Rosetta vs. Boot Camp vs. VT

When the Intel based Macs were first introduced, the only option for running non-native binaries was through Rosetta, a binary translator. As we showed in our iMac Core Duo review, performance using Rosetta was pretty horrible, but that was compared to an iMac G5. Would someone upgrading from a PowerBook G4 really notice the performance difference between an application running natively on their G4 and an application running on top of Rosetta on a MacBook Pro?

To find out I ran the same MS Word tests I ran a couple of months ago on the iMac Core Duo, and compared the results between the native version of the application running on the PowerBook G4 and the Rosetta powered MacBook Pro. As an additional pair of reference points I also ran the same benchmark using Word 2003 on a Windows XP partition using Boot Camp as well as Parallels' Workstation VM.

Non-Native Performance - Microsoft Word

Although Rosetta on the MacBook Pro wasn't that much worse than the G4 in opening the test document the HTML conversion process was significantly slower. What's truly impressive is how much faster everything is under both Boot Camp and running under Parallels' VM. Although we can't explain it, the VM consistently posted slightly faster times than under Boot Camp, although it wasn't perceivable other than by stopwatch.

Non-Native Performance - Microsoft Word

What these tests show is a very interesting alternative to Rosetta - simply running the non-native application in Windows instead. While you don't get all of the benefits of running something in OS X, you do get a huge increase in performance. It's also quite impressive to see that the VM solution truly is basically as fast as running Windows XP natively using Boot Camp.

I dug a little deeper to see how the Boot Camp and VM offerings stood up to one another in a couple more tests. The first being a script that Intel uses to measure performance using Adobe Photoshop CS2 (if I could run the script under OS X I would, but it is not supported).

Windows XP Performance - Adobe Photoshop CS2

Once again the Parallels' Workstation VM manages to be a bit faster than Boot Camp, which doesn't exactly make sense but the results were consistent.

I wanted to run both sets of Winstone benchmarks but Business Winstone 2004 consistently failed to complete running in the VM. The applications themselves didn't crash but I suspect there may have been a timing issue related to the VM. Luckily, I could at least get Multimedia Content Creation Winstone running and got a set of results from it:

Windows XP Performance - MMCC Winstone 2004

While Boot Camp is slightly faster, the difference is hardly noticeable. This is quite an important benchmark because it shows that disk as well as CPU performance isn't hampered by running Windows XP on a virtual machine. Unfortunately what these benchmarks don't measure is the responsiveness of the UI, which truly does suffer under the VM option.

Using Windows XP on a VM Network Performance and MacBook Pro vs. PC Notebooks
POST A COMMENT

48 Comments

View All Comments

  • ss284 - Friday, April 14, 2006 - link

    Which pretty much drops macworld's benchmark credibility to zero. Actually, their credibility was already really low so its no big deal. Reply
  • ss284 - Friday, April 14, 2006 - link

    Unless the benchmark was how fast it could burn a hole in your pants and sterilize your important parts. Reply
  • jbb132 - Thursday, April 13, 2006 - link

    Sadly, even the most recent MacBook Pro's continue to suffer from the hardware problems you noticed. I've now had two units with the "whine," particularly when the laptop is running on battery power. The only way to stop it is to turn photobooth on and leave it on. Various hacks (magicnoisekiller in particular) help but really...
    Reply
  • Pirks - Thursday, April 13, 2006 - link

    Hey Anand!

    Thanks for an interesting review! Now, what was the video source codec and resolution in your H.264 encoding test? I've got a Mac zealot here saying he's got only 2 frames per second in similar H.264 encoding task. He also has Mac OS X 10.4.6 and Quicktime Pro 7.0.4. I wonder is you used source with resolution like 160 by 120 for that test??

    To other readers: take with a grain of salt Anand's view on running multiple apps simultaneously in XP. He says something like "uh oh we can't run multiple apps all the time on win coz win can't manage its RAM blah blah", but in fact if you disable XP swap AND if you stick 2 or 3 gigs or RAM in your mobo you'll get my experience of running dozens of apps open at the same time and even some games while at it (DOOM 3 + Matlab + Maya 7 + other little apps like VDub open at the same time is a no brainer on my XP). Stick 4 gigs and open even more, without loss of performance. In fact if you wanna get Mac OS X experience, just stuff your mobo with RAM and turn off swap, that's it, no need to spend $$$ on Mac to get this "experience" :-))

    One last minor correction - Anand tells you "anything more than 2GB of RAM on your PC is useless" but he does not know about /3GB switch (google up "/3GB switch") which again allowed me to run Matlab with THREE gigs or RAM consumed, not two.

