From:	jcoffin AT taeus DOT com (Jerry Coffin)
Newsgroups:	alt.lang.asm,comp.os.msdos.djgpp,comp.lang.asm.x86,alt.os.assembly
Subject:	Re: Benchmarking of NASM
Date:	10 Jul 1999 13:40:35 GMT
Organization:	TAEUS
Lines:	27
Approved:	<johnfine AT erols DOT com>
Message-ID:	<7m7igj$1gv$11@autumn.news.rcn.net>
References:	<7m645n$f0a$1 AT autumn DOT news DOT rcn DOT net>
X-Trace:	41EMUBSEbjwoXBaM1NqSFweUUf+J/ScX/BlNseB/ejE=
X-Complaints-To:	abuse AT rcn DOT com
NNTP-Posting-Date:	10 Jul 1999 13:40:35 GMT
X-Newsreader:	MicroPlanet Gravity v2.10
X-Server-Date:	10 Jul 1999 09:51:18 GMT
To:	djgpp AT delorie DOT com
DJ-Gateway:	from newsgroup comp.os.msdos.djgpp
Reply-To:	djgpp AT delorie DOT com

In article <7m645n$f0a$1 AT autumn DOT news DOT rcn DOT net>, hardwork AT freemail DOT c3 DOT hu 
says...
> Please forgive me if this has been mentioned before.
> 
> Is there any benchmark we can run to compare the performance of NASM with
> other x86 assemblers like A86, MASM, TASM and OPTASM?

I haven't got solid numbers from a benchmark, but NASM is NOT a 
particularly fast assembler -- the most recent versions of the others 
are clustered close enough together that you have to keep pretty close 
track of the exact system in which tests are conducted before they 
mean anything at all (though, of course, the most recent version of 
optasm isn't really very recent at all).  NASM is slower than the rest 
by a fairly wide margin.  Of course, on a modern computer, you're 
unlikely to notice much real difference unless you're assembling a LOT 
of code.

The number and size of files you use will make a difference as well -- 
MASM (for example) takes a while to load and start up, but assembles 
really fast once it's working.  NASM tends to load faster, but doesn't 
run terribly fast.

Therefore, if you have a small number of really large files, MASM will 
win by a large margin.  If you have a lot of smaller files, things 
will be much closer, though I'm pretty sure MASM will still normally 
win.

- Raw text -