Safari performance issues

Safari performance issues

Summary: Safari, Apple's Web browser, has always been considered a lean and powerful and is the default browser of most Mac users. But recent performance questions are putting Safari under the microscope.

SHARE:
TOPICS: Browser, Hardware
31

Safari IconSafari, Apple's Web browser, has always been considered a lean and powerful and is the default browser of most Mac users. But recent performance questions are putting Safari under the microscope.

Macenstein recently ran some performance tests of the venerable Apple browser and the results aren't exactly flattering. According to the site using Safari can slow your system down as much as 76% vs Firefox.

It seems that it's not the load time of Web pages that's the issue, its that Safari is using too many processor cycles, even when it's idle.

I noticed this quite by accident the other day while rendering out an After Effects animation I had done. I had made a spelling change in one of the graphics, and re-rendered a composition I had rendered the day before. When it was finished, I noticed in the Render Queue that it had taken 15 minutes longer to render than it had the day before.

So I began to think why this would be. I had not restarted the computer since the day before, nothing had changed hardware wise. The only thing different was that I had been surfing the web a bit while the render was going on that day, where the day before I had not. “Surely surfing the web on a multi-processor machine shouldn’t add 15 minutes to a render”, I thought. Well, yes it does actually, if you’re using Safari.

CNet's Mac blog has some feedback from around the blogosphere on the "Sluggish Safari" issue including a response from Surfin' Safari's David Hyatt, one of the engineers responsible for creating and updating Safari:

1) Browsers rarely get served the same content, even on very popular sites. Without spoofing it's hard to know if Safari is being served some buggy content from one of the pages in question.

2) A small sample set isn't enough to draw general conclusions. Try a bunch of other different Web sites and see if a slowdown still occurs. If so, then maybe there is a systemic problem. Until then, though, all we know is that something is hogging CPU in one of five Web pages.

3) Reduce reduce reduce! Reduce the problem if possible. Cut it down to one page. Don't go back/forward (just go right to the pages instead of clicking through to them).

I haven't noticed the issue myself because I'm a devoted Flock user (a FireFox derivative) but I'm curious if you have experienced it. Sound off in the TalkBack below.

[poll id=39]

Topics: Browser, Hardware

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

31 comments
Log in or register to join the discussion
  • If it is performance you want, Opera is the best choice

    [url=http://www.opera.com]
    Scrat
  • Yep, Opera works great on both the Mac and PC (NT)

    No text (NT)
    iPad-awan
  • Flock is best in my book!

    However I don't know that this story is true, Safari has never given me any problems whatsoever.
    Reverend MacFellow
  • Ever stop to think it's not CPU?

    But oh, I dunno, the HDD? Could very likely be I/O related. In fact I'd bet it is
    because CPU wise neither FF nor Safari are that bad, and yes it HUGELY depends
    on the particular CSS, JavaScript, and Flash being loaded. Safari definitely tends to
    use quite a bit of RAM though and that could also cause your rendering app to get
    swapped out.

    In any event, repeatable testing is hard since FF and S are totally different and are
    generally being fed different pages on todays dynamic sites. Come up with a
    simple set of static HTML test cases that they behave differently on, then re-run
    your rendering test. That would be a little more definitive. Better still, run FF or S
    under a profiler such as Shark or Sample and then you'll have real numbers to
    report instead of heresay.
    mloftis@...
    • oops wrong word

      *hearsay. (Yup, grown too close to my spell checker, and it didn't help me a bit this
      time)
      mloftis@...
  • Safari seems to have a massive memory leak

    When I use Safari it keeps ballooning in memory usages until it exceeds 400mb and around that point it stops working and slows down my whole system until I can force quit it. That's why I've switched to Firefox for most of my browsing, even though in many ways I like Safari better. I do have to say the extensions available to Firefox are quite amazing though, so I don't expect to be switching back...
    philip@...
  • My experience is that any

    Apple software I've used, uses more system resources than any comparable program, at least on Windows. <br>
    I've noticed this with Quicktime and iTunes esp. and both of which use a tremendous amount of RAM. Even their respective services use an over abundant amount of RAM when you've not even started either program. Not sure about processor cycles but something to look at I guess.
    xuniL_z
  • If you are worried about rendering time, you should shut off all

    unecessary applications. I run a bunch of memory hog applications from Adobe
    and Quark, sometimes 8 or 10 at a time. OSX works better with multitasking than
    Windows, but occasionally I find apps start running more slowly. I simply shut
    down unnecessary apps or occasionally restart to clear everything. I've never
    noticed a problem with Safari in particular. Acrobat and Word seem to be pretty
    hoggish.

    I read recently that Safari had a 4.7 percent browser market share. I use Safari and
    Firefox, but sometimes I use Camino and have tried others. Safari and Firefox
    seem to be the best all around. The worst browser I've ever used is Internet
    Explorer. I don't like it on PC but IE for Mac is pathetic!
    MacGeek2121
  • About that IE 5% ...

    IE has not been developed for the Mac in roughly two and a half years. It never got to
    version [b]6[/b], never mind the current version 7. It is missing half a dozen fairly
    significant features and is as buggy on OS9 and OSX as on Windows. Yet ... 5% claim
    to be using it? Those would be the Windows users, yes?

    DLMeyer - the Voice of [url=http://glhorton.podomatic.com][b]G.L.Horton's Stage
    Page[/b][/url]
    davidmeyer@...
    • Slight correction

      ---IE has not been developed for the Mac in roughly two and a half years.---
      http://en.wikipedia.org/wiki/Internet_Explorer_for_Mac

      Development of IE for the Mac stopped in June 2003.

