Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!mcsun!cernvax!chx400!chx400!urz.unibas.ch!ganter From: ganter@urz.unibas.ch Newsgroups: comp.sys.ibm.pc.hardware Subject: Re: Various 486 questions Message-ID: <1991Feb21.142350.1389@urz.unibas.ch> Date: 21 Feb 91 13:23:50 GMT References: <708.27c232c6@malibu.sedd.trw.com> Organization: University of Basel, Switzerland Lines: 41 In article <708.27c232c6@malibu.sedd.trw.com>, rose@malibu.sedd.trw.com writes: > 1) I have seen a lot of recent price drops in 486/25 machines. ..stuff deleted... > Is there any specific > reason that prices have dropped so drastically recently? And > any opinion on whether these prices will continue down or start > back up? I think, there are still some 486/25 with B9 mask around. These are told to have bugs, that prohibit multiprocessor use without external logic (delivered by Intel with the processor). The main problem wouldn't be the actual processor, but an eventual newer one (connect pin x with ground ..., etc). I don't think, that a C1 or higher would run in such an environement. Because of these bugs, the machines containing them are sometimes much cheaper. If You are kind with Your B9-486 all that stuff perhaps doesn't matter for You. But I'd be carefull. Another reason is certainly the 486/33 entering the market now. > Does the 486 machine truly contain a built in math co-processor? > Will "any" software that requires or can use a co-processor > be able to run on the 486? If the above is true: then why does > Tri-Star advertise a "CAD" system which contains an 80387? > This was in the computer shopper. The system contains a digitizer > in addition to the basic computer. Yes, the 486 contains a full 387 coprocessor. The 486 built in FPU is faster (i.e. fever clock counts for some instructions) and has the same parallel facilites as the 386/387 processors have. Perhaps the 387 is meant as the built in 387 (Or do they use a second one only for the digitizer part ?). Most 486 boards have a second socket for the Weitek 4167 FPU (not compatible to the 387, but faster). Robert Robert Ganter University of Basel Switzerland ganter@urz.unibas.ch