opus.git
10 years agoMake ectest use random random numbers. Add a new test
Gregory Maxwell [Fri, 20 Feb 2009 02:28:33 +0000 (21:28 -0500)]
Make ectest use random random numbers. Add a new test
"tandem-test" which tests the encoder and decoder in
many modes.

10 years agoFixing the encoder's handling of no-folding; with folding enabled
Gregory Maxwell [Wed, 27 May 2009 01:15:12 +0000 (21:15 -0400)]
Fixing the encoder's handling of no-folding; with folding enabled
the encoder could attempt to create folding-free short blocks which
is not permitted by the bitstream.

10 years agoEnabling more than 128 pulses for N=3 and N=4.
Jean-Marc Valin [Wed, 27 May 2009 01:50:38 +0000 (21:50 -0400)]
Enabling more than 128 pulses for N=3 and N=4.

10 years agoSwitch the N=5 case of CWRS to also use a binary search.
Gregory Maxwell [Tue, 26 May 2009 15:56:37 +0000 (11:56 -0400)]
Switch the N=5 case of CWRS to also use a binary search.
This avoids the need for 64-bit addition and is faster on x86_64.
It may be slower on some platforms so the direct solution is still
available in the source.

10 years agoCWRS clean-ups and optimizations.
Timothy B. Terriberry [Tue, 26 May 2009 13:09:27 +0000 (09:09 -0400)]
CWRS clean-ups and optimizations.

Adds specialized O(N*log(K)) versions of cwrsi() and O(N) versions of icwrs()
 for N={3,4,5}, which allows them to operate all the way up to the theoretical
 pulse limit without serious performance degredation.
Also substantially reduces the computation time and stack usage of
 get_required_bits().
On x86-64, this gives a 2% speed-up for 256 sample frames, and almost a 16%
 speed-up for 64 sample frames.

10 years agoThis fixes a bug in the new stereo code triggered only at ridiculously high
Jean-Marc Valin [Sun, 24 May 2009 02:28:04 +0000 (22:28 -0400)]
This fixes a bug in the new stereo code triggered only at ridiculously high
bitrates. It was caused by excessive "balance" bits leading to qb>14.

10 years agoSlightly increasing the range of value by making sure the "start fs" is even
Jean-Marc Valin [Sun, 24 May 2009 01:53:57 +0000 (21:53 -0400)]
Slightly increasing the range of value by making sure the "start fs" is even
(rounding towards zero).

10 years agoDocumentation update: removed lookahead from the doxygen comment long after it
Jean-Marc Valin [Sat, 23 May 2009 19:09:15 +0000 (15:09 -0400)]
Documentation update: removed lookahead from the doxygen comment long after it
was removed from celt_mode_create()

10 years agoMaking use of previously unused range of the laplace encoder to have a slightly
Jean-Marc Valin [Sat, 23 May 2009 18:17:02 +0000 (14:17 -0400)]
Making use of previously unused range of the laplace encoder to have a slightly
wider range of values.

10 years agoSetting ft=32768 instead of 32767 so that implementations can use shifts and
Jean-Marc Valin [Sat, 23 May 2009 14:29:55 +0000 (10:29 -0400)]
Setting ft=32768 instead of 32767 so that implementations can use shifts and
thus save a few divisions.

10 years agoShould be enough to handle signals with a 144 dB (24-bit) dynamic range
Jean-Marc Valin [Sat, 23 May 2009 04:07:08 +0000 (00:07 -0400)]
Should be enough to handle signals with a 144 dB (24-bit) dynamic range

10 years agoAdds support for up to 32767 pulses for the N=3 case (disabled for now)
Jean-Marc Valin [Thu, 21 May 2009 02:50:46 +0000 (22:50 -0400)]
Adds support for up to 32767 pulses for the N=3 case (disabled for now)

10 years agoThe change to FLOAT2INT16 in a8734e0f would break float input for
Gregory Maxwell [Fri, 22 May 2009 03:08:46 +0000 (23:08 -0400)]
The change to FLOAT2INT16 in a8734e0f would break float input for
fixed point compilation because SCALEIN is a no-op in fixed point
mode but the float interface is still normally +/- 1.0.

This patch adds a seperate define for the scaling factor so people
can adjust it for unusual input levels.

