2 * Floating point AAN DCT
3 * Copyright (c) 2003 Michael Niedermayer <michaelni@gmx.at>
5 * This library is free software; you can redistribute it and/or
6 * modify it under the terms of the GNU Lesser General Public
7 * License as published by the Free Software Foundation; either
8 * version 2 of the License, or (at your option) any later version.
10 * This library is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13 * Lesser General Public License for more details.
15 * You should have received a copy of the GNU Lesser General Public
16 * License along with this library; if not, write to the Free Software
17 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
19 * this implementation is based upon the IJG integer AAN DCT (see jfdctfst.c)
25 * Floating point AAN DCT
26 * @author Michael Niedermayer <michaelni@gmx.at>
36 # define SCALE(x) postscale[x]
41 //numbers generated by simple c code (not as accurate as they could be)
44 printf("#define B%d %1.20llf\n", i, (long double)1.0/(cosl(i*acosl(-1.0)/(long double)16.0)*sqrtl(2)));
47 #define B0 1.00000000000000000000
48 #define B1 0.72095982200694791383 // (cos(pi*1/16)sqrt(2))^-1
49 #define B2 0.76536686473017954350 // (cos(pi*2/16)sqrt(2))^-1
50 #define B3 0.85043009476725644878 // (cos(pi*3/16)sqrt(2))^-1
51 #define B4 1.00000000000000000000 // (cos(pi*4/16)sqrt(2))^-1
52 #define B5 1.27275858057283393842 // (cos(pi*5/16)sqrt(2))^-1
53 #define B6 1.84775906502257351242 // (cos(pi*6/16)sqrt(2))^-1
54 #define B7 3.62450978541155137218 // (cos(pi*7/16)sqrt(2))^-1
57 #define A1 0.70710678118654752438 // cos(pi*4/16)
58 #define A2 0.54119610014619698435 // cos(pi*6/16)sqrt(2)
59 #define A5 0.38268343236508977170 // cos(pi*6/16)
60 #define A4 1.30656296487637652774 // cos(pi*2/16)sqrt(2)
62 static FLOAT postscale[64]={
63 B0*B0, B0*B1, B0*B2, B0*B3, B0*B4, B0*B5, B0*B6, B0*B7,
64 B1*B0, B1*B1, B1*B2, B1*B3, B1*B4, B1*B5, B1*B6, B1*B7,
65 B2*B0, B2*B1, B2*B2, B2*B3, B2*B4, B2*B5, B2*B6, B2*B7,
66 B3*B0, B3*B1, B3*B2, B3*B3, B3*B4, B3*B5, B3*B6, B3*B7,
67 B4*B0, B4*B1, B4*B2, B4*B3, B4*B4, B4*B5, B4*B6, B4*B7,
68 B5*B0, B5*B1, B5*B2, B5*B3, B5*B4, B5*B5, B5*B6, B5*B7,
69 B6*B0, B6*B1, B6*B2, B6*B3, B6*B4, B6*B5, B6*B6, B6*B7,
70 B7*B0, B7*B1, B7*B2, B7*B3, B7*B4, B7*B5, B7*B6, B7*B7,
73 void ff_faandct(DCTELEM * data)
75 FLOAT tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7;
76 FLOAT tmp10, tmp11, tmp12, tmp13;
77 FLOAT z1, z2, z3, z4, z5, z11, z13;
83 for (i=0; i<8*8; i+=8) {
84 tmp0= data[0 + i] + data[7 + i];
85 tmp7= data[0 + i] - data[7 + i];
86 tmp1= data[1 + i] + data[6 + i];
87 tmp6= data[1 + i] - data[6 + i];
88 tmp2= data[2 + i] + data[5 + i];
89 tmp5= data[2 + i] - data[5 + i];
90 tmp3= data[3 + i] + data[4 + i];
91 tmp4= data[3 + i] - data[4 + i];
98 temp[0 + i]= tmp10 + tmp11;
99 temp[4 + i]= tmp10 - tmp11;
101 z1= (tmp12 + tmp13)*A1;
102 temp[2 + i]= tmp13 + z1;
103 temp[6 + i]= tmp13 - z1;
109 z5= (tmp10 - tmp12) * A5;
117 temp[5 + i]= z13 + z2;
118 temp[3 + i]= z13 - z2;
119 temp[1 + i]= z11 + z4;
120 temp[7 + i]= z11 - z4;
123 for (i=0; i<8; i++) {
124 tmp0= temp[8*0 + i] + temp[8*7 + i];
125 tmp7= temp[8*0 + i] - temp[8*7 + i];
126 tmp1= temp[8*1 + i] + temp[8*6 + i];
127 tmp6= temp[8*1 + i] - temp[8*6 + i];
128 tmp2= temp[8*2 + i] + temp[8*5 + i];
129 tmp5= temp[8*2 + i] - temp[8*5 + i];
130 tmp3= temp[8*3 + i] + temp[8*4 + i];
131 tmp4= temp[8*3 + i] - temp[8*4 + i];
138 data[8*0 + i]= lrint(SCALE(8*0 + i) * (tmp10 + tmp11));
139 data[8*4 + i]= lrint(SCALE(8*4 + i) * (tmp10 - tmp11));
141 z1= (tmp12 + tmp13)* A1;
142 data[8*2 + i]= lrint(SCALE(8*2 + i) * (tmp13 + z1));
143 data[8*6 + i]= lrint(SCALE(8*6 + i) * (tmp13 - z1));
149 z5= (tmp10 - tmp12) * A5;
157 data[8*5 + i]= lrint(SCALE(8*5 + i) * (z13 + z2));
158 data[8*3 + i]= lrint(SCALE(8*3 + i) * (z13 - z2));
159 data[8*1 + i]= lrint(SCALE(8*1 + i) * (z11 + z4));
160 data[8*7 + i]= lrint(SCALE(8*7 + i) * (z11 - z4));