Xref: utzoo comp.sys.ibm.pc.misc:7971 comp.sys.mac.misc:10274 Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!wuarchive!cec2!news From: jyp@wucs1.wustl.edu (Jerome Yvon Plun) Newsgroups: comp.sys.ibm.pc.misc,comp.sys.mac.misc Subject: Re: Open Letter to an MS-DOS User Message-ID: <1991Mar29.042136.8302@cec1.wustl.edu> Date: 29 Mar 91 04:21:36 GMT References: <1991Mar27.200613.4423@jato.jpl.nasa.gov> <1991Mar27.215516.20770@gn.ecn.purdue.edu> <1991Mar29.015546.24193@jato.jpl.nasa.gov> Sender: jyp@wucs1.wustl.edu Organization: Washington University, St. Louis MO Lines: 57 In article <1991Mar29.015546.24193@jato.jpl.nasa.gov> dave@elxr.jpl.nasa.gov writes: [things not related deleted] >>>Gee..I could if you tell me how much slower it runs, how much > >>AutoCAD illustrations (can you say FOREVER redraw on a 286), or... the list >>goes on. There are numerous applications where the speed of a PC is far >>inferior to the Mac, and these happen to be the same applications that I use > >I see. Most of my comparisons are a 386/25 to a Mac SE. Think about it. Well, as a Mac user, I am very proud to find out that you need to compare a PC with a 386 running at 25Mhz with a poor little SE having only a mere 6800 at 16MHz. Before I start getting hardware characteristics and other not very useful details if you are not playing with the raw hardware (give me a 500MHz 68070 without a computer around it and some software to use the beast and I will have a very fast paper holder :-), let me tell you that I really don't care about the fact that application X works slower on machine Y than machine Z. In fact: 1) I can buy a faster machine Z, 2) I can buy a different program that will have an inverse speed ratio, 3) A program contains more than one feature usually. Being slower for one doesn't imply to be slower for all. Can anybody dare to tell me that MS-DOS, Windows or OS/2 are _ALWAYS_ faster/slower (pick your favorite) than MacOS/Unix/Xenix/VMS/... 4) Speed is not the only characteristic of a program (a car can be a lot faster if it only has 4 tires, an engine and a steering wheel. You might beat everybody else but you might have some trouble in winter!). Similarly, would you prefer a compiler that spits out poor code very fast or a compiler that takes much longer but generates optimized code. You won't waste time waiting for your code but you might waste that time using the code later on. I would like both compilers, the fast to write run-once test routines and debugging, the slow to produce the final code. To finish, let me come back to this general trend of considering a particular machine has being above the others. Each religion tries the same approach and milleniums of religious behavior with respect to the others didn't (and are still not) leave much nice and pleasant things. Constructive and instructive informations are more interesting than "I know your computer stinks". If some Cray user were to show up and state that PCs are useless because they compute weather maps or fluid dynamics so slowly, everybody would "beat" on her/him. If the same user states that PCs are useless for supercomputing, most of us would agree but we might start looking at what computational intensive tasks a PC and a Cray can both attack. In one case, religious wars, in the other mutual exchange of information. Being an engineer, I would rather see the latter taking over the former. Just my 2 cents worth. Jerome Jerome Plun jyp@wucs1.wustl.edu // Tell me why is it so hard to say Dept of Computer Science // Brother don't you walk away Washington University // St. Louis, Mo // Hooters "Zig Zag"