before :

[INFO][???]> 12 : t = 1.099 ms, 0.9 MOPS, 3433.0 cycles/element
[INFO][???]> 12 : t = 1.249 ms, 0.8 MOPS, 3900.5 cycles/element (reference)
Maths_Cell_Streams_SinCos_Hi 8176/ 8176 [ OK ]
[INFO][???]> 11 : t = 0.541 ms, 1.9 MOPS, 1688.0 cycles/element
[INFO][???]> 11 : t = 0.555 ms, 1.8 MOPS, 1733.0 cycles/element (reference)
Maths_Cell_Streams_Cos_Hi 4088/ 4088 [ OK ]
[INFO][???]> 11 : t = 0.541 ms, 1.9 MOPS, 1689.9 cycles/element
[INFO][???]> 11 : t = 0.546 ms, 1.9 MOPS, 1705.0 cycles/element (reference)
Maths_Cell_Streams_Sin_Hi 4088/ 4088 [ OK ]


after:

[INFO][???]> 12 : t = 0.022 ms, 47.4 MOPS, 66.8 cycles/element
[INFO][???]> 12 : t = 1.212 ms, 0.8 MOPS, 3786.1 cycles/element (reference)
Maths_Cell_Streams_SinCos_Hi 8176/ 8176 [ OK ]
[INFO][???]> 11 : t = 0.019 ms, 53.4 MOPS, 59.2 cycles/element
[INFO][???]> 11 : t = 0.538 ms, 1.9 MOPS, 1681.6 cycles/element (reference)
Maths_Cell_Streams_Cos_Hi 4088/ 4088 [ OK ]
[INFO][???]> 11 : t = 0.018 ms, 55.3 MOPS, 57.5 cycles/element
[INFO][???]> 11 : t = 0.539 ms, 1.9 MOPS, 1682.7 cycles/element (reference)
Maths_Cell_Streams_Sin_Hi 4088/ 4088 [ OK ]
Maths_Cell_Float4_Compares 122/ 122 [ OK ]