Re: FAST CLIM 2.0

• To: SWM@stony-brook.scrc.symbolics.com
• Subject: Re: FAST CLIM 2.0
• From: clarisse@iexist.att.com
• Date: Thu, 23 Dec 93 13:04:09 CST
• Cc: clim@BBN.COM

We have some good news about CLIM2.0 performance
at least on one or two vendors [on yet another vendor,
bugs prevent us from getting any performance feel].

What:
An application that defines its output record mechanism:
check the CAD-demo (and see previous discussion on
the mailing list).
Mostly graphical objects using low level drawing primitives,
multicolors, multifonts, patterns, transformations etc.
Each object IS its own output record.

The application uses memoizing for a few
things (color table, font table, and patterns).
It also uses static objects for graphical computation
(vectors, contexts etc.).

We found it necessary to memoize transformed patterns
(this means patterns derived from other patterns using
a finite set of rotations, and rounded up scaling)
as well as patterns based on their designs (keyed on background
and foreground colors).

With that in mind CLIM takes care of all the rest.
We know what the *rest* is since we build a similar
application on top of straight XLIB (CLX) in the past.

Results: CLIM wins, it's faster, slicker, simpler
(about 5 times less code) for the same type of functionality.
Drag and drop, refresh, run-time everything ALL is much faster.

Congratulations to the designers and implementors of the
CLIM versions that work!

Happy holidays and see you next year.

• Prev: Re: multiple frames and activities
• Next: Re: reducing time overhead of text display (in 1.1) Date: Wed, 22 Dec 1993 19:01-0500 From: "Bruce R. Miller" <miller@cam.nist.gov> Cc: clim@bbn.com
• Index: CLIM Mailing list (1-Jan-93 to 31-May-94)
• Thread: CLIM Mailing list Index (1-Jan-93 to 31-May-94)