shithub: opus

cd29b027 – Jean-Marc Valin <[email protected]> authored on 2009/07/01 05:59

fixed-point: fixing two overflows that didn't really affect quality

7c422653 – Gregory Maxwell <[email protected]> authored on 2009/06/30 23:50

Fix compilation of test cases with fixed-debug enabled

da1156af – Jean-Marc Valin <[email protected]> authored on 2009/06/30 21:27

fixing a floating point bug for very small signals

63396f39 – Jean-Marc Valin <[email protected]> authored on 2009/06/30 19:54

ietf doc: misc

84966954 – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/06/29 21:06

Update valgrind support to match the latest valgrind API.

c57475b4 – Jean-Marc Valin <[email protected]> authored on 2009/06/30 11:27

ietf doc: range coder

17ea4b7e – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/06/30 07:29

For some frame sizes there can be 5 shorts in a frame.

6d454d8d – Jean-Marc Valin <[email protected]> authored on 2009/06/30 06:31

This fixes a potential assertion failure with some very weak signals

02428d6c – Jean-Marc Valin <[email protected]> authored on 2009/06/29 19:55

ietf doc: decoder overview (ASCII art)

d4a907b2 – Jean-Marc Valin <[email protected]> authored on 2009/06/29 19:42

ietf doc: encoder overview (ASCII art)

224824b0 – Jean-Marc Valin <[email protected]> authored on 2009/06/29 13:35

ietf doc: pitch description

66a57c8e – Gregory Maxwell <[email protected]> authored on 2009/06/29 10:03

Fixes for celt_header_to_packet / celtdec for bigendian systems.

a80958b0 – Gregory Maxwell <[email protected]> authored on 2009/06/29 08:48

Note some more platforms where float-approx is tested, fix a bug in the prediction ctl,

b92dce32 – Gregory Maxwell <[email protected]> authored on 2009/06/28 15:51

Additional mathops.h tests.

8d2c51af – Jean-Marc Valin <[email protected]> authored on 2009/06/28 10:46

ietf doc: final fine bits, some fixes to the references

065bc152 – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/06/24 12:43

Numerous autoconf changes. Configure option for FLOAT_APPROX

6ca7f10e – Jean-Marc Valin <[email protected]> authored on 2009/06/23 17:23

Bump version to 0.6.0 and increase bitstream version.

640f7fd8 – Jean-Marc Valin <[email protected]> authored on 2009/06/21 05:47

Gives an error when trying to create a stereo mode with DISABLE_STEREO defined.

2014ca38 – Jean-Marc Valin <[email protected]> authored on 2009/06/18 19:33

Minor simplification to the transient code

0dd2cd49 – Jean-Marc Valin <[email protected]> authored on 2009/06/18 19:18

Simplifies the stereo mixing code, throwing away some old code.

e7dea13c – Jean-Marc Valin <[email protected]> authored on 2009/06/18 07:20

ietf doc: misc corrections

e2b7616e – Jean-Marc Valin <[email protected]> authored on 2009/06/17 18:39

Fixing a few bugs introduced in the energy interleaving changes and due to

b0b23a35 – Jean-Marc Valin <[email protected]> authored on 2009/06/17 17:24

Interleaving the left and right final fine energy bits

fb1b4dae – Jean-Marc Valin <[email protected]> authored on 2009/06/17 16:37

Interleaving the left and right fine energy

76bda753 – Jean-Marc Valin <[email protected]> authored on 2009/06/17 13:47

ietf doc: security, VBR, stereo

1b7e9c41 – Jean-Marc Valin <[email protected]> authored on 2009/06/17 04:02

intealeaving the left and right coarse energy

5b9b51bc – Jean-Marc Valin <[email protected]> authored on 2009/06/16 18:27

Forced intra now overrides intra avoidance at low rate. Also, using the number

a20500db – Jean-Marc Valin <[email protected]> authored on 2009/06/16 12:45

ietf doc: stereo description

9dff0218 – Jean-Marc Valin <[email protected]> authored on 2009/06/15 18:40

Minor stuff: preventing float underflow in celt_exp2(), preventing the use of

9edb7b42 – Jean-Marc Valin <[email protected]> authored on 2009/06/15 07:22

This fixes a folding issue with stereo

08a82ffb – Jean-Marc Valin <[email protected]> authored on 2009/06/14 10:05

More de-interleaving: denormalised MDCT no longer stored with interleaved

66fb2462 – Jean-Marc Valin <[email protected]> authored on 2009/06/14 08:31

removing redundant calls to exp2()

