Path: utzoo!attcan!utgpu!utstat!jarvis.csri.toronto.edu!rutgers!apple!well!pokey From: pokey@well.UUCP (Jef Poskanzer) Newsgroups: comp.unix.wizards Subject: Re: Optimal for loop on the 68020. Message-ID: <12080@well.UUCP> Date: 8 Jun 89 21:45:08 GMT References: <19945@adm.BRL.MIL> Reply-To: Jef Poskanzer Organization: Paratheo-Anametamystikhood Of Eris Esoteric, Ada Lovelace Cabal Lines: 21 In the referenced message, rbj@dsys.ncsl.nist.gov (Root Boy Jim) wrote: }Hey, aren't most programs *I/O bound* anyway? And you can sometimes }unroll your loops. Any statement about "most programs" is guaranteed false. In my case, I have a large suite of programs called PBM. Most of them (note difference) are I/O bound (but less so now that I've improved the external data format). Some are CPU bound. I estimate that these would go about 1% faster if they used a dbra in their inner loop. }I seem to remember Chris Torek stating something about using } printf("n = %d\n"); Somehow I doubt Chris would make such a stupid mistake. What is your point, anyway? I didn't see it in your article. --- Jef Jef Poskanzer pokey@well.sf.ca.us {ucbvax, apple, hplabs}!well!pokey "Sound impressive? It should. It is."