Use Caution when Benchmarking the Intel Mac

Because so many people on the Mac platform use Photoshop, some folks in various Mac related forums have been using Photoshop CS or CS2 to benchmark the machines’ performance under Rosetta. That’s entirely appropriate. It demonstrates what user performance today will be like and establishes a foundation from which to later measure gains in native application performance. What’s not appropriate, however, is to pick just one test of Photoshop prowess and then bandy that about as a measure of your Intel Mac’s performance. This is exactly what’s happened in one forum I won’t call by name but uses Photoshop’s Radial Blur to perform a test that turns out to be favorable to the Intel iMac. But along came a reviewer at ARS Technica who performed a bevy of Photoshop tests, and his results painted an entirely different picture. He showed that using Photoshop under Rosetta the Intel iMac was generally much slower than its G5 counterpart with several exceptions, and one of those was radial blur performance.

If you’re thinking about buying an Intel iMac, do. But if you’re basing that decision on benchmarks, then make sure you’re finding everyone you can before you make that decision. The performance picture actually looks pretty good for the most part, but there are pitfalls in moving to the Intel iMac, and software performance under Rosetta is one of them. Be sure you know how the packages you most need to use work to your satisfaction (as opposed to “just working”), and that the information you’re using to tell you that comes from more than one source.