      The last real "new" version of IE for the Mac came out in March 2000, with only minor tweaks since then until development stopped altogether.
      tic swayback
      • Silly

        Wha? 5% use IE?
        Unless of course they are an ignorant pre-Panther user. But.. then youd have to be silly anyway to not upgrade.
        :)
        slayr-cxf50@...
        • Yep IE is STILL in use on the Mac

          I might ask the same question, but I was surprised to find that 80% of the teachers in the all Mac school district that I now work for still use IE as their primary browser. None of these people knew that MS was no longer supporting IE on the Mac 4 months ago when I came to the district. They have all been "programmed" to click on the "Blue E" to get to the Internet. I am currently having to pry the "E" off of their machines and teach them that there a number of better alternatives...
          tigoree
  • 7 minutes in an Apple store (I wish I could spend more)

    It happened two weeks ago at the END OF THE DAY.

    First I checked one of the weakest Mac spots ? Internet surfing.

    I took two MacBooks Pro 2.33GHz Intel Core 2 Duo, 2 GB memory, ATI Mobility Radeon X1600 graphics with 256MB SDRAM:

    One of them loaded a [b]12 KB page (Internet) about 1 min[/b] (I checked it 3 times); another one ? with a second delay. I guess it?s because Mac OS X still cannot handle the memory properly ? the more memory-hungry applications were used ? the slower Mac even if those apps are not opened (but I did not have time to investigate this case).

    Second, I came to Mac Pro Quad-core Intel Xeon 3GHz, ATI Radeon X1900 X 512 MB with 4GB memory and 30? Apple Cinema Display:

    The loading time of Apple?s Logic Express on was noticeably slower than it was on a MacBooks Pro. I did not count how long did it take because I did not expect it. I guess it?s because of 30? Apple Cinema Display (again - I did not have time to investigate).

    By the way, according to Mac magazines - 24" iMac is faster than Mac Pro Quad-core Intel Xeon 2GHz. It's not because Quad-core Intel Xeon is bad; it's means because Apple (Mac OSX) handle their new hardware properly. E.g. Apple?s Logic Pro 7.2 is unable to take advantages of Quad/Dual-core Intel Xeon.

    Third, I checked claims of an authorized Apple dealer that [b]each third iPod shuffle is defected[b]. I checked two [i]iPods shuffle[/i] and both did not worked (I did not check others).

    It looks like some of those who ?tested? MacBook Pro and other Mac products for the public had really ?independent? from Steve Jobs opinions.

    Nothing new - for many corporations customers are not people who deserve truth.
    Vily Clay
    • I wish you could have spent more time there too

      Next time, start loudly announcing the results of your inquiries. Or make appointments at the Genius Bar and try to pick fights with the employees. That'll show those fascists!
      tic swayback
      • Tic, reading truth about Apple is not healthy for you. Do not do it. (NT)

        (NT)
        Vily Clay
        • The truth is out there

          We need truthseekers like you Vily. Keep up the good work.
          tic swayback
          • I know ? truth about Apple doesn?t have place in your mind. (NT)

            (NT)
            Vily Clay
    • 7 minutes to do all those checks?

      Wow! That was real quick. Either that or you were sloppy.
      labarker
      • I see, for you - nothing is more important than empty talking. Go on. (NT)

        (NT)
        Vily Clay
  • Not able to reproduce these hypothesis'

    I was surprised to read this story, and posts, as I have a MB C2D w/2GBs RAM, and
    I have loaded both Safari and Firefox, and have not noticed much difference at all
    between them. During the last fifteen minutes, because of this story, I opened
    AcMon and watched CPU useage on Safari, it has not gone above 24%, and even
    then, only for a few seconds. Normally it's running about 6 to 8%. CPU useage on a
    multicore machine w/a multitasking OS is a very hard measurement, unless it
    goes to 100 and stays there. Many applications will use as much RAM as they can,
    and let the OS prioritize as other events come online. I'm not sure about OSX but
    my experience in tuning SQL Server Db's in the past led me to not assume that a
    program using 100% of a CPU was misbehaving.


    Also, It does *not* appear to be exhibeting RAM leakage, on my machine, as the
    Real Mem used is fluctuating but holding steady around 67MB. Granted, a slow
    leak could take weeks to show significant growth, but there is nothing here that
    jumps out at me.

    Thoughts: How much RAM do these folks have in their machines? Is this paging
    activity that is going on? Is there a bug in the web pages that they are visiting?
    That's actually more likely the cause.

    Is this a network issue? Is the WiFi being the bottleneck here?
    Other issues that I'm not aware of other than those already mentioned?

    Going into a computer store to do a "test" on download speeds seems an odd way
    to carry it out. How do you have any baseline at all, since you don't know how the
    machines are really configured?

    As to IE 5, I would guess that there are people out there running it because there
    are a few web sites and corporate users that need it for compatbility with
    programs, perhaps Siebel, etc. However, I would never use it, as there hasn't been
    bug fixes or security patches to it in ages, and it's a time bomb, if it's not already
    been used to insert a rootkit., IMHO.

    Given that I could use either Firefox (and do on my Windows desktop) or Safari,
    and assumed I *would* use Firefox when I bought the machine, I have been
    surprised at how nice Safari is, featues, speed, etc. I've almost exclusely gone to
    using it.

    I know it's easy to jump on a bandwagon but this seems like a hypothesis and not
    a fact.
    8string