Re: gallium surface still maintained ?
"Enrico Weigelt, metux IT consult" <[email protected]>
| Newsgroups | gmane.comp.lib.cairo |
|---|---|
| Message-ID | <[email protected]> |
On 05.08.2016 09:35, Guillermo Rodriguez wrote: Hi, > From my own experience, as I said in a previous mail: > > === > I am using Cairo on embedded targets without a usable GPU, but a > reasonably powerful CPU, and actually using many of the above features > in my application, and it works just fine. > === Maybe you could give my twtk a little try on your hw - especially moving the windows tends to be slow (even on an i7 machine!). The interesting part is that the final composition is the slowest part (first the dirty widgets are rendered into a group, which is saved to a pattern - finall ending up w/ a big root pattern, which then is rendered with clipping to the actually changed regions). https://github.com/metux/twtk By the way: Where exactly is the clipping applied ? Will clipped-out objects still appear in the saved regions or do they get thrown out earlier ? Or turn the question around: on the actual rasterization (in image surface) are the clipped-out objects still calculated (but their pixels just not painted down to the surface) or are they skipped completely ? --mtx -- cairo mailing list [email protected] https://lists.cairographics.org/mailman/listinfo/cairo