Chrome 10 vs. Internet Explorer 9 Reconsidered

Chrome 10 vs. Internet Explorer 9 Reconsidered

Summary: I, and a lot of my readers, were puzzled about why my recent SunSpider results showed IE 9 doing so badly compared to Chrome 10, so I took a closer look. This time IE 9 took first by a nose.

SHARE:

When I recently took a look at the brand new Chrome 10 Web browser, the results for the latest release candidate of Internet Explorer (IE) 9 puzzled me because they were so bad. Some of my readers were more than puzzled. Some were outraged and accused me of deliberating trying to set up IE to fail. Please. I call them like I see them and, in my tests, the IE 9 RC was just dreadfully slow compared to the others.

Still, as I said, I wasn't happy with my results. When I started hearing from people that I respected, like my ZDNet bud Ed Bott, that he was also seeing results that put IE 9 RC just ahead of Chrome, I decided to re-run my benchmarks and take a closer look at my results.

The first thing I did was to start talking with Microsoft, who, as you might have guessed, had wanted to talk to me anyway, and to set up a new benchmark PC. This time around I picked a Gateway SX2802-07 desktop. This PC uses a 2.6GHZ Intel Pentium Dual-Core E5300 processor and has 6GBs of RAM and a 640GB hard-drive.

I burned out everything that currently lived on that system with my favorite system repair Linux distribution SystemRescueCD. I then installed on Windows 7 Ultimate 64-bit from the partitions on up. After that I brought it up to date with Windows 7 SP1. In passing, let me say that while I know some people have had trouble with Windows 7 SP1, SP1 is a good patch. While it's not as big a comparative improvement as Windows XP SP3was in its day, I think most Windows 7 users would be better off going with SP1.

OK, so I now have a fresh system with nothing on it except Windows. Once more I installed the latest IE 9 release candidate. This time though there was nothing else on the system, not even Chrome 10 yet, that might cause trouble. And, once again I sent IE 9 64-bit on its trip through the SunSpider JavaScript benchmark 0.91 and this time it came in 1111.3ms (milliseconds). Yuck!

I was getting closer to working out the riddle though. I had used the 64-bit version of IE9 because generally speaking 64-bit applications run marginally faster on 64-bit architectures. Now, with a Web browser that's not likely to make a big difference. The real win for the current generation of most 64-bit software is its access to more than 4GBs of RAM. Or, in Windows case, about 3.25GB, once you've taken out the RAM used for the BIOS, the PCI and PCI Express buses, etc. etc. With a 64-bit system, you have a theoretical maximum of 16 Exabytes, or about 16-billion GBs, but Microsoft currently puts a 16TB limit on address space and allows only 128GB of physical RAM. Only HD photography and video editors are likely to run into limits.

Still, 32-bit IE 9 would need to run on Windows on Windows 64-bit (WOW64)--which had nothing to do with World of Warcraft--and WoW64 could have caused slightly reduced 32-bit IE 9's speed. Well, that's what I thought anyway.

But, then I ran 32-bit IE 9 on my 64-bit Windows 7. Wow. Now, IE 9 came in at 245.4ms. This was actually faster than Chrome 10, which came in at 266.7ms. So, for the moment at least IE9 is actually the fastest browser I've tested to date.

Page 2: [Why 32-bit IE is faster than 64-bit IE] »

Why 32-bit IE is faster than 64-bit IE

So what the heck was going on here? This time, guided by comments from Ed and others, I looked at the JavaScript JIT (just-in-time) compiler. Ah-ha! It turns out that IE 9 64-bit, which I presumed was the most advanced version of IE, was actually using an older, much slower JavaScript JIT engine, while IE 9 32-bit was using the far more efficient Chakra JIT.

Sure enough, when I asked, Tim Sneath, Microsoft's Senior Director of Windows and Web Evangelist, "The 64-bit IE 9 RC doesn't use the Chakra JIT does it? It uses the older, far less efficient one, right?" Sneath replied, "That's correct--we put all our JIT compiler investment in this release of Chakra into the 32-bit compiler, since that's the one that is near-universally used today. The 32-bit browser is the default even on 64-bit systems, since almost all add-ons are 32-bit only at this time. The 64-bit version provides no real advantage, since no browser scenario requires access to >2GB of user memory in a single process."

The mystery is solved!

For more on this Microsoft IE design decisions see Eric Law's Q&A: 64-Bit Internet Explorer. Even though this piece is almost two years old, Sneath assures me it's still accurate.

The takeaway is that, for now anyway, IE9 32-bit is unquestionably the best version of IE to run on any version of Windows Vista or 7. For the moment at last, IE 9, and not Chrome 10, is the fastest Web browser around.

I'm sure we'll see the lead changing back and forth in the next few weeks as IE 9 goes to final, Firefox 4 finally arrives, and minor updates are expected from all the Web browser makers in response to the Pwn2Own Web browser cracks.

