home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!shady!kevin
- From: kevin@shady.UUCP (Kevin Smith)
- Newsgroups: biz.sco.opendesktop
- Subject: Re: Math coprocessor with ODT
- Message-ID: <69@shady.UUCP>
- Date: 30 Dec 92 21:54:51 GMT
- References: <1992Dec29.023225.23804@beaver.cs.washington.edu> <1992Dec29.190445.1878@allink.com>
- Organization: ShadeTree Software, Inc.
- Lines: 28
-
- In article <1992Dec29.190445.1878@allink.com> leonard@allink.com (Leonard Primak) writes:
- :>In article <1992Dec29.023225.23804@beaver.cs.washington.edu> drupp@cs.washington.edu (Douglas Rupp) writes:
- :>>Can anyone tell me how much adding a math coprocessor will help response
- :>>time in ODT? I have a 386/40 with 64K cache and an Paradise 8514/A video
- :>>card. I do alot of heavy duty graphics development using a commercial
- :>>GKS binding to X-windows. I am wondering just how much floating point
- :>>is used in X? What response I get now is not objectionable, but maybe I'm
- :>>missing out.
- :>
- :> You are missing out. ODT's X server uses floating-point operations
- :>for many graphics operations. I checked this out with xeyes, if you don't
- :>have a math co-processor, the eyeball movement is jumpy and slow, but when
- :>I added a MathCo to my 386/33, the eyeball movement was smooth, and so was
- :>ICO's, and other animation stuff.
- :>
- :> leonard@allink.com
- :
-
- I was led to believe that the X server itself uses mostly integer math.
- I did not notice any particular performance increase in my usual stuff
- (xterms, xdt, zmail, moving, redrawing,...). I suspect it is application
- dependant. Xeyes and ico, themselvs, may use floating point to compute the
- positions but X may not use floating point to do the actual drawing.
- --
- | Email - !shady!kevin uunet!shady!kevin kevin%shady@uunet.uu.net
- Kevin Smith | Voice - (+1) (908) 874-7980
- | Mail - ShadeTree Software, Inc., 192 Capricorn Dr. #10,
- | Somerville, NJ 08876, USA
-