shithub: opus

Info  •  Files  •  Log  •  Branches

9d312b97 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/05 12:34

Defining RADIX_TWO_ONLY removes all butterflies, except for radix 2 and 4.

f93747c4 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/05 12:20

fixed-point: changed find_spectral_pitch() to use single-precision (16-bit) FFT.

40244448 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/05 10:18

nothing to see here

b624cdd2 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/05 08:40

fixed-point: masking curve computation now converted. ***Fixed a bug in the

b0c93c66 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/05 08:12

oops, pgain_table now has half the number of elements

afa5eb38 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/05 07:22

fixed-point: quantised the pitch gain table.

92376697 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/05 06:31

fixed-point: converted the pitch gain quantisation, except for the codebook

a78dc946 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/05 03:15

Making 16-bit compilers happy

d9de5931 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/05 03:11

Fixed a bunch of warnings

a1bc18a3 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/04 17:52

compute_band_energies() merged with normalised_bands()

069c3de6 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/04 15:38

Removed deprecated mode interface and added missing include

7137c26e – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/04 12:43

fixed-point: exp_rotation() now fully converted, using an approximation of the

7e94cc4d – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/04 12:31

fixed-point: converted denormalise_bands()

a31899f0 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/03 17:28

added suffix to real fft. The celtclient can now be built with an unmodified

300cb5b8 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 12:52

First attempt at adding a suffix to kiss_fft functions so we can have multiple

bc60f8b3 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 12:23

making sure __GNUC_PREREQ() is only used when __GNUC__ is defined

07586892 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 12:19

pre-emphasis coef now a constant

31213278 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 11:52

A bunch of const qualifyers and a few comments

730c6e39 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 11:38

Applying the window properly in find_spectral_pitch()

ddb181b7 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 09:53

A bit more reordering in stack memory allocation -- saved 1024 bytes on peak use

137ec8e9 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 09:44

applying the pitch windowing directly in find_spectral_pitch()

05080b4b – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 09:23

No longer storing the zero parts of the in[] array in the encoder

632f218b – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 09:08

Pitch predictor now uses a larger range of offset by not being limited to the

0141723d – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 08:59

compute_mdcts() no longer reads the part of the input that should be zeros

3e08a883 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 08:49

Only the (rising) overlap part of the window is stored now. No need to hold the

f28062fe – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 08:24

Don't bother multiplying when the window's value is one or zero

227ad2f8 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 05:07

cwrs32-test should be 16-bit clean now.

4c2787a4 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/03/03 05:04

Laplace encoder now works with 16-bit frequencies

eb5a245a – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/02 23:11

Hadn't realised the bitr* stuff wasn't needed.

7ff2cbc1 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/02 16:25

oops, fixed the Laplace testcase

37942649 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/01 17:55

Saved 4 kB of stack usage in find_spectral_pitch() by doing the FFT in-place

7cdc5a34 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/01 15:56

Making it easier to debug laplace testcase

a556694e – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/01 15:14

Stereo pitch search no longer requires twice the memory/complexity.

86cb7e86 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/01 03:45

oops. Another 16-bit fix

9e006b6d – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/01 03:33

16-bit fix

d748cd55 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/03/01 02:27

Another C90-fying pass. Fixed some warnings in the process.

81b38c22 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/29 16:08

Moved mdct state and window to the mode so it can be shared between multiple

37131e3a – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/29 15:03

Just realised half the twiddles in the real FFT weren't even used. Removing

8d4ac155 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/29 12:24

Saved on stack usage by changing the order of the allocation

9b9e9862 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/29 12:07

Bit of memory reduction in the pitch search. More needed.

cac91ec4 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/29 12:03

Fixed temp arrays that were allocated too large (no change in peak mem)

877b1975 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/29 11:40

fixed-point: converted intra prediction and folding, unb0rked mixed-precision

2aaa0fee – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/29 11:07

fixed-point: overflow debugging now works again.

8600f69f – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/29 10:14

Initial support for a managed stack/scratchpad. Still needs some work.

8b2a5923 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/28 19:32

fixed-point: energy quantisation is entirely converted now.

adf87e26 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/28 19:15

fixed-point: more energy quantisation stuff.

47bf175f – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/28 19:07

fixed-point: a few more bits converted from the energy quantisation. Almost done

68b02b10 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/28 18:37

fixed-point: copied the exp2 implementation from Speex, using it for dB2Amp()

2eaba8ac – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/28 17:14

fixed-point: log approximation

98193470 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/28 12:46

fixed-point: unquant_energy_mono() has received the fixed-point code from

d0316201 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/28 11:52

fixed-point: second check-point on quant_energy_mono()

43b3537d – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/28 11:41

fixed-point: first check-point on quant_energy_mono() conversion

4e1328b0 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/28 10:14

fixed-point: alg_quant() entirely converter -- may require revisiting...

