From: twpierce@unix.amherst.edu (Tim Pierce) Subject: 486SXs as Unix Iron? Date: Tue, 27 Oct 1992 16:52:38 GMT
I'm looking for a machine on which to run Linux in the $1500-$2000
range. I was looking at a lot of 486SX boxes, until I ran across Eric
Raymond's PC Unix hardware buyer's guide, which includes the following
paragraph:
> Don't bother with SX machines. Under UNIX the 16-bit bus-to-CPU path
> can choke your throughput. Anyway there's not enough of a 386DX
> premium to matter in the desktop market any more (laptops are a whole
> 'nother story). The 486SX is even worse, a stupid marketing crock
> with no technical justification whatsoever. It's a 486DX with the
> floating-point unit missing; the difference *will* bite you in
> unobvious ways, for example if you use X which does a lot of FP for
> graphics. (One respondent opined that "SX" is Intel-internal code for
> "sucks".)
Now I'll grant you that the 486SX is quite the marketing gimmick.
However, with the 8K on-chip cache (and on-board cache), a 486SX-25 is
supposed to perform about as well as a 386-33. Part of the reason I'm
interested in the 486SX is for its resale value, and I'm not keen on a
DX because I want to conserve my money for memory and disk.
My purpose in owning a Linux PC is to do kernel hacking and associated
computer science work. I have no interest in running X or any other
windowing system at this time (in fact, am trying to find vendors who
will sell low-end graphics cards and monitors). So it seems like a
486SX is not a bad idea for what I have in mind. Am I mistaken, or is
the above simply a highly subjective portrayal of SX CPUs?
-- ____ Tim Pierce / "You are just naive and repressed because \ / twpierce@unix.amherst.edu / penis envy is here and it's now and it's \/ (BITnet: TWPIERCE@AMHERST) / all around you." -- Neal C. Wickham