Since the release of Parallels 5 I have been a huge fan.  Previous to that, I really liked VMware Fusion as the product was much more mature and seemed more stable.  I really had no means to measure performance other than to gauge it by my perception over regular use.

Earlier this week, VMware announced their VMware Fusion 3.1 beta, which I had to run out and try.  I like to keep tabs on both of these products and am always curious when either of them releases an update.

That being said, I downloaded the beta ad built up a new Windows 7 virtual machine, installed the machine additions and played around a little bit.

In the past I’ve done things like measure the boot time for the operating system on each product and in most cases Parallels topped Fusion.  Especially after Parallels released version 5, the gap between those numbers seemed to grow especially when it came to the amount of time it takes to come back on a resume.

As part of my test I also built a new Parallels virtual machine with Windows 7 installed.  I thought an equal way to measure the differences would be by running the tests that provide the results of the Windows Experience Index.  The scale of the Windows Experience Index ranges from 1.0 to 5.9. A higher base score generally means that your computer will perform better and faster than a computer with a lower base score, especially when performing more advanced and resource-intensive tasks.

This host hardware was an iMac with a Core 2 Duo 2.8 GHz processor and equipped with 4 GB of RAM.  Each virtual machine (although not ran simultaneously) had 1 GB of RAM allocated to it.

So, how did each virtual machine fair?  I was actually very surprised.  As VMware has touted lots of improvements around its graphics driver and performance, it actually seemed slower than the 3.0 release.

Here’s how the numbers stacked up:

VMware Fusion Results:

Parallels Desktop Results:

Interestingly enough, it seems that Parallels does a better job in more of the categories than not.  Parallels implementation and support of Aero in Windows 7 seems to be more solid as well.

What has your experience been with these products?  Have you found the optimal virtual machine configuration on your Mac?

It’s a bitter battle between the two – which provides the best feature set while maintaining optimal performance?  I’ve tried both, been back and forth, and can’t quite decide for myself.  Have you tried either or both?  Which do you prefer and why?

A couple of weeks ago I posted a story about virtualized performance where I did a comparison between VMware Fusion and Sun’s VirtualBox.

I wanted to update that story to include some details on Windows performance when Parallels is added to the mix.

Interestingly enough, Parallels really isn’t that much further behind Fusion in terms of average boot time. To be precise, the difference is about .8 seconds between the two. Probably not that noticeable to the average user.

It comes down to features in that case. I also think there is a lot to be said about the stability of Fusion given VMware’s years of experience and product maturity.

This will be a very important year in the Parallels development cycle. We’ll see what they have up their sleeves.