    Just my 2 cents to debunk some myths that Mac zealots love to spread ;) Don't get caught in that stuff, read docs/mans and be smart.
    Reply
  • Anand Lal Shimpi - Friday, April 14, 2006 - link

    The source for my H.264 test was the Hoodwinked trailer from Apple's Quicktime trailers site.

    My comments about memory usage and Windows XP have nothing to do with the /3GB switch. The point I was trying to make is that Windows XP does not do a good job of caching to reduce disk accesses. Microsoft itself has admitted that there's lots of room for improvement, which is why you hear about all of the caching improvements that will be introduced in Vista.

    Take care,
    Anand
    Reply
  • kleinwl - Thursday, April 13, 2006 - link

    BTW: if you don't think that anandtech knows about the /3GB switch you certainly haven't been here very long. Reply
  • Pirks - Thursday, April 13, 2006 - link

    quote:

    if you don't think that anandtech knows about the /3GB switch you certainly haven't been here very long.
    Did Anand ever mention this switch in any of his reviews? Or anyone else besides him from AT staff? I've never seen it before. How about you? Would you provide me with a link or some other proof? I'd love to be corrected here, since AT guys should know about it, and I wish I were wrong stating the opposite, especially about Anand himself.

    As for the stability - I run Matlab with 3GB consumed routinely, it also loads some Maya 7 stuff internally via my own DLL and there's another Maya 7 hanging around to check results from Matlab quickly, and I never saw a slightest glitch. Of course YMMV but I heard too much "omg windoze is 2GB limited and mac is TrUe 64-bit WoNdEr" and I've got some experience with Matlab on both Mac OS X (no 64-bit matlab there) and WinXP (now this is true 64-bit product) to let Anand repeat that kind of sh.t. Mac guys are ok to say that, they live like this, so no big deal but not Anand please :-) So let's just say "please Anand be a bit more correct in some places and don't sound like a dumb Mac user", saying windows can't properly run a lot of apps at once and stuff like that.

    Speaking of Vista I've read somewhere on MSDN that a lot of XP 64 code is in there so it _should_ be as robust as XP 64 with regard to RAM handling. I'm sure after SP1 or SP2 it'll be absolutely rock solid :-)
    Reply
  • kleinwl - Thursday, April 13, 2006 - link

    I've used the /3GB switch as well... however it can (and does) degrade the stability of your system (running fluent (a CFD program) on XP SP2). I wouldn't run around recommending the normal use of the /3GB switch. Some programs don't even support it (ie. Catia V15). Ultimately we upgraded to XP x64 (which came with it's own headachs).

    2.5GB seems to be the most that XP SP2 can really handle well.. everything else is a waste. OS X just does a better job handing large amounts of ram and not "losing" it with time.

    I am interested is seeing how VISTA will perform and if will be as good as OSX or XP x64 with ram... or something quite better (or worse).
    Reply
  • boinkle - Thursday, April 13, 2006 - link

    Great review, apart from making me think "that's still quite pretty!", at the end of the day it's just a PC in an Apple enclosure. It's shiny, fast, but has all the problems of an Apple 1st gen product, with few of the benefits of Core duo showing up...

    How I wish someone had given Freescale some incentive to develop the G4 further (to a reasonablt timescale). It's amazing that it's still even *reasonably* competitive. Imagine dual-core, 65nm G4 derivative production? Don't TI have a 65nm fab up and running now? That's where your 5 hours would come from, Anand... pie in the sky, I know.

    *sigh*
    Reply
  • littlebitstrouds - Thursday, April 13, 2006 - link

    How about some gaming benchmarks. If you run windows and game are you getting good performance numbers? If I could have a Mac for everyday use and boot up windows when I want to game I might jump on this. Reply

Log in

Don't have an account? Sign up now