10 years agoThis code does a very naive reset of the state. It is possible (even likely)
John Ridges [Fri, 22 May 2009 02:38:39 +0000 (22:38 -0400)]
This code does a very naive reset of the state. It is possible (even likely)
that I'm resetting more than is necessary, but I would have to know a lot
more about how CELT works to determine that.

10 years agoProper use of SCALEIN() as suggested by John Ridges
Jean-Marc Valin [Fri, 22 May 2009 02:29:09 +0000 (22:29 -0400)]
Proper use of SCALEIN() as suggested by John Ridges

10 years agoVBR support. VBR API and VBR support in celtenc.
Gregory Maxwell [Thu, 21 May 2009 08:21:53 +0000 (04:21 -0400)]
VBR support. VBR API and VBR support in celtenc.

10 years agoMake it possible to use more than 255 pulses (not used yet).
Jean-Marc Valin [Tue, 19 May 2009 14:12:34 +0000 (10:12 -0400)]
Make it possible to use more than 255 pulses (not used yet).

10 years agomaking codec draft more compliant with IETF submission rules
Jean-Marc Valin [Tue, 12 May 2009 19:39:37 +0000 (15:39 -0400)]
making codec draft more compliant with IETF submission rules

10 years agoVersion -01 of the RTP draft
Jean-Marc Valin [Mon, 11 May 2009 20:34:58 +0000 (16:34 -0400)]
Version -01 of the RTP draft

10 years agoChange to ipr="trust200902" to make the experimental xml2rfc happy. Also, added
Jean-Marc Valin [Mon, 11 May 2009 20:33:05 +0000 (16:33 -0400)]
Change to ipr="trust200902" to make the experimental xml2rfc happy. Also, added
(old) version -00 of the draft

10 years agoUpdated RTP draft
Jean-Marc Valin [Mon, 11 May 2009 16:53:46 +0000 (12:53 -0400)]
Updated RTP draft

10 years agoFixing stereo: Do not attempt to use more bits than are available.
Gregory Maxwell [Mon, 4 May 2009 19:58:40 +0000 (15:58 -0400)]
Fixing stereo: Do not attempt to use more bits than are available.
This change breaks the bitstream.

Make the first frame out of the encoder an intra-frame. (While not required this may help
in the case that the decoder has old state laying around, and it shouldn't hurt)

10 years agoReduce rounding in the mode allocator and remove a scaling loop.
Gregory Maxwell [Mon, 4 May 2009 18:55:40 +0000 (14:55 -0400)]
Reduce rounding in the mode allocator and remove a scaling loop.
This breaks the bitstream.

10 years agoThe intra decision is now made for the following frame, not the current one.
Jean-Marc Valin [Sat, 2 May 2009 13:57:50 +0000 (09:57 -0400)]
The intra decision is now made for the following frame, not the current one.
Also, setting transient shift to zero when not using short blocks (this causes a
failed assertion otherwise)

10 years agoDynamically selecting intra energy based on energy variations from the previous
Jean-Marc Valin [Sat, 2 May 2009 01:28:58 +0000 (21:28 -0400)]
Dynamically selecting intra energy based on energy variations from the previous
frame

10 years agoAdd support for intra-coding of the coarse energy.
Jean-Marc Valin [Fri, 1 May 2009 23:58:55 +0000 (19:58 -0400)]
Add support for intra-coding of the coarse energy.

10 years agoApparently the 0b notation used for the flags isn't standard C.
Jean-Marc Valin [Thu, 30 Apr 2009 02:16:26 +0000 (22:16 -0400)]
Apparently the 0b notation used for the flags isn't standard C.

10 years agofix testcelt to use loss simulation
Torben Hohn [Thu, 23 Apr 2009 19:51:54 +0000 (21:51 +0200)]
fix testcelt to use loss simulation

10 years agoNew code for encoding the flags
Jean-Marc Valin [Wed, 29 Apr 2009 11:44:13 +0000 (07:44 -0400)]
New code for encoding the flags

10 years agoMaking it easier to Torben to develop his new PLC code
Jean-Marc Valin [Mon, 27 Apr 2009 23:35:09 +0000 (19:35 -0400)]
Making it easier to Torben to develop his new PLC code

10 years agoAddressing some of Randell Jesup's comments
Jean-Marc Valin [Wed, 15 Apr 2009 20:52:25 +0000 (16:52 -0400)]
Addressing some of Randell Jesup's comments

10 years agoFix (unexploitable) buffer overrun when _m=1 during the cwrs table init, as
Timothy B. Terriberry [Sun, 12 Apr 2009 02:15:43 +0000 (22:15 -0400)]
Fix (unexploitable) buffer overrun when _m=1 during the cwrs table init, as
reported by Bjoern Rasmussen.

