Survex 0.95 prerelease 7 and xcaverot's rubber band.

[Mark Shinwell]

On Fri, Oct 27, 2000 at 11:44:45AM +0100, Olly Betts wrote:
> In message <20001027104553.H4554@mrs30.quns.cam.ac.uk>, Mark Shinwell writes:
> >On Fri, Oct 27, 2000 at 10:38:28AM +0100, Olly Betts wrote:
> >> In message <39F94BBC.97EF62A4@uts.edu.au>, Michael Lake writes:
> >> >Olly Betts wrote:
> >> >> Aven avoids this problem by purging the event queue - Xcaverot should
> >> >> probably do the same.
> >> 
> >> Thinking some more, rather than purging the queue it probably ought to read
> >> out any pending events (mouse movements, key presses, etc) and apply them
> >> all before redrawing.
> >
> >However, I tried that (in addition to quite a lot of variations) and under
> >GTK it's just too slow.
> 
> I was actually thinking of xcaverot.  But any idea why it is too slow with
> GTK?  I have trouble visualising what could take the time.

Well it has all the crud in the way in GDK with removing events and
copying structures, etc.  I think even raw Xlib is slow this way too;
XFlush or whatever I used is very much quicker (it's probably just a
pointer assignment or something).

> >The code there at the moment is vulgar, but at least
> >it works (but may not with future GTK+ releases).
> 
> That's a non-issue at present, since aven won't compile with the gtk-- 1.2
> C++ bindings anyway (prod prod).

I shall have to get onto that soon.

Mark

-- 
Mark Shinwell -- http://mrs30.quns.cam.ac.uk -- Mark.Shinwell@cl.cam.ac.uk
Theory and Semantics Group, University of Cambridge Computer Laboratory