branches/koffice/1.6/koffice/krita/plugins/filters

[Boudewijn Rempt]

On Thu, 7 Sep 2006, Cyrille Berger wrote:

> > SVN commit 581621 by coppens:
> >
> > All of these don't work sufficiently (IMO) with the adjustment layers. So
> > disable them. I'm pretty sad about some of them (oilpaint, wave), but as
> > long as they don't work, they'll be disabled. If people have suggestions on
> > fixing any of them, feel free to do so, I'd be very happy to see them
> > reappear.
> 
> We have to rethink the filter architecture a little bit for adjustement layers 
> (and for the preview too), but first we need to identify the list of types of 
> layers: 
> 1) filter who works only on one pixels (invert, convolution, wave...)
> 2) filter who works only on one pixels but needs global statistics (color 
> transfert)
> 3) filter who need all the image all the time (wavelet noise reduction)
> 
> Did I miss something ?

Some filters act on only one pixel, but need information from others, like the
convolution filters.

> My opinion is that all three should be available with the adjustement layer, 
> eventually with a warning of slowliness for type 3.
> 
> Most of type 1 are allready available. Once CoW tiles are available, all of 
> them would be possible.
> 
> For type 2, there is two possibilities:
> - we consider that the statistics aren't much changed by each strokes, and we 
> only recompute them occasionally

Strokes? That's painting with filters, right?

> - or, this can work at least for fast color transfert and for 
> brightness/contrast, somehow we give the area to apply to the filter, and he 
> removes the old statistics (rereading the image) and add the new one, and can 
> process on the new pixels for a small computational cost, it works best, but 
> I don't know if it will covers all case of type 2
> 
> As for type 3, we can use a timer before the last stroke and the update, 
> or(and ?) have the filter be applied in background.
> 

Photoshop handles this kind of thing by showing something very simplified and 
only calculating things on mouse-up.

> Note: for random filters, I think the best solution for them is to use a seed 
> base random generator, even if the end results is correct, painting 
> get "annimated" with them
> 
>