Review Request: Set Antialiasing in Kig

Maurizio Paolini paolini at dmf.unicatt.it
Sat Apr 14 21:17:49 UTC 2012


On Sat, Apr 14, 2012 at 06:40:31PM -0000, Albert Astals Cid wrote:
> I am not so sure this is wanted, antialiasing makes things thicker meaning that if the line had to be 1px wide it might be wider, and for a program that tries to be an exact representation of math this may not be desirable. I'd appreciate if you could attach a few screenshots of various geometrical objects with and without antialias so we could see the difference

Let me argue a little bit: if exactness is what we want, then 1px is just as bad as 2px
(and I think that antialiasing is not much more than 2px)... On the other hand, I personally often use
kig to produce pictures to include in papers, and in that case having nice pictures
is important.

Perhaps possible performance loss should be taken into account, but I have no idea of the impact of
dealing with antialiasing.
There are interesting constructions (mainly involving fractals and that can be obtained mainly
by using the external scripting pykig) that require the construction of a considerable amount
of geometric objects (segments, typically, or circular arcs in examples involving tessellations
in hyperbolic geometry) that could be tested.
I have a couple of examples in http://dmf.unicatt.it/~paolini/pykig-examples/, in particular

http://dmf.unicatt.it/~paolini/pykig-examples/poincare/disco/heptagonal_tessellation_disk.kpy

that can be opened with "pykig.py heptagonal_tessellation_disk.kpy" is a tessellation of the
hyperbolic plane (disk model).
It requires the presence of "poincare_disk.kpyh" in the same directory:
http://dmf.unicatt.it/~paolini/pykig-examples/poincare/disco/poincare_disk.kpyh

a fractal can be found in

http://dmf.unicatt.it/~paolini/pykig-examples/fractals/sierpinski.kpy

Unfortunately I do not have at the moment the possibility to test the patch myself...

Maurizio
 
> - Albert Astals Cid
> 
> 
> On April 13, 2012, 11:48 a.m., David Narváez wrote:
> > 
> > -----------------------------------------------------------
> > This is an automatically generated e-mail. To reply, visit:
> > http://git.reviewboard.kde.org/r/104584/
> > -----------------------------------------------------------
> > 
> > (Updated April 13, 2012, 11:48 a.m.)
> > 
> > 
> > Review request for KDE Edu.
> > 
> > 
> > Description
> > -------
> > 
> > Pretty straightforward, the question is actually if anybody disagrees. The one issue I would be concerned about is performance, but I would expect this not to be an issue in modern computers, yet I don't have any stress case to test this against.
> > 
> > 
> > This addresses bug 122376.
> >     http://bugs.kde.org/show_bug.cgi?id=122376
> > 
> > 
> > Diffs
> > -----
> > 
> >   misc/kigpainter.cpp 18d4396 
> > 
> > Diff: http://git.reviewboard.kde.org/r/104584/diff/
> > 
> > 
> > Testing
> > -------
> > 
> > Compare drawing a large circle with and without this patch to see antialiasing in action.
> > 
> > 
> > Thanks,
> > 
> > David Narváez
> > 
> >
> 

> _______________________________________________
> kde-edu mailing list
> kde-edu at mail.kde.org
> https://mail.kde.org/mailman/listinfo/kde-edu



More information about the kde-edu mailing list