095abafc – Jean-Marc Valin <[email protected]> authored on 2009/06/14 05:29

Simplifying the stereo energy quantisation functions now that energy is

83c26e71 – Jean-Marc Valin <[email protected]> authored on 2009/06/14 05:19

Now storing the band energies in de-interleaved format when doing stereo

7a56741b – Jean-Marc Valin <[email protected]> authored on 2009/06/13 18:59

removing redundant calls to log2()

f1ce9007 – Jean-Marc Valin <[email protected]> authored on 2009/06/13 18:30

float approximations for log2() and exp2()

495ea676 – Jean-Marc Valin <[email protected]> authored on 2009/06/12 13:41

ietf doc: characteristics, bit-stream

c10565bd – Jean-Marc Valin <[email protected]> authored on 2009/06/12 12:52

ietf doc: PVQ search

59f67687 – Jean-Marc Valin <[email protected]> authored on 2009/06/12 07:49

ietf doc: better description of the bitstream

f1b5006d – Jean-Marc Valin <[email protected]> authored on 2009/06/11 13:40

ietf doc: fine energy

a76a5e8f – Jean-Marc Valin <[email protected]> authored on 2009/06/10 18:46

Better ebits rounding and making it possible to add remaining ebits even to

eb5aa094 – Jean-Marc Valin <[email protected]> authored on 2009/06/10 13:48

ietf doc: synthesis

9ac1673c – Jean-Marc Valin <[email protected]> authored on 2009/06/09 13:44

PVQ doc

52cb5fb3 – Jean-Marc Valin <[email protected]> authored on 2009/06/10 04:08

Adding extra fine bits only when we have rounded down in the allocation

b469e601 – Jean-Marc Valin <[email protected]> authored on 2009/06/09 10:17

cwrs documentation

24c25a23 – Jean-Marc Valin <[email protected]> authored on 2009/06/09 07:56

Folding, coarse energy, source cleanup

c871c8de – Jean-Marc Valin <[email protected]> authored on 2009/06/08 20:57

The 010101 end of stream pattern is no longer needed now that we use the

39710536 – Jean-Marc Valin <[email protected]> authored on 2009/06/08 20:10

This makes it possible to use the bits left after PVQ for fine energy.

cbc2afb1 – Jean-Marc Valin <[email protected]> authored on 2009/06/08 13:45

Illustrating the bit-stream

6ed927f2 – Jean-Marc Valin <[email protected]> authored on 2009/06/08 12:50

Describing pitch and short blocks

028ba369 – Jean-Marc Valin <[email protected]> authored on 2009/06/08 10:27

Defining mode flags

7144200a – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/06/05 11:01

Add support for interframe prediction control in celtenc.

68280546 – Jean-Marc Valin <[email protected]> authored on 2009/06/08 05:33

Merge commit 'origin/master'

695ab700 – Jean-Marc Valin <[email protected]> authored on 2009/06/07 19:54

Minor simplification to the fine energy code

d8dcd244 – Jean-Marc Valin <[email protected]> authored on 2009/06/07 09:01

This simplifies the energy quantisation by converting to the log2() scale

66a88594 – Jean-Marc Valin <[email protected]> authored on 2009/06/05 14:10

Indent options

52b4cb21 – Jean-Marc Valin <[email protected]> authored on 2009/06/05 13:21

Merge commit 'greg/master'

56522add – Jean-Marc Valin <[email protected]> authored on 2009/06/05 13:17

IETF doc update, including better source code formatting

2dd3d325 – Gregory Maxwell <[email protected]> authored on 2009/06/05 10:05

For celt_encoder_ctl CELT_SET_LTP is replaced with CELT_SET_PREDICTION

f43488cd – Jean-Marc Valin <[email protected]> authored on 2009/06/05 07:07

Fix non-C90 declaration after statement

033e6091 – Jean-Marc Valin <[email protected]> authored on 2009/06/04 20:02

Improving code conversion to publishable form.

dc67fa9a – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/06/04 13:17

Handle malloc failure in mode_create gracefully without leaking memory.

1716999b – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/06/04 11:15

Add paranoid checking for the validity of the encoder and the decoder

083883e6 – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/06/03 16:02

Fixes for the tests in pseudo-stack mode.

dc663473 – Jean-Marc Valin <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/06/04 12:39

Updated scripts for converting the source code to something more publishable

f3b44ef4 – Gregory Maxwell <[email protected]> authored on 2009/06/03 09:37

Add get_mode CTL for the encoder and decoder.

9a44cde2 – Jean-Marc Valin <[email protected]> authored on 2009/06/02 16:21

Removed support for band-per-band stereo_mode that was never really used

7353203b – Jean-Marc Valin <[email protected]> authored on 2009/06/02 16:07

Updating DISABLE_STEREO to the new stereo code (i.e. disabling more stuff)

dcd580d7 – Jean-Marc Valin <[email protected]> authored on 2009/06/02 04:02

Disabling K>128 until it can be shown to be useful.

