Linked by Thom Holwerda on Sun 7th May 2006 14:05 UTC, submitted by Mystilleef
Benchmarks Jasjeet Sekhon benchmarked Linux and MacOS X on the MacBook Pro using his statistical software, and finds that "Linux is found to be much faster than Apple's OS X for statistical computing. For example, in one benchmark Linux is more than twice as fast." Earlier, he ran tests on a G5 and an Opteron, and conlcuded: "Those results were terrible for OS X and not particularly good for the G5 (970) chip. For example, my 2.7 pound Pentium-M Linux laptop is faster than my 44 pound G5 running OS X. The floating point performance of the 970 chip leaves much to be desired, but OS X makes the performance problem significantly worse."
Permalink for comment 122331
To read all comments associated with this story, please click here.
What is more important...
by markus on Mon 8th May 2006 17:30 UTC
Member since:

I agree that some portions of the Mac OS X kernel need optimizations but:

- If I could get a nice statistical software that runs fine under Mac OS X and plays well with other software I use AND the workflow under Mac OS X would safe MY time (and I suppose it would) I would run that software under Mac OS X and not under Linux to save some cpu cycles.

- I also run some file / directory servers under Mac OS X because integrating Mac / Win / Linux works for me really well (and nobody complaned about performance so far)

- I also think that the statistical software package may have been tweaked (over time) for Linux perfomance - Mac OS X is not Linux and that does not means that it is slower but that it has a different performance profile.
There is an article on Apples developer pages how to optimize MySQL performance by turning off a MySQL feature that dynamically changes thread priority - the very same feature that increases performance under Linux.

- This is not to say that Apple should not optimize the kernel that has 16 year old code in it (Apple has allready done this successfully with launchd) - but this article draws just the wrong conclusions.

Reply Score: 1