Scaling algorithms in Krita Bad (at least at downscaling)

[Boudewijn Rempt]

On Saturday 11 November 2006 20:50, Sander Koning wrote:
> Bart Coppens wrote on 2006-11-11 19:07 +0100 regarding Re: Scaling 
algorithms in Krita Bad (at least at downscaling):
> > As said by pippin on #koffice:
> > <pippin> interpolation is wrong when scaling down, I think I've stated
> > that here two or three times,..
> > <pippin> bilinear uses 4 source pixels/destination pixel, when scaling to
> > 10% that means discarding 96% of the image data (using 4 instead of at
> > <pippin> interpolation only works for filling in data between existing
> > data points, the other form of resampling is usually called decimation
> > (though that would be the nearest neighbour approach if the word is taken
> > literally) <pippin> decimation  n : destroying or killing a large part of
> > the population (literally every tenth person as chosen by lot)
> >
> :)
>
> Resampling by just taking the average of points 1 to N, then N+1 to 2N,
> etc., should already be a better solution.

So what can and should we do, preferably before 1.6.1 gets retagged?
-- 
Boudewijn Rempt 
http://www.valdyas.org/fading/index.cgi
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 191 bytes
Desc: not available
Url : http://mail.kde.org/pipermail/kimageshop/attachments/20061112/6073cee8/attachment.pgp