From: James Henrickson (ujlh@pool.info.sunyit.edu)
Date: 09/03/92


From: ujlh@pool.info.sunyit.edu (James Henrickson)
Subject: Re: Does a 387 help Linux's performance?
Date: 3 Sep 1992 17:09:57 GMT

In article <1992Sep1.201149.29493@serval.net.wsu.edu> hlu@poly2.eecs.wsu.edu (H.J. Lu) writes:
>In article <1992Sep1.160537.11089@afterlife.ncsc.mil>, swaliff@afterlife.ncsc.mil (Steve Aliff) writes:
>|> I have a 33MHz 386DX. Would it improve my system performance to add
>|> a 387? I plan to run X11 on top of Linux if that makes any
>|> difference.
>|> --
>|> Steve Aliff (swaliff@afterlife.ncsc.mil [144.51.1.1])
>
>Get a 387 if you consider X11.
>
>H.J.

I asked this before, but got no responses so....

I have a 386SX-25 that will act as an X server. I have a 386DX-33 that
will run most of the X clients. If I were to get ONE math coprocessor,
which machine would benefit the most?

Or maybe nobody understood the way I worded it...

Where is most of the floating point taking place, in the server or the
clients?

-- 
Jim H.
*
* James L. Henrickson
* ujlh@sunyit.edu        "Some day I might have a real .signature!"