On Mon, 2012-04-30 at 13:05 +0200, peter sikking wrote:
[...]
> > 4. Its implementation is very simple and well designed. This is the
> > most important things for developers!
>
> ...but one million users do not care.
They do if people stop working on GIMP.
So, there are several parts here.
(1) the user experience;
(2) the technology (the C/C++ library) and how easy it
is to maintain and enhance;
(3) the API between GIMP and the technology.
I don't think there's any question that the mypaint brush editor user
experience could easily be improved. Heck, it would be improved by
having some categories, whether as tabs or just headings/groups --
colour, shape, opacity, motion, and so on.
I also don't think GIMP's brush interface is perfect. It's a lot better
than it used to be, but it's still not very streamlined.
It's for sure true that painting in GIMP is slower than it used to be.
Clearing the Undo History every now and then gives a dramatic speedup,
at last with the older 2.7 preview I'm using, and profiling suggests
there's an O(n^2) list insertion going on, although it's hard to
understand why that shouldn't be a matter of microseconds, not a
noticeable fraction of a second.
The gimp brush outline rendering-in-the-idle-loop also makes painting
_feel_ much slower than it is.
I remember dealing with this in the past in another application by
discarding motion events if the pointer got too far behind. It meant
that the drawing became less accurate, but once the pointer outline
isn't where you're drawing, there's no accuracy in any case.
Getting a shared core with MyPaint might be really interesting - and the
user interface doesn't need to be the same in both programs.
Liam
--
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/
Pictures from old books: http://fromoldbooks.org/
The barefoot ankh.
_______________________________________________
gimp-developer-list mailing list
gimp-developer-list@xxxxxxxxx
http://mail.gnome.org/mailman/listinfo/gimp-developer-list