445720f9 – Jean-Marc Valin <[email protected]> authored on 2009/06/02 03:59

It is no longer necessary to have a minimum of one bit for fine energy

527db5c4 – Jean-Marc Valin <[email protected]> authored on 2009/06/02 03:56

Adding a decision mechanism for turning folding on or off depending on the

d9458cd0 – Gregory Maxwell <[email protected]> authored on 2009/05/30 13:04

Add a mode_info call to obtain the sampling rate of the current mode.

e4c05a4d – Jean-Marc Valin <[email protected]> authored on 2009/05/30 06:09

Making the modified transient code work with stereo as well

18c1c22e – Jean-Marc Valin <[email protected]> authored on 2009/05/30 05:50

Fix for a mode memory leak -- was still allocating memory for the energy

aa936256 – Jean-Marc Valin <[email protected]> authored on 2009/05/29 18:14

New transient handling code that weights the MDCTs directly unless there's

2b73a27e – Jean-Marc Valin <[email protected]> authored on 2009/05/26 20:05

1024 pulses ought to be enough for everybody

6391fe82 – Timothy B. Terriberry <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/05/26 19:15

Minor optimisation -- using do-while() instead of for() in isqrt32()

7aa0ed40 – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/02/19 16:28

Make ectest use random random numbers. Add a new test

c4b146c4 – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/05/26 17:15

Fixing the encoder's handling of no-folding; with folding enabled

5aaf1233 – Jean-Marc Valin <[email protected]> authored on 2009/05/26 17:50

Enabling more than 128 pulses for N=3 and N=4.

8eadcdc6 – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/05/26 07:56

Switch the N=5 case of CWRS to also use a binary search.

d98d8ae0 – Timothy B. Terriberry <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/05/26 05:09

CWRS clean-ups and optimizations.

1cca1516 – Jean-Marc Valin <[email protected]> authored on 2009/05/23 18:28

This fixes a bug in the new stereo code triggered only at ridiculously high

331130e0 – Jean-Marc Valin <[email protected]> authored on 2009/05/23 17:53

Slightly increasing the range of value by making sure the "start fs" is even

95400774 – Jean-Marc Valin <[email protected]> authored on 2009/05/23 11:09

Documentation update: removed lookahead from the doxygen comment long after it

be607256 – Jean-Marc Valin <[email protected]> authored on 2009/05/23 10:17

Making use of previously unused range of the laplace encoder to have a slightly

ce02e1b4 – Jean-Marc Valin <[email protected]> authored on 2009/05/23 06:29

Setting ft=32768 instead of 32767 so that implementations can use shifts and

d1d74056 – Jean-Marc Valin <[email protected]> authored on 2009/05/22 20:07

Should be enough to handle signals with a 144 dB (24-bit) dynamic range

6f18d111 – Jean-Marc Valin <[email protected]> authored on 2009/05/20 18:50

Adds support for up to 32767 pulses for the N=3 case (disabled for now)

0ac2b2fe – Gregory Maxwell <[email protected]> authored on 2009/05/21 19:08

The change to FLOAT2INT16 in a8734e0f would break float input for

454d1d0c – John Ridges <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/05/21 18:38

This code does a very naive reset of the state. It is possible (even likely)

a8734e0f – Jean-Marc Valin <[email protected]> authored on 2009/05/21 18:29

Proper use of SCALEIN() as suggested by John Ridges

888d8ce9 – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/05/21 00:21

VBR support. VBR API and VBR support in celtenc.

381d05aa – Jean-Marc Valin <[email protected]> authored on 2009/05/19 06:12

Make it possible to use more than 255 pulses (not used yet).

fa4a3bf5 – Jean-Marc Valin <[email protected]> authored on 2009/05/12 11:39

making codec draft more compliant with IETF submission rules

8161f9e9 – Jean-Marc Valin <[email protected]> authored on 2009/05/11 12:34

Version -01 of the RTP draft

db5f38ef – Jean-Marc Valin <[email protected]> authored on 2009/05/11 12:33

Change to ipr="trust200902" to make the experimental xml2rfc happy. Also, added

e95de9a2 – Jean-Marc Valin <[email protected]> authored on 2009/05/11 08:53

Updated RTP draft

8842fdee – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/05/04 11:58

Fixing stereo: Do not attempt to use more bits than are available.

ec836da2 – Gregory Maxwell <[email protected]> authored and Jean-Marc Valin <[email protected]> committed on 2009/05/04 10:55

Reduce rounding in the mode allocator and remove a scaling loop.