As for me, I'll still be using Chrome. It may not be the fastest, but it is fast enough, has great security-the Pwn2own hackers wouldn't even try to break Chrome even though Google offered a $20,000 prize for any successful hacks, and it works on all my commonly used operating systems: various Linux distributions, Mac OS X and Windows XP and 7. That said, if all you use is Windows, IE 9 is a much faster choice than I had first thought and is a gigantic step up from IE 8.

Topics: Hardware, Browser, Microsoft, Processors, Windows

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

Talkback

149 comments
Log in or register to join the discussion
  • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

    strange. I got 219ms on my pc last time. I?m pretty sure i?m always running 64bit but i?ll recheck tonight. btw my pc is intel Quad q9550 (2.8ghz) + 4gb ram (800mhz).
    bnlf
    • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

      @bnlf Not really your Quad's a faster than what I used.

      Steven
      sjvn@...
      • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

        @sjvn@...
        I have 12 quads on my PC, beat that!!!
        Hasam1991
      • So Steven, when you tested IE9 on Windows 7 *32* bit yesterday

        @sjvn@... <br>This you wrote at 03/09/2011 11:51 AM:<br><i>@G-Systems You know I thought my IE9 numbers were quite slow also, so before I wrote that tale, I tried it several more times--on Win 7 SP1 <b>32</b> and 64-bit systems and on several different PCs. It still was slow as heck.</i><br>Are the results you get "today" different now that you know the explanation? What changed? <br><br>IE9 64 bit will not run on 32 bit Windows, so if you tested on a 32 bit Windows 7 SP1, how come you still had it "slow as heck" when we now know that the explanation is that the 64bit RC is not yet using the same JIT'er?

        Steven, did you <b>really</b> test IE9 on a 32 bit Windows 7? If you did, then I am really puzzled that you are satisfied with the 64/32 bit explanation.<br><br>Or, were you (gosh!) just <b>lying</b> a little bit about the lengths you went to?
        honeymonster
      • An unclear correction has led to more confusion.

        @ honeymonster

        Vaughan-Nichols's initial deception, and less than clear correction, has how generated a new round of confusion, as readers try to figure out how to install/run the 32-bit version of IE9 on 64-bit Windows. What they don't realise, of course, is that 32-bit IE9 is the one they've been running all along -- and that all the other major browsers are 32-bit too.

        This fact that 32-bit is the default for all the major browsers, including IE9, is central to any open and honest correction. Unfortunately, this key point was only mentioned in a quote from Tim Sneath on page two, amongst a barrage of irrelevant details and confused nonsense (e.g. confusing RAM limits in 32- versus 64-bit OSes with the address-space limits for 32- versus 64-bit processes running on a 64-bit OS, and the implication that things like reinstalling the OS had anything to do with the results).

        Sheer incompetence may explain Vaughan-Nichols's implication that 64-bit IE9 is the norm, and that running 32-bit IE9 requires some extra effort. However, in view of the record, I'm afraid I'm inclined to suspect otherwise.
        WilErz
      • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

        @WilErz ... Sheer incompetence may explain your implication that 32-bit browsers are the norm on 64-bit systems, since of course on a Linux 64-bit system you're almost certainly running a 64-bit version of the browser as well. Oh wait, you meant on Windows! That's right, maybe, just maybe, when you know one platform better than the other, you transfer assumptions. And maybe it's not incompetence, maybe it's being human. And maybe we should all cut each other some slack. Sheesh, it's a blog, not a frickin' peer-reviewed technical journal.
        daboochmeister
      • Deceit, not incompetence

        @WillErz

        My point was that SJVN <b>claimed</b> that he had even tried running IE9 on a 32bit Windows 7 SP1 - and that it still produced the same "slow as heck" results. He made that claim not in his article but in the talkback to back up why he stood by his numbers.

        We now know this was a <b>lie</b>. It is the 64bit IE9 which does not feature the speedy JIT. <b>But SJVN could not have tried 64bit IE9 on a 32bit Windows</b>. If it was true that he even tried the 32bit Windows he would have gotten the same fast results as everybody else.

        @daboochmeister

        For certain the Windows platform is not SJVNs strong side. That does not hold him back from posting frivolous claims about it, though. Sometimes he will post pure speculation (like Windows being the culprit of the LSE network problems), but at other times he will post claims such as that Windows 7 was originally designed as a single-user system.

        You may excuse incompetency, but only so far. When you pass yourself off as tech savvy there are some minimal qualifications you have to meet.

        But even if you allow some room for errors, when those errors show a clear tendency to err in a certain direction, incompetency can not be the lone explanation. Ignorance or arrogance coupled with bias comes to mind. Although personally I suspect he is knowing playing to myths to advance personal agendas.
        honeymonster
      • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

        @ honeymonster

        I wasn't disagreeing with you at all, just pointing out that, by avoiding a simple admission of his 'mistake' (genuine or malicious), Vaughan-Nichols has created yet more confusion.