8b158f56 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/28 09:44

fixed-point: (hopefully) last check-point for alg_quant() conversion

642ff944 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/28 09:33

fixed-point: The cross-products in alg_quant() are now all converted.

c9d606f5 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/28 08:46

fixed-point: third alg_quant() checkpoint

ed9e4236 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/28 07:22

fixed-point: second check-point in the conversion of alg_quant()

f675adce – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/28 07:15

fixed-point: First check-point in alg_quant() conversion

a02ca1ee – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/28 06:33

fixed-point: Added a ROUND() operator, no real change to the code

6ba69782 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/27 19:28

Oops. Fixed a missing \ in a Makefile.am

79fa8b6c – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/27 19:14

Just in case sizeof(char)!=1

5e171490 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/27 19:07

Removed some unused parameters and fixed a couple -W warnings

3ca9b1d2 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/27 18:50

fixed-point: Moved sqrt and cos approximations to mathops.h

a847b775 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 12:46

fixed-point: done with mix_pitch_and_residual() though a bit of cleaning up

1ca07225 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 12:23

fixed-point: mix_pitch_and_residual() check-point #3

b50c541d – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 12:05

fixed-point: another mix_pitch_and_residual() check-point

d17edd35 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 11:52

fixed-point: half-way converting mix_pitch_and_residual() -- just check-pointing

9d8d9b3f – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 11:17

fixed-point: compression factor (alpha) now a 16-bit value (still internally

9fbb56a5 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 10:44

fixed-point: converted pitch_quant_bands() -- that one was an easy one-liner

ff74e396 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 10:35

fixed-point: converted compute_pitch_gain() and removed the energy-based

3f382cae – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 09:07

Fixing pi again before another massive public outcry

387a20d7 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 08:49

fixed-point: exp_rotation() mostly converted (still need to convert the cos/sin)

e40a19cb – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 07:09

fixed-point: no float vars left (and nearly no float ops left) in celt.c

5d561834 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 06:59

fixed-point: log-energy for previous frame now a 16-bit value. This currently

1d6ad108 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 06:25

fixed-point: converted window to 16-bit value

42074389 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 06:08

fixed-point: pre/de-emphasis converted.

d4018c3a – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/27 05:09

Reduced code duplication in vq.c by adding mix_pitch_and_residual() which

2f5ccf60 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/27 02:48

C90-fying ectest

7a455d4e – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/26 18:59

Another 16-bit fix.

b30dd9e1 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/26 15:32

fixed-point: got stereo to work again by fixing renormalise_bands()

2805a14e – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 12:02

fixed-point: celt_pgain_t now a 16-bit value (Q15)

5f09ea56 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 11:43

celt_mask_t for masking curves

8835a4df – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 11:35

Removed useless masking curve for current frame and ignored Fs argument to

017d4455 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 11:19

pitch gain is now celt_pgain_t

b60340f7 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 10:41

fixed-point: band energy now a 32-bit value. It might have (barely) fix into

e901fe35 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 09:46

fixed-point: added a celt_ener_t type for band energy.

e046c18a – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 08:51

Minus a bunch of warnings when enabling alloca()

e995c31f – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 08:48

Put the channel count outside of the energy calculation. Increased the allowed

2fa8affd – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 06:38

fixed-point: celt_norm_t now a 16-bit value.

9a5f3d2e – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 05:38

16-bit fixes and warning fixes

508de38d – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/26 05:28

Trying to be nice with 16-bit chips.

65d79e44 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/26 02:38

Corrected the last digit of pi after massive public outcry :-)

b5a4c40d – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/25 12:58

Ogg decoder does gapless for the beginning of the file.

b4dfce46 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/25 12:41

fixed-point: Using a NORM_SCALING of 16384, sig_norm_t is still a float though.

5c032a4c – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/25 09:55

fixed-point: celt_sig_t now a 32-bit value.

49ca99ef – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/25 09:05

fixed-point: initial support for using the fixed-point MDCT (rest is still all

bd43729e – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/25 07:07

MDCT is in fixed-point now

2e8a3b20 – Jean-Marc Valin <Jean-Marc.Valin@csiro.au> authored on 2008/02/25 06:49

MDCT conversion, part I.

42667b0a – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/25 04:47

real fft no longer needs an internal buffer.

4508ebd5 – Jean-Marc Valin <jean-marc.valin@usherbrooke.ca> authored on 2008/02/24 18:48

real fft snr comparison testcase includes DC