Volume Number: 26
Issue Number: 01
Column Tag: Virtualization

Head-to-Head: Parallels Desktop for Mac vs. VMware Fusion

How do VMware Fusion 3 and Parallels Desktop 5 for Mac compare?

By Neil Ticktin, Editor-in-Chief/Publisher

< Previous Page Start | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 Next Page>

After installing the game through Steam (http://store.steampowered.com/), we tested the game by measuring frames per second while running a consistent demo mode.

In short, the speed difference in FPS was noticeably faster in Parallels Desktop when compared to VMware Fusion. As expected, both environments took advantage of the extra graphics hardware of the MacBook Pro and Mac Pro, but Parallels took greater advantage of it. See the graph.

Here were the results:

• Portal, Frames Per Second in demo mode
  • XP: Parallels Desktop 130.4% faster (28.5 frames out of 50.26 average FPS)

  • Windows 7: Parallels Desktop 498.3% faster (94.4 frames out of 113.39 average FPS)

• Civilization IV: Colonization, FPS in demo mode
  • XP: Parallels Desktop 52.6% faster (18.2 frames out of 52.73 average FPS)

  • Windows 7: Parallels Desktop 45.2% faster (13.7 frames out of 44.12 average FPS)

• Quake Wars, FPS in demo mode
  • XP: Parallels Desktop 28.1% faster (4.1 frames out of 18.92 average FPS)

  • Windows 7: VMware unable to run. Parallels ran game fine.

Figure 21: Gaming Performance

Our advice? There's no doubt, Parallels Desktop 5 is a huge leap forward not only from Parallels Desktop 4, but also all versions of VMware Fusion.

If gaming is your primary reason for a virtual machine, then figure out what games you want to play, double check the online discussions about them, and go for it. Furthermore, you may want to use a virtualization setup that allows you to share a Boot Camp volume with your virtual machine so that you can have even greater performance when you want it.

Either way, we were pleasantly surprised at just how well these games played in Parallels Desktop 5. And we highly recommend the games we looked at and tested here.

There are additional conclusions that we can extract from the results as well. Specifically, we looked at the differences between XP and Windows 7, as well as multiple virtual processors and 64-bit Windows 7.

In our last virtualization benchmarking article, there was a significant difference between running Microsoft XP vs. Vista. Now, the difference is much less significant. In fact, while XP continues to be faster for most things, it's probably not enough to matter. You should make your choice based on which OS you want or for compatibility reasons. That said, anecdotally, we find XP under virtualization to be snappier … and definitely less annoying to run.

There's a big push right now for multiple virtual processors, and 64-bit Windows. While there are times that you may need them, most people will not.

Multiple virtual processors are helpful for when you have a computationally intensive application, and you need to split the work. The types of applications that you normally would need this for include video, Photoshop, CAD, etc... Frankly, if speed is that important to you, you should be asking yourself about whether to run the app native on your Mac instead of in a virtual machine. Sometimes, like for CAD, you may not have an option.

64-bit is another issue. Primarily, the big benefit here is that you can address a lot more memory. For most virtualization users, this is likely not relevant. Many Windows applications and drivers still are not 64-bit compatible. This is the future of where Windows is going, but it's not necessary for most users, especially under virtualization.

That said, we wanted to give you a look at what performance looked like for both of these for both IO and virtual machine performance. As you can see, while there's some difference, you have to judge if it's enough to be worthwhile. For most, 32-bit XP is likely adequate.

Figure 22: Multiple Virtual Processors, 64-bit Performance

Both VMware Fusion and Parallels Desktop for Mac are excellent products, and both allow you to run Windows XP and Windows 7 quite well (except for graphics in VMware Fusion under 7). In the end, your decision as to which product you should take into account what's most important to you: speed, footprint, graphics capabilities, features, user interface, OS you want to run, and more all come into play.

While the Windows Vista "penalty" that we saw in prior tests is now mostly gone (presumably because both Windows 7 made improvements as well as both of these virtualization products), we would advise that you stick with Windows XP given that it runs slightly better overall (not to mention how much less annoying it is).

When it comes to whether you should use multiple processors or 64-bit virtual machines that depends on your use. If you have a real need for either, and can articulate a reason for it, than use them. They do work well. That said, if you don't have a specific need, then don't bother, it's not worth it; just stick with 32-bit Windows XP (or Windows 7) on a single virtual processor.

Many people have the feeling of "more is better," but when it comes to RAM in the virtual machine, that is not necessarily the case. More RAM means longer virtual machine launch times, suspends and resumes. For most users, 512MB to 1GB of virtual machine RAM will work best. Use more than that only if you really know you need it.

Here's how things look in general terms for each of the test suites that we ran:

Figure 23: Chart: Performance Winner in Each Test Suite

In the majority of overall averages of our tests, Parallels Desktop 5 is the clear winner running 30% faster than VMware Fusion 3.0.1 with Windows XP, and 43% faster with Windows 7. And, the difference is even more apparent when looking at graphics. If gaming, graphics, and 3D are your thing, you have no choice. Parallels Desktop 5 has so much better graphics support, and is so much faster in most of the comparisons, there's simply no contest.

One thing is clear: virtualization for the Mac works well. Really well—even for casual gamers. Even with that, given the track record, I expect we'll see it keep getting better and better.

About the author...

Neil is the Editor-in-Chief and Publisher of MacTech Magazine. Neil has been in the Mac industry since 1985, has developed software, written documentation, and been heading up the magazine since 1992. When Neil does a benchmark article, he likes to test the features that people will use in real-life scenario and then write about that experience from the user point of view. Drop him a line at publisher@mactech.com

< Previous Page Start | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 Next Page>