Comments on: ...and statistics

by: corbin

Tue, 16 May 2006 15:50:18 +0000

super duper, again ;) (corby-d around the corner)

by: Zac

Tue, 16 May 2006 16:28:08 +0000

Very nice. I looked on digg to post this story, and someone had already.

http://digg.com/linux_unix/Debunked:_Sehkon_s_Mac_OS_X_Ubuntu_WinXP_Core_Duo_benchmarking_results

Digg it.

by: Michael Tsai - Blog - OS Speed

Tue, 16 May 2006 16:50:51 +0000

[...] ai - Blog - OS Speed Tuesday, May 16, 2006 OS Speed
Peter Ammon:

Now, when allocating memory, malloc can either manage the memory blo [...]

by: Matt

Tue, 16 May 2006 19:13:06 +0000

Can we assume you sent Mr. Sekhon a helpful email?

by: Justin Olbrantz (Quantam)

Tue, 16 May 2006 19:27:47 +0000

Interesting article (particularly the malloc part). Though I’m not so sure you’ve disproved his claim that “For example, in Linux, the variables for a system call are passed directly using the register file. In OS X, they are packed up in a memory buffer, passed to a variety of places, and the results are then passed back using another memory buffer before the results are written back to the register file.”

What you posted makes me wonder if he might have been referring to something that happens in the system call trap before the correct system call handler gets a hold of the CPU. Now, I’ve never looked at the OS X source, so you may already know that isn’t the case, either. In either case, you should check that (if you haven’t) and post that information as well.

by: Travis Cripps

Tue, 16 May 2006 22:50:11 +0000

Very nicely done, sir. I applaud your thorough disputation of the Mr. Sekhon’s claims and the gentle, factual manner in which you debunked them.

Also... It’s great to hear from you again.

by: Mark

Wed, 17 May 2006 01:57:10 +0000

Excellent article, thanks.

by: kermit

Wed, 17 May 2006 02:49:32 +0000

by the end of the day, after all the technical explanations, the results are there and the actual users just have to cope with the speed there is...
I have a 800MHZ G4/17″ and was disappointed with the speed from day 1... my old G3 Imac with os9 was much faster then the new iMac (at that time) I just bought...

by: ssp

Wed, 17 May 2006 05:59:22 +0000

Thanks for the explanations. They’re very interesting. As a casual programmer at best I wasn’t aware of those problems at all. That said, I haven’t done anything performance-critical yet...

As in this case the performance differs significantly, I wonder how often we can expect to see such phenomena? Would there be an easy way to increase OS X’s switching level to a higher number and see how/whether performance or normal applications is affected?

by: Wincent Colaiuta

Wed, 17 May 2006 08:16:52 +0000

As always, fascinating.