<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div style="-webkit-text-size-adjust: auto;">Thanks so much. It works perfectly now. </div><div style="-webkit-text-size-adjust: auto;">May I suggest the following modification in pixman.h so that it is clear for all users ?</div><div style="-webkit-text-size-adjust: auto;"> </div><div>/*<o:p></o:p>* Important note: pixman colors are alpha-premultiplied<o:p></o:p>* and must conform: red <= alpha ; green <= alpha ; blue <= alpha<o:p></o:p>* So that for instance pure white with 0x80 alpha should<o:p></o:p> * be r=0x80, g=0x80, b=0x80, a=0x80<o:p></o:p>*/<o:p></o:p>struct pixman_color<o:p></o:p>{<o:p></o:p> uint16_t red;<o:p></o:p> uint16_t green;<o:p></o:p> uint16_t blue;<o:p></o:p> uint16_t alpha;<o:p></o:p>}; Thanks</div><div style="-webkit-text-size-adjust: auto;"> Le 12 juin 2014 à 17:01, <a href="mailto:soren.sandmann@gmail.com">soren.sandmann@gmail.com</a> (Søren Sandmann) a écrit : </div><blockquote type="cite" style="-webkit-text-size-adjust: auto;"><div>Eric Nicolas <<a href="mailto:erik.nicolas@gmail.com">erik.nicolas@gmail.com</a>> writes: <blockquote type="cite">Hello, Thanks for your answer. </blockquote><blockquote type="cite"> </blockquote><blockquote type="cite">Below is a small test program (works on little endian machines). </blockquote><blockquote type="cite"> </blockquote><blockquote type="cite">As is the result is already strange to me : </blockquote><blockquote type="cite"> in: r=20 g=20 b=80 a=FF </blockquote><blockquote type="cite"> out: r=90 g=90 b=C0 a=FF </blockquote><blockquote type="cite">Where as I expected the fill color #FFFFFFFF x the mask 0x80 = #FFFFFF80 </blockquote><blockquote type="cite">and so an output of #9090FFFF. </blockquote><blockquote type="cite"> </blockquote><blockquote type="cite">However, if I set the fill color to #FFFFFF80, the result seems really </blockquote><blockquote type="cite">wrong : </blockquote><blockquote type="cite">in: r=20 g=20 b=80 a=FF </blockquote><blockquote type="cite">out: r=98 g=98 b=E0 a=FF </blockquote><blockquote type="cite"> </blockquote><blockquote type="cite">I would expect fill x mask => #FFFFFF40 and thus out: #606060C0FF. </blockquote><blockquote type="cite"> </blockquote><blockquote type="cite">I especially do not understand how a lower alpha input color can end up in </blockquote><blockquote type="cite">a lighter output on the target image. </blockquote><blockquote type="cite"> </blockquote><blockquote type="cite">What I am doing wrong here ? </blockquote> The thing you are missing is that Pixman works with premultiplied alpha. That is, if the rgb values are R, G, and B, and the alpha value is A, then pixman will assume that pixels are stored as R * A, G * A, B * A, A and it will also generate pixels in this format. The reason for doing this is that it avoids divisions when compositing. When you set the fill value to 0xffffff80 pixman will interprete it as (1.0 / 0.5), (1.0 / 0.5), (1.0 / 0.5), 0.5 = (2, 2, 2, 0.5) ie., you get out-of-range color values, which is why the result will be lighter. Normally, the alpha value in premultiplied pixels should not be bigger than the color values, though it can occasionally be useful. Premultiplied alpha also means that when applying an alpha mask, it must be applied to all four channels, so in this case you get 0xffffff80 x 0x80 = 0x80808040 when applying the mask. Søren </div></blockquote></body></html>