Am 2014-03-20 23:22, schrieb Keith Packard:
I think this functions are worth to reimplement. eg horizontal lines
are
used a lot and very slow on mi.
Having sped up the underlying functions, they're not terrible at this
point:
2000000 reps @ 0.0035 msec (288000.0/sec): 500-pixel horizontal
line segment
But, yeah, doing a 'real' implementation would be awesome. For
comparison, fbdev does:
30000000 reps @ 0.0002 msec (4850000.0/sec): 500-pixel horizontal
line segment
Sorry, my bad, I wanted to talk about vertical lines:
10000000 trep @ 0.0046 msec (219000.0/sec): 500-pixel horizontal line
segment
2500000 trep @ 0.0119 msec ( 84100.0/sec): 500-pixel vertical line
segment
How are the interpolation requirement for a line? Is it possible to
achive them with an opengl quad?
zero-width line pixelization semantics are pretty weak and we should be
able to hit them with GL quads. Getting the 'notlast' cap semantic
working is about the only trick necessary.
It shouldn't be that hard:
in vec4 pos //x1,y1,x2,y2
in int notlast;
const float line_width = 1.0;
void main() {
vec2 direction = 0.5 * normalize(vec2(x2,y2) - vec2(x1,y1));
vec2 normal = vec2(-direction.y, direction.x) * line_width;
vec2 pos = (gl_VertexId&1==0) ? pos.xy - direction : notlast ? pos.zw
- direction : pos.zw + direction;
pos += (gl_VertexId&2==0) ? -normal : normal;
gl_Position = vec4(pos, 0.0, 1.0);
}
_______________________________________________
[email protected]: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
