summaryrefslogtreecommitdiff
path: root/vp8/encoder/dct.c
diff options
context:
space:
mode:
Diffstat (limited to 'vp8/encoder/dct.c')
-rw-r--r--vp8/encoder/dct.c209
1 files changed, 51 insertions, 158 deletions
diff --git a/vp8/encoder/dct.c b/vp8/encoder/dct.c
index 5207e39c4..b5a11ae34 100644
--- a/vp8/encoder/dct.c
+++ b/vp8/encoder/dct.c
@@ -1,172 +1,64 @@
/*
- * Copyright (c) 2010 The VP8 project authors. All Rights Reserved.
+ * Copyright (c) 2010 The WebM project authors. All Rights Reserved.
*
- * Use of this source code is governed by a BSD-style license and patent
- * grant that can be found in the LICENSE file in the root of the source
- * tree. All contributing project authors may be found in the AUTHORS
- * file in the root of the source tree.
+ * Use of this source code is governed by a BSD-style license
+ * that can be found in the LICENSE file in the root of the source
+ * tree. An additional intellectual property rights grant can be found
+ * in the file PATENTS. All contributing project authors may
+ * be found in the AUTHORS file in the root of the source tree.
*/
#include <math.h>
-
-static const short dct_matrix2[4][4] =
-{
- { 23170, 30274, 23170, 12540 },
- { 23170, 12540, -23170, -30274 },
- { 23170, -12540, -23170, 30274 },
- { 23170, -30274, 23170, -12540 }
-};
-
-static const short dct_matrix1[4][4] =
-{
- { 23170, 23170, 23170, 23170 },
- { 30274, 12540, -12540, -30274 },
- { 23170, -23170, -23170, 23170 },
- { 12540, -30274, 30274, -12540 }
-};
-
-
-#define _1STSTAGESHIFT 14
-#define _1STSTAGEROUNDING (1<<( _1STSTAGESHIFT-1))
-#define _2NDSTAGESHIFT 16
-#define _2NDSTAGEROUNDING (1<<( _2NDSTAGESHIFT-1))
-
-// using matrix multiply
void vp8_short_fdct4x4_c(short *input, short *output, int pitch)
{
- int i, j, k;
- short temp[4][4];
- int sumtemp;
- pitch >>= 1;
-
- for (i = 0; i < 4; i++)
- {
- for (j = 0; j < 4; j++)
- {
- sumtemp = 0;
-
- for (k = 0; k < 4; k++)
- {
- sumtemp += input[i*pitch+k] * dct_matrix2[k][j];
-
- }
-
- temp[i][j] = (short)((sumtemp + _1STSTAGEROUNDING) >> _1STSTAGESHIFT);
- }
- }
-
-
- for (i = 0; i < 4; i++)
- {
- for (j = 0; j < 4; j++)
- {
- sumtemp = 0;
-
- for (k = 0; k < 4; k++)
- {
- sumtemp += dct_matrix1[i][ k] * temp[k][ j];
- }
-
- output[i*4+j] = (short)((sumtemp + _2NDSTAGEROUNDING) >> _2NDSTAGESHIFT);
- }
- }
-
-}
-
-
-void vp8_short_fdct8x4_c(short *input, short *output, int pitch)
-{
- vp8_short_fdct4x4_c(input, output, pitch);
- vp8_short_fdct4x4_c(input + 4, output + 16, pitch);
-}
-
-
-static const signed short x_c1 = 60547;
-static const signed short x_c2 = 46341;
-static const signed short x_c3 = 25080;
-
-void vp8_fast_fdct4x4_c(short *input, short *output, int pitch)
-{
int i;
int a1, b1, c1, d1;
- int a2, b2, c2, d2;
short *ip = input;
-
short *op = output;
- int temp1, temp2;
for (i = 0; i < 4; i++)
{
- a1 = (ip[0] + ip[3]) * 2;
- b1 = (ip[1] + ip[2]) * 2;
- c1 = (ip[1] - ip[2]) * 2;
- d1 = (ip[0] - ip[3]) * 2;
-
- temp1 = a1 + b1;
- temp2 = a1 - b1;
-
- op[0] = ((temp1 * x_c2) >> 16) + temp1;
