Re: Question on the CAD-Demo

• To: clarisse@iexist.att.com, york@parc.xerox.com, SWM@stony-brook.scrc.symbolics.com
• Subject: Re: Question on the CAD-Demo
• From: Scott McKay <SWM@stony-brook.scrc.symbolics.com>
• Date: Mon, 22 Nov 1993 17:06 -0500
• Cc: clim@BBN.COM
• In-Reply-To: <9311221614.AA24344@tenet.lab5523>

    Date: Mon, 22 Nov 1993 11:14 EST
    From: clarisse@iexist.att.com

    William, Scott:

    Thanks to both of you for your replies.

No problem.

    For a cad like application containing several hundred
    graphical objects, I am not sure this approach
    is efficient enough. I saw some mail on the CLIM mailing list
    about the use of special tree structure for output records
    (I assumed to get efficient redisplay, highlighting,
    and fast mouse selections).

As Bill said, the CAD demo is just meant to show the possibilities,
not to be wonderful itself.  CLIM does provide an R-tree output
record, which does indeed have efficient access (order log N), but it
has a higher algorithmic complexity.  So R-trees won't always be
faster than, say, coordinate sorted sets, but for highly overlapping
output like you would get from a real CAD application, R-trees will
have better performance.

I agree with everything else he said.

    P.S. By the way, the CLIM 2.0 implementations I have seen are
	 VERY impressive. Congratulations to the designers and implementors,
	 it's high quality, flexible, efficient, quite robust, highly productive.
	 I only wish I had switched from 1.1 to CLIM 2.0 earlier!

Thanks.

    This reflects my personal opinion, not necessarily my employer's...

I hope someday they share your opinion.

• Re: Question on the CAD-Demo
• From: clarisse@iexist.att.com

• Prev: Question on the CAD-Demo
• Next: More Pull-Down Questions
• Index: CLIM Mailing list (1-Jan-93 to 31-May-94)
• Thread: CLIM Mailing list Index (1-Jan-93 to 31-May-94)