Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!yale!mintaka!bloom-beacon!eru!hagbard!sunic!cs.umu.se!dvlmfs
From: dvlmfs@cs.umu.se (Michael Forselius)
Newsgroups: comp.sys.mac.misc
Subject: Re: Not another NeXT defector???!!!
Message-ID: <1990Nov8.150943.20325@cs.umu.se>
Date: 8 Nov 90 15:09:43 GMT
References: <F.*m?7$2@cs.psu.edu> <1990Nov5.095930.27753@cs.umu.se> <cbBQbw200UgKEa9GNv@andrew.cmu.edu>
Sender: news@cs.umu.se (News Administrator)
Organization: Dep. of Info.Proc, Umea Univ., Sweden
Lines: 26

In article <cbBQbw200UgKEa9GNv@andrew.cmu.edu> vd09+@andrew.cmu.edu (Vincent M. Del Vecchio) writes:
>> Excerpts from netnews.comp.sys.mac.misc: 5-Nov-90 Re: Not another NeXT
>> defect.. Michael Forselius@cs.umu (798)
>
>> And remember that no compiler NEVER generates code as efficient as
>> hand-optimized assembler.
						^^^^^^^^^^^^^^^^^^^^^^^
Michael speaking:

	Hey guys, gimme' a break will you - don't blame me for that, if
	you were to write this in Swedish...

	If you've ever waded through tons of compiler generated code then
	you'd know what I'm talking about, If not - come back when you have.

	Write a simple loop in your favorite language,  dissasemble the code
	made by the compiler, what do you see??? Pretty bad code I suppose.
	Now consider that this is just ONE loop taking perhaps 10-... CLK
	(clock cycles) more than it has to and we loop a lot don't we?
	Compilers in general doesn't make especially fast code for
	array indexing either and ...

	Now cut off this discussion and move it to comp.compilers.

Have a Nice Day
	/Michael