SSE4.1 slow

[peplegal]

Is there anything that can be done to improve performance for SSE4.1 code ? (without parallelization).

PRECISE_COVERAGE is disabled already...and I'm getting 14 milliSeconds for rendering about 85kTri.

Too close of the limit of 60 FPS (16.6 mS). An entire CPU core dedicated only for that.

Thanks anyway. This algorithm is really great ! The Test Rectangle function is incredibly fast !
.