On Mon, Apr 11, 2016 at 12:35 PM, Søren Sandmann <soren.sandm...@gmail.com> wrote:
> On Mon, Apr 11, 2016 at 2:43 PM, Bill Spitzak <spit...@gmail.com> wrote: > > >> I feel this can be fixed. It is already correct for subsample_bits==0. >> Since both the filter generator and filtering code would be changed in the >> same version, only programs that generate their own filters would actually >> be incompatible. And as you point out the error is very tiny for large >> numbers of subsamples, and Cairo is already using an excessively large >> number of subsamples (likely because I was trying to remove the difference >> between identity filters and nearest filtering, and did not realize this >> was the underlying problem). >> > > It is technically an ABI break, and cairo 1.14 did ship with a > copied-and-pasted filter generator that assumes the current subpixel > positioning. > > But yeah, maybe we can just ignore that, if we make sure a there is a > cairo 1.14.x update that uses pixman_filter_create_separate_convolution() > instead of the copied-and-pasted filter generator. > > Other than that, the fix should be straight-forward enough. > As I wrote that Cairo patch I can state that I think it would be perfectly fine to make this change. The subsamples are set unnaturally high in that patch and hide the problem. Also it is possible my filter generator is producing centered samples so this could be an improvement, need to check. In addition, other than the (abandoned) idea of getting good/best into pixman, this patch series is also designed so that Cairo can use the pixman filter generator rather than it's own.
_______________________________________________ Pixman mailing list Pixman@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/pixman