        @ daboochmeister

        Please do at least try to think before you post. The article is about IE9 and other browsers on Windows.
        WilErz
    • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

      @bnlf yea a E5200 is pretty slow. So if your running a C2Q then its not surprise that your results were alot better. I have a Phenom II so if I ran it I'm sure I would get a much better score aswell.
      Jimster480
    • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

      @bnlf
      I just ran the benchmark and got 176.1. I'm running 64 bit ie9 on an SSD.
      Dmachine
  • This was a problem during the IE9 beta as well!

    Scroll down to the beta benchmarks on this page: http://arstechnica.com/microsoft/news/2010/10/windows-browsers-benchmarked-october-2010-edition.ars
    Empro
  • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

    @cabdriverjim - it's a question of resources.

    If you had a team with a finite number of people and a finite amount of time available to create a brand new product, would you opt to spend the majority of your time optimizing your product for 99% of your customers or for 1% of your customers?

    As the article correctly states, only apps that require > 2GB RAM per instance will benefit from being ported to 64-bit. 64-bit code and data is bigger than 32-bit code and data. This means that 64-bit code and data takes up more disk space, more memory, more IO to read and write to/from disk and reduces cache efficiency.

    Rememberin that IE9 largely runs each browser tab in its own process. Thus, a 64-bit browser isn't REQUIRED unless an individual page required > 2GB RAM.

    Further, until Adobe and others port their plug-ins for Flash, Reader, etc. to 64-bit, 64-bit browsers (like IE8/9 64-bit) won't be able to render flash, video, PDF's, etc.

    Microsoft have done absolutely the right thing by spending all their budget on optimizing IE9 for 99% of its customers and partners. They're already working on optimizing Chakra for 64-bit and will release that update later.
    bitcrazed
    • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

      @bitcrazed There are already 64 bit plugins. I use 64 bit minefield with no issues. And the 64 bit app uses no more ram than a 32 bit app. Because with 64 bit processing, things that would take 2 sets of instructions can take only 1 (because of the larger addressing).
      Jimster480
    • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

      @Jimster480 - you may be using 2 plug-ins, but there are tens of thousands out there that are still only 32-bit.

      You clearly don't understand what happens when you compile your app to x64: 64-bit apps are on average around 25-30% larger than the exact same code compiled to 32-bit. This is because every pointer is now doubled in size (to 64-bits) and that every int is now 64 bits long rather than 32 bits long.

      64-bit code can, in some situations, make better use of the larger number of registers available on x64 chips.

      As you (partially correctly stated) x64 code can sometimes optimize calculations by using larger operands, thus reducing the number of instructions used, but this generally constitutes a microscopically small proportion of most app code.

      As an example, compare the sizes of the common files in "C:\Program Files\Internet Explorer" (64-bit IE) and "C:\Program Files (x86)\Internet Explorer" (32-bit IE).
      bitcrazed
  • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

    @cabdriverjim thats just IE. I run 64 bit minefield with 64 bit Flash. And there are no issues. Its very fast aswell. And it has HW acceleration.
    Jimster480
  • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

    Have you ever actually done anything at all with 32/64bit development? For one thing, _every_ program on _every_ system that runs on x86_64 (at the least) has to be rearchitectured for 64bit. When SuSE GMBH brought out the first 64bit Linux they had to provide both 32 and 64bit libraries and guess what, the browser plugins only worked in 32bit. This is _not_ a Windows specific problem, this is a problem borne of silicon which provides 64bit instructions as an extension to a natively 32bit marchitecture.

    Next time, get a clue before spooging your ignorance across the internet.
    odubtaig
    • RE: Chrome 10 vs. Internet Explorer 9 Reconsidered

      @odubtaig

      I was in Sun Microsystems' Solaris OS/Net group when we did the first 64 bit version of Solaris. While I don't want to trivialize the effort, to say that "_every_ program on _every_ system that runs on x86_64 (at the least) has to be rearchitectured for 64bit" is a gross exaggeration.
      philip.robar@...
  • PS

    No-one except a few noobs give a crap about 64bit browsers but a lot of people can only run 32bit. Let's count the number of browser developers officially releasing 64bit builds...

    No-one cares, but damn you numerically obsessed nerdlingers sure are clogging up the internet with your self-important crap.
    odubtaig
    • OK, Let me see if I have this correct...

      We applaud MS for keeping up with Jones by putting out a product that still directly competes with every other browser ever made. (Oh, but it's faster (for now)). MS has deep enough pockets that they could have taken the risk and been a pioneer, a champion for moving forward with 64bit development. The fact that MS has lost ground in the marketplace (Slight, but noteworthy) is because they are consistantly last to the table with competing products. The whole Andriod/IPAD revolution has success because it has a champion that was willing to push a few boundries. Declaring those who value innovation, and help prod it along to be self-important - is simply just ignorant. The self -important person see's technology only in the here and now. Without the "noobs" you wouldn't even have 32bit - you'd still be counting your brain cells on your fingers. I am just saying...
      cnejenkins
  • The MS story keeps getting funnier and funnier

    And to think the MCSEs mocked Apple with Snow Leopard.

    MS declares, in writing, they put more effort into past generation technology (decades old). Technology retired on other platforms. Lets go fanboys;-)
    Richard Flude