- op[2] = ((temp2 * x_c2) >> 16) + temp2;
-
- temp1 = (c1 * x_c3) >> 16;
- temp2 = ((d1 * x_c1) >> 16) + d1;
+ a1 = ((ip[0] + ip[3])<<3);
+ b1 = ((ip[1] + ip[2])<<3);
+ c1 = ((ip[1] - ip[2])<<3);
+ d1 = ((ip[0] - ip[3])<<3);
- op[1] = temp1 + temp2;
-
- temp1 = (d1 * x_c3) >> 16;
- temp2 = ((c1 * x_c1) >> 16) + c1;
+ op[0] = a1 + b1;
+ op[2] = a1 - b1;
- op[3] = temp1 - temp2;
+ op[1] = (c1 * 2217 + d1 * 5352 + 14500)>>12;
+ op[3] = (d1 * 2217 - c1 * 5352 + 7500)>>12;
ip += pitch / 2;
op += 4;
- }
+ }
ip = output;
op = output;
-
for (i = 0; i < 4; i++)
{
-
a1 = ip[0] + ip[12];
b1 = ip[4] + ip[8];
c1 = ip[4] - ip[8];
d1 = ip[0] - ip[12];
+ op[0] = ( a1 + b1 + 7)>>4;
+ op[8] = ( a1 - b1 + 7)>>4;
- temp1 = a1 + b1;
- temp2 = a1 - b1;
-
- a2 = ((temp1 * x_c2) >> 16) + temp1;
- c2 = ((temp2 * x_c2) >> 16) + temp2;
-
- temp1 = (c1 * x_c3) >> 16;
- temp2 = ((d1 * x_c1) >> 16) + d1;
-
- b2 = temp1 + temp2;
-
- temp1 = (d1 * x_c3) >> 16;
- temp2 = ((c1 * x_c1) >> 16) + c1;
-
- d2 = temp1 - temp2;
-
-
- op[0] = (a2 + 1) >> 1;
- op[4] = (b2 + 1) >> 1;
- op[8] = (c2 + 1) >> 1;
- op[12] = (d2 + 1) >> 1;
+ op[4] =((c1 * 2217 + d1 * 5352 + 12000)>>16) + (d1!=0);
+ op[12] = (d1 * 2217 - c1 * 5352 + 51000)>>16;
ip++;
op++;
}
}
-void vp8_fast_fdct8x4_c(short *input, short *output, int pitch)
+void vp8_short_fdct8x4_c(short *input, short *output, int pitch)
{
- vp8_fast_fdct4x4_c(input, output, pitch);
- vp8_fast_fdct4x4_c(input + 4, output + 16, pitch);
+ vp8_short_fdct4x4_c(input, output, pitch);
+ vp8_short_fdct4x4_c(input + 4, output + 16, pitch);
}
void vp8_short_walsh4x4_c(short *input, short *output, int pitch)
@@ -177,17 +69,18 @@ void vp8_short_walsh4x4_c(short *input, short *output, int pitch)
short *ip = input;
short *op = output;
+
for (i = 0; i < 4; i++)
{
- a1 = ip[0] + ip[3];
- b1 = ip[1] + ip[2];
- c1 = ip[1] - ip[2];
- d1 = ip[0] - ip[3];
-
- op[0] = a1 + b1;
- op[1] = c1 + d1;
- op[2] = a1 - b1;
- op[3] = d1 - c1;
+ a1 = ((ip[0] + ip[2])<<2);
+ d1 = ((ip[1] + ip[3])<<2);
+ c1 = ((ip[1] - ip[3])<<2);
+ b1 = ((ip[0] - ip[2])<<2);
+
+ op[0] = a1 + d1 + (a1!=0);
+ op[1] = b1 + c1;
+ op[2] = b1 - c1;
+ op[3] = a1 - d1;
ip += pitch / 2;
op += 4;
}
@@ -197,25 +90,25 @@ void vp8_short_walsh4x4_c(short *input, short *output, int pitch)
for (i = 0; i < 4; i++)
{
- a1 = ip[0] + ip[12];
- b1 = ip[4] + ip[8];
- c1 = ip[4] - ip[8];
- d1 = ip[0] - ip[12];
-
- a2 = a1 + b1;
- b2 = c1 + d1;
- c2 = a1 - b1;
- d2 = d1 - c1;
-
- a2 += (a2 > 0);
- b2 += (b2 > 0);
- c2 += (c2 > 0);
- d2 += (d2 > 0);
-
- op[0] = (a2) >> 1;
- op[4] = (b2) >> 1;
- op[8] = (c2) >> 1;
- op[12] = (d2) >> 1;
+ a1 = ip[0] + ip[8];
+ d1 = ip[4] + ip[12];
+ c1 = ip[4] - ip[12];
+ b1 = ip[0] - ip[8];
+
+ a2 = a1 + d1;
+ b2 = b1 + c1;
+ c2 = b1 - c1;
+ d2 = a1 - d1;
+
+ a2 += a2<0;
+ b2 += b2<0;
+ c2 += c2<0;
+ d2 += d2<0;
+
+ op[0] = (a2+3) >> 3;
+ op[4] = (b2+3) >> 3;
+ op[8] = (c2+3) >> 3;
+ op[12]= (d2+3) >> 3;
ip++;
op++;
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-11 13:23:08 +0000