[cairo] Write ups on Microsoft Avalon
cworth at east.isi.edu
Fri Jun 4 18:08:18 PDT 2004
On Mon, 17 May 2004 10:06:13 -0700, Bill Spitzak wrote:
> I strongly recommend that this be at the corners of the pixels, though I am
> not absolutely certain this is the CTM chosen for Cairo.
Yes, this is the case.
Or, for the sake of those who don't consider pixels as squares, one
might say that with the default CTM that integer grid points in user
space correspond to points offset by one-half device-space unit in each
dimension from the pixel sample points.
Clearly, thinking of pixels as squares makes this description simpler if
> In the other mail I also recommended that a
> special "0" case be provided that eliminates the need to put .5 in the
> coordinate, the reason is to discourage people from moving the CTM by .5 and
> messing up all the optimized drawing of rectangles.
A special "0" width like that of PostScript (where width 0 means the
thinnest possible line)? With modern printers the thinnest line is
already too thin to ever be desirable, and we're getting close to the
same situation with displays. I don't see how that would be helpful.
More information about the cairo