devs/foxbox
1
0
Fork 0
forked from len0rd/rockbox
Code Pull requests Activity

Merged in first part of ffmpeg accuracy improvments. No change on my samples, but brings us closer to ffmpeg's source and clears the way for additional bug fixes.

Browse source 
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@13804 a1c6a512-1295-4272-9138-f99709370657
This commit is contained in:
Michael Giacomelli 2007-07-06 02:44:58 +00:00
parent e0473717e5
commit 2f1da8d24a
2 changed files with 18 additions and 38 deletions
Download patch file Download diff file Expand all files Collapse all files

1
apps/codecs/libwma/wmadec.h
View file

@ -124,6 +124,7 @@ typedef struct WMADecodeContext
int block_pos; /* current position in frame */
uint8_t ms_stereo; /* true if mid/side stereo mode */
uint8_t channel_coded[MAX_CHANNELS]; /* true if channel is coded */
int exponents_bsize[MAX_CHANNELS]; // log2 ratio frame/exp. length
fixed32 exponents[MAX_CHANNELS][BLOCK_MAX_SIZE];
fixed32 max_exponent[MAX_CHANNELS];
int16_t coefs1[MAX_CHANNELS][BLOCK_MAX_SIZE];

55
apps/codecs/libwma/wmadeci.c
View file

@ -1192,6 +1192,7 @@ int wma_decode_init(WMADecodeContext* s, asf_waveformatex_t *wfx)
return 0;
}
#if 0
/* interpolate values for a bigger or smaller block. The block must
have multiple sizes */
static void interpolate_array(fixed32 *scale, int old_size, int new_size)
@ -1227,7 +1228,7 @@ static void interpolate_array(fixed32 *scale, int old_size, int new_size)
}
}
}
#endif
/* compute x^-0.25 with an exponent and mantissa table. We use linear
interpolation to reduce the mantissa table size at a small speed
expense (linear interpolation approximately doubles the number of
@ -1397,7 +1398,7 @@ static int decode_exp_vlc(WMADecodeContext *s, int ch)
static int wma_decode_block(WMADecodeContext *s)
{
int n, v, a, ch, code, bsize;
int coef_nb_bits, total_gain, parse_exponents;
int coef_nb_bits, total_gain;
//static fixed32 window[BLOCK_MAX_SIZE * 2]; //crap can't do this locally on the device! its big as the whole stack
int nb_coefs[MAX_CHANNELS];
fixed32 mdct_norm;
@ -1564,15 +1565,8 @@ static int wma_decode_block(WMADecodeContext *s)
}
}
/* exposant can be interpolated in short blocks. */
parse_exponents = 1;
if (s->block_len_bits != s->frame_len_bits)
{
parse_exponents = get_bits(&s->gb, 1);
}
if (parse_exponents)
{
/* exponents can be reused in short blocks. */
if ((s->block_len_bits == s->frame_len_bits) || get_bits(&s->gb, 1)) {
for(ch = 0; ch < s->nb_channels; ++ch)
{
@ -1589,24 +1583,11 @@ static int wma_decode_block(WMADecodeContext *s)
{
decode_exp_lsp(s, ch);
}
s->exponents_bsize[ch] = bsize;
}
}
}
else
{
for(ch = 0; ch < s->nb_channels; ++ch)
{
if (s->channel_coded[ch])
{
interpolate_array(s->exponents[ch],
1 << s->prev_block_len_bits,
s->block_len);
}
}
}
//ok up to here!
//printf("got here!\n");
//rb->splash(HZ, "in wma_decode_block 2");
/* parse spectral coefficients : just RLE encoding */
for(ch = 0; ch < s->nb_channels; ++ch)
{
@ -1693,8 +1674,6 @@ static int wma_decode_block(WMADecodeContext *s)
}
//rb->splash(HZ, "in wma_decode_block 3");
/* finally compute the MDCT coefficients */
for(ch = 0; ch < s->nb_channels; ++ch)
{
@ -1706,15 +1685,14 @@ static int wma_decode_block(WMADecodeContext *s)
fixed64 mult;
fixed64 mult1;
fixed32 noise;
int i, j, n, n1, last_high_band;
fixed32 exp_power[HIGH_BAND_MAX_SIZE];
int i, j, n, n1, last_high_band, esize;
fixed32 exp_power[HIGH_BAND_MAX_SIZE];
//double test, mul;
//total_gain, coefs1, mdctnorm are lossless
//total_gain, coefs1, mdctnorm are lossless
coefs1 = s->coefs1[ch];
exponents = s->exponents[ch];
esize = s->exponents_bsize[ch];
mult = fixdiv64(pow_table[total_gain],Fixed32To64(s->max_exponent[ch]));
// mul = fixtof64(pow_table[total_gain])/(s->block_len/2)/fixtof64(s->max_exponent[ch]);
@ -1829,13 +1807,14 @@ static int wma_decode_block(WMADecodeContext *s)
atemp = (fixed32)(coefs1[i]*mult>>16);
//atemp= ftofix32(coefs1[i] * fixtof64(exponents[i]) * fixtof64(mult>>16)); //this "works" in the sense that the mdcts converge
*coefs++=fixmul32(atemp,exponents[i]); //this does not work
//this can still overflow in rare cases
//running a full scale value square wave through here does bad things
*coefs++=fixmul32(atemp,exponents[i<<bsize>>esize]);
//atemp = ftofix32( coefs1[i]*mul* fixtof64(exponents[i]) ); //this doesn't seem to help any at all.
// *coefs++=atemp;
} //coefs1 could underflow?
}
n = s->block_len - s->coefs_end[bsize];
for(i = 0;i < n; ++i)
*coefs++ = 0;