Slashdot Mirror
Parallels Desktop For Mac Vs. VMware
neilticktin writes "MacTech performed an exhaustive set of benchmarks comparing Parallels Desktop 4 to VMWare Fusion 2 to run Windows on a Mac. To tackle this problem, MacTech undertook a huge benchmarking project starting in December — over 2500 tests by stopwatch. The goal was to see how the recent versions of VMWare Fusion and Parallels Desktop performed on different levels of Mac hardware, using XP, Vista, 64-bit, multi-procs, games, etc. ... As usual, results vary by what's important to you."
When people by a Mac and then run Windows on it.
I always laugh. Like now.
I'll try anything once. Twice if it tastes good
That would be 99.99% of home users. It's hard to conceive of an application for ESX[i] at "home," given Fusion and Workstation. ESX is heavyweight and particular in its hardware requirements, nontrivial to manage (especially with the free license), and just generally not the right thing if you don't have a spare tower server or DC handy, a full license, and someone else to pay your power bill. Although, in those circumstances, it's pretty cool.
(A bunch of the remaining .01% are going to explain why I'm wrong now.)
Disclaimer: I work for VMware. (And I would run ESX at home if there was a reason to.)
When you're comparing performance of compressing an 8GB folder with 1000 files, or total time to encode a 2-hour movie, it's perfectly acceptable to use a stopwatch, and have your margin of error be +/- 1 second.
Convert FLACs to a portable format with FlacSquisher
I also couldn't run Parallels more than 15 minutes on my Mac Pro without it causing a kernel panic. I'm glad to hear someone else had the same problem.
I switched to VMWare Fusion and haven't looked back.
Gee, who would have thought that spreading your article across TEN BLOODY PAGES would increase the load on your servers? Idiots and their ad impressions...
They do it because people were posting benchmarks based on mis-configured systems. It would be like running a 3D benchmark on the latest-n-greatest new $600 video card, but without installing 3D accelerated drivers. If it were your product, you'd want competent people posting "authoritative" benchmarks (that laymen would consider "authoritation").
Think about it.
Yeah, that's a huge market. And it's full of customers with deep pockets!
What makes you so sure it's not? I bet it's pretty common for a company to have an in-house app that was custom-written for them back in the day, which works fine for them. Ten years later, they still want to run that app, but they can no longer easily find hardware that supports the OS the app was written for. So now their choice is either re-write the app from scratch (an expensive and risky project; the people who wrote the original version are long gone of course) or spend $70/seat on a virtualization product to keep the original program running. For a company with thousands of seats, that would be a major opportunity for VMWare or Parallels or whomever to make a large amount of money.
I don't care if it's 90,000 hectares. That lake was not my doing.
This is the general consensus from everyone I know who uses their VM product for more than the never-switched-on safety net of a stale Windows install:
Parallels gets you there faster if it manages to get you there at all. Fusion just works. I had largely the same experience. If I wanted it to be as fast as possible, I'd bother to BootCamp it. Speed is always secondary to reliability for me.