[gs-bugs] [Bug 690250] Performance slower than other PCL s/w

bugs.ghostscript.com-bugzilla-daemon at ghostscript.com bugs.ghostscript.com-bugzilla-daemon at ghostscript.com
Wed Mar 11 07:01:13 PDT 2009


http://bugs.ghostscript.com/show_bug.cgi?id=690250





------- Additional Comments From ray.johnston at artifex.com  2009-03-11 07:01 -------
hmm... The initclip does mean that we will be merging with an outer box that
does not intersect the complex path, but it makes it more difficult to use
the method of checking to see if the clip path being set is the same as the
one already being used.

Either we have to speed up the way we decompose the path into the 'shape' list
(preferred), or we need to keep paths and their shapes laying around in a cache
to avoid rebuilding them (at least the complicated paths), or both.

I suspect that using the general fill logic to decompose the clip path into
the shape list is part of the problem -- we could probably do a much faster
decomposition.

A bigger architectural change is to look at changing the 'shape' into a list
of trapezoids (not rectangles). Clipping/interesecting against trapezoids is
not that hard and we will be dealing with fewer elements on the list and
less decomposition.



------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.



More information about the gs-bugs mailing list