[Mesa-dev] [PATCH 3/9] radeonsi: improve quality of 16 sample locations

Nicolai Hähnle nhaehnle at gmail.com
Mon May 7 11:53:29 UTC 2018


On 02.05.2018 06:13, Marek Olšák wrote:
> From: Marek Olšák <marek.olsak at amd.com>
> 
> This results in better 16x and 8x quality when using these locations.
> Verified with the piglit MSAA accuracy test.

How did you come up with those? :)

Reviewed-by: Nicolai Hähnle <nicolai.haehnle at amd.com>


> ---
>   src/gallium/drivers/radeonsi/si_state_msaa.c | 4 ++--
>   1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/src/gallium/drivers/radeonsi/si_state_msaa.c b/src/gallium/drivers/radeonsi/si_state_msaa.c
> index b6504d53a2a..5066c31319e 100644
> --- a/src/gallium/drivers/radeonsi/si_state_msaa.c
> +++ b/src/gallium/drivers/radeonsi/si_state_msaa.c
> @@ -100,26 +100,26 @@ static const uint32_t sample_locs_8x[] = {
>   	FILL_SREG( 0, 0,   0, 0,   0, 0,   0, 0),
>   };
>   static const uint64_t centroid_priority_8x = 0x3542017635420176ull;
>   
>   /* 4x and 16x MSAA
>    * (the first 4 locations happen to be optimal for 4x MSAA, better than
>    *  the standard DX 4x locations)
>    */
>   static const uint32_t sample_locs_4x_16x[] = {
>   	FILL_SREG(-5,-2,   5, 3,  -2, 6,   3,-5),
> -	FILL_SREG(-7,-8,   1, 1,  -6, 4,   7,-4),
> +	FILL_SREG(-6,-7,   1, 1,  -6, 4,   7,-3),
>   	FILL_SREG(-1,-3,   6, 7,  -3, 2,   0,-7),
>   	FILL_SREG(-4,-6,   2, 5,  -8, 0,   4,-1),
>   };
>   static const uint64_t centroid_priority_4x = 0x2310231023102310ull;
> -static const uint64_t centroid_priority_16x = 0x497ec6b231d0fa85ull;
> +static const uint64_t centroid_priority_16x = 0x49e7c6b231d0fa85ull;
>   
>   static void si_get_sample_position(struct pipe_context *ctx, unsigned sample_count,
>   				   unsigned sample_index, float *out_value)
>   {
>   	const uint32_t *sample_locs;
>   
>   	switch (sample_count) {
>   	case 1:
>   	default:
>   		sample_locs = &sample_locs_1x;
> 


-- 
Lerne, wie die Welt wirklich ist,
Aber vergiss niemals, wie sie sein sollte.


More information about the mesa-dev mailing list