if in page changes are not as quick i tend to mentally gloss over them..Emotionally its clearly snappier than Safari. Snappy seems to equate to fast in the real world.
  • Reply 13 of 69
    solipsismsolipsism Posts: 25,726member
    Just to be clear, SunSpider is testing WebKit?s Nitro v. Google?s V8 JavaScript engines not the WebKit browser engine themselves.





    Quote:
    Originally Posted by Quadra 610 View Post


    Still very impressive for a Beta.



    If you run Safari on SL in 32-bit mode you get much slower results on SunSpider. When Chrome moves to 64-bit this may also help their V8 JS engine fly past Nitro.





    Quote:
    Originally Posted by Quadra 610 View Post


    What's funny is that NetNewsWire's built-in browser is about equal to Chrome (or extremely close) on the Sunspider tests. I'm not even sure how current NNW's engine is, either. Interesting . . .



    I am under the impression that NNW uses the WebKit.framework for its browser, which means that it?s the same as your Safari?s browser engines.





    Quote:
    Originally Posted by DJRumpy View Post


    Still no adblock and extensions are still disabled in Chrome, and although enabled in Chromium, they don't seem to work at all in the latest build (or at least the Adblock extension doesn't). I find all the ads far to distracting to use. Get the extensions working, and I'll consider it.



    As to the speed tests, simply testing Java speed isn't all that interesting to me. In addition, as to the numbers above, a tenth of a second isn't all that significant to the end user. Web pages are not 100% java. I find that Safari, Firefox, and Chrome all perform well enough that any would work from a speed perspective. I do like the light weight of Chrome though. Something to watch.



    In the end it?s all about convenience, which to me means interoperability. Safari is my preference on OS X and IE8 on Windows. However, with IE I have installed Chrome as the browser engines and made it load this by default by editing a DLL.
  • Reply 14 of 69
    I'm not so sure about this....in my experience, Chrome is a LOT faster than Safari - and it doesn't create a huge memory leak a la Safari. It's my main browser now - I've been using it since the Developer release.
  • Reply 15 of 69
    BOO!! Down with Apple.
  • Reply 16 of 69
    djrumpydjrumpy Posts: 1,116member
    Quote:
    Originally Posted by nondual View Post


    I'm not so sure about this....in my experience, Chrome is a LOT faster than Safari - and it doesn't create a huge memory leak a la Safari. It's my main browser now - I've been using it since the Developer release.



    Memory leak? You do realize it's not called a memory leak if the software purposely uses the memory, and then gives it back when closed?
  • Reply 17 of 69
    solipsismsolipsism Posts: 25,726member
    edit: Pipped by DJRumpy
  • Reply 18 of 69
    Well, I tried Chrome on the Mac out - (I use it regularly when I'm on a PC) and I am so glad it fixed that annoying clicking on a link and opening to a tab instead of a new window as Safari always does.



    However, one thing I wish Safari and Chrome did was block annoying flash ads and allow custom pop-up blocking instead of all or nothing.



    I love how fast Safari is and Chrome is darn good in my eyes too.



    I also gave the new Opera try and I must say I am so impressed by it that I am considering using it as my primary browser because it has all the features I like such as the built-in ad flash blocking that I wanted so much. I also like how Opera does bookmarks more similar to IE which is about the only thing I ever liked about IE. Opera seems to have so much more in comparison to many of the other browsers out there.



    Now to each their own, i'm not advocating or trying to convert anyone to anything.
  • Reply 19 of 69
    Quote:
    Originally Posted by DJRumpy View Post


    Memory leak? You do realize it's not called a memory leak if the software purposely uses the memory, and then gives it back when closed?



    I'm using iFreeMem and Safari uses EVERYTHING it can. Yeah, it gives it back when it's closed, but it hogs everything while it's working. Chrome does not do this. iFreeMem shows huge chunks of 'free' memory while using Chrome the entire time I keep it up. I'm constantly reclaiming memory back when I'm using Safari, which makes multi-tasking a real chore. I like to leave my browser up to pages and leave and come back to some of my faves instead of closing and reloading everything again and again.



    Also, when I close Safari lately, I get a message from Safari that says something about a 'world leak'.
  • Reply 20 of 69
    djrumpydjrumpy Posts: 1,116member
    Quote:
    Originally Posted by nondual View Post


    I'm using iFreeMem and Safari uses EVERYTHING it can. Yeah, it gives it back when it's closed, but it hogs everything while it's working. Chrome does not do this.



    That still doesn't make it a memory leak. I'm running safari right now with 4 tabs open. It's using 205 MB.



    Hardly "everything".
Sign In or Register to comment.