Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!pdn!tscs!tct!chip From: chip@tct.com (Chip Salzenberg) Newsgroups: comp.arch Subject: Re: Compilers and efficiency Message-ID: <283334FE.3C0@tct.com> Date: 17 May 91 01:42:21 GMT References: <1991May8.205106.6039@Stardent.COM> <28297C23.6984@tct.com> Organization: Teltronics/TCT, Sarasota, FL Lines: 26 According to rockwell@socrates.umd.edu (Raul Rockwell): >Chip Salzenberg: > That kind of problem is not "generally important" in that it does not > come up in the "general computing base". > >harumph. > >At work we have a hand-optimized assembly routine to convert from a >bit array to a list of indices. It gets used a lot. Mr. Rockwell's point is presumably that it would be more widely used were it more widely implemented. I cannot help but agree. Unfortunately, it is impossible to do cost/benefit analyses on possible universes of computation in which given tools are more or less available or efficient. A pity. I've been taken to task by another reader for my use of the term "general computing base", since I can't define it. That point is also well taken. Things are not as simple as I had believed... -- Brand X Industries Custodial, Refurbishing and Containment Service: When You Never, Ever Want To See It Again [tm] Chip Salzenberg ,