Path: utzoo!utgpu!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!stanford.edu!leland.Stanford.EDU!leland.Stanford.EDU!fangchin From: fangchin@leland.Stanford.EDU (Chin Fang) Newsgroups: comp.sys.ibm.pc.hardware Subject: Re: 486SX - Intel now telling lies Message-ID: <1991Jun2.041512.29546@leland.Stanford.EDU> Date: 2 Jun 91 04:15:12 GMT References: <1991May29.230433.10095@maverick.ksu.ksu.edu> <1991May30.164751.16585@thyme.jpl.nasa.gov> <1991May31.183111.16505@ux1.cso.uiuc.edu> <1991Jun2.030215.11584@unixland.natick.ma.us> Sender: news@leland.Stanford.EDU (Mr News) Organization: AIR, Stanford University, CA 94305 USA Lines: 69 In article <1991Jun2.030215.11584@unixland.natick.ma.us>, bill@unixland.natick.ma.us (Bill Heiser) writes: |> In article <1991May31.183111.16505@ux1.cso.uiuc.edu> tmkk@uiuc.edu (K. Khan) writes: |> > |> >It means that Intel is scared silly of the industry's upcoming move to |> >the MIPS RISC chips, and is going to milk us for every single cent it |> >can while it is still able to do so. |> |> How does the i486 compare with the 20 - 30 MIPS of the SUN Sparc chips? |> How about the MIPS chips? |> Hi Bill :-) and everyone, I try to stay objective. Let's use i486 33Mhz as baseline for the CISC side, and IBM POWER archecture as the RISC side baseline. Assuming that everything runs in memory, then i486 33 Mhz offers you about 21 MIPS and less then 2 MegaFLOPS. Now IBM RISC POWERstation 320 20Mhz offers you 29.5 MIPS and 8.5 MFLOPS. So given enough memory to both, i486 33Mhz should be able to keep up for integer intensive operations. In fact, i486, due to it's more frugal use of memory and CISC, I suspect sometimes it may even win in a few areas (based on my past experience running 386 UNIX on my 386 box) In addition, If you haven't used a RISC machine, I assure you that the first time you are made aware of the disk file size and the output of size(1), you WILL feel Dismayed! SPARC executables so far in general are smaller than IBM POWER archecture's. But both are FAT! Implication is that if you use RISC, you need both more memory and disk space. The typical rule, 30% more than their CISC counterpart, is a understatement. The more, the better. 16 Megs seem to be abs necessary for IBM POWERstations vs 8 Megs for an. i486/386 box running X windows X11R4 on top of a 386 UNIX. Sun SPARC is no better in memory usuage, all our SUN SPARCs equipped with 8 Megs swap like crazy, really disgusting :-( Floating point operation wise, i486 is buried in the dust. No way it can win! even with all goodies from Cyrix, IIT, Weitek and friends. No way! Unless they also go RISC too. RISC 600 320 is the bottom end of POWER line and at 20Mhz already offers 8.5 MFLOPS, the newer 320H offer 13.5MFLOPS. CISC stuff will be *REALLY* hard pressed to squeeze out such a outstanding FP performace. I have a suspecision that Intel CISC machines can best be used as general purpose computers. They are popular, lots third party components, easy to configure, software availability is great, a combo really hard to beat by any RISC entry indeed. However, the prestigious Technical Workstations will always be dominated by RISC I guess. After all, most scientists/engineers crunch numbers, they are not going to do much with anything else. Remember even Intel had to venture into RISC and made this i860? And lot's people use it as graphic coprocessor or numerical coprocessor, why? RISC, => superior floating point performance (> 30 MegaFLOPS), that is. SUN SPARC 2 doesn't offer as good as floating point performance as either IBM POWER line or the newcomer HP 9000 series. The later is the newest screamer on the block and from what I know (from IBM inside info) that IBM Workstation Group is working overtime to try to top HP these days. (an IBM senior programmer told me that their face were green with envy these days seeing the blazingly fast X Window performance of HP 9000s :-) I love this :-) Competition is always good for customers :-) :-) :-) Later... Chin Fang Mechanical Engineering Department Stanford University fangchin@leland.stanford.edu