10 years agoMaking the "data" argument to celt_decode() const as pointed out by Bjoern
Jean-Marc Valin [Sun, 12 Apr 2009 02:01:20 +0000 (22:01 -0400)]
Making the "data" argument to celt_decode() const as pointed out by Bjoern
Rasmussen.

10 years agoCleanup: getting rid of some old bits of stereo code that are no longer useful
Jean-Marc Valin [Mon, 30 Mar 2009 02:29:27 +0000 (22:29 -0400)]
Cleanup: getting rid of some old bits of stereo code that are no longer useful

10 years agoSquashed commit of the following:
Jean-Marc Valin [Fri, 27 Mar 2009 00:23:14 +0000 (20:23 -0400)]
Squashed commit of the following:

commit ea807b68678dd76175def2c5eb006c6bdb16679e
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Wed Mar 25 23:24:41 2009 -0400

    cleanup before merge

commit 73ad1a0202641be6a23903e464ece21fe332a131
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Sat Mar 21 00:02:16 2009 -0400

    Some tuning of the new stereo

commit c05057eb57a7723045214a2f830fd561388ae48a
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Sun Mar 15 19:56:11 2009 -0400

    fixed-point: atan2() converted

commit a8476cf8be55b0612d42df98d9807ca1335adfe3
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Sat Mar 14 23:10:04 2009 -0400

    fixed-point: Getting the new stereo code working in fixed-point (still more
    work left)

commit 70a452761a5ce15700664e7167886dce5914cbd0
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Mon Mar 2 23:36:25 2009 -0500

    Coding left and right independently for lower bands

commit 4efd1e6385c7d036749080265a8d26668312b91b
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Sun Mar 1 23:56:46 2009 -0500

    Removed the sqrt(C) from the normalisation, which simplifies a lot of things.

commit a4f3c5c60bc396bf644afa49b49e6b24ccf144f8
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Fri Feb 20 20:49:38 2009 -0500

    Better point stereo calculation when we don't encode the side anyway

commit f08525de4739f4017d19ec2e2022883deda8f826
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Thu Feb 19 19:43:09 2009 -0500

    Apparently, Timothy's calculations for fine energy allocation also apply
    to the quantisation of theta.

commit 6548cffc9d3f996b8a8dbfab982f0da0bc6c2dc2
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Thu Feb 19 07:35:24 2009 -0500

    Better handling of the "theta bits" and disabling the orthogonalize()

commit 7aa82c694967afa85dd30be8cc670663f98829fe
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Wed Feb 18 08:01:07 2009 -0500

    tuning the new stereo

commit c2b780a773de66fd9613c7cd54c09b705fe9ce45
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Sun Feb 15 22:24:52 2009 -0500

    The new stereo coupling actually decodes properly now.

commit 85513c203d773bebcf0a6055f953170d563d890c
Author: Jean-Marc Valin <jean-marc.valin@usherbrooke.ca>
Date:   Sun Feb 15 21:31:16 2009 -0500

    First attempt at a new "constrained" MS stereo scheme

10 years agoAllow --without-ogg to address issue raised by Ian Davidson.
Gregory Maxwell [Wed, 11 Mar 2009 17:13:19 +0000 (13:13 -0400)]
Allow --without-ogg to address issue raised by Ian Davidson.

10 years agoCleanup, acknowledgments
Jean-Marc Valin [Fri, 6 Mar 2009 21:28:17 +0000 (16:28 -0500)]
Cleanup, acknowledgments

10 years agoMention the possibility of increasing the packetization interval in response to conge...
Gregory Maxwell [Fri, 6 Mar 2009 20:15:29 +0000 (15:15 -0500)]
Mention the possibility of increasing the packetization interval in response to congestion.

10 years agolow-overhead mode
Jean-Marc Valin [Fri, 6 Mar 2009 15:42:07 +0000 (10:42 -0500)]
low-overhead mode

10 years agoC code example
Jean-Marc Valin [Fri, 6 Mar 2009 14:45:39 +0000 (09:45 -0500)]
C code example

10 years agocleanup
Jean-Marc Valin [Thu, 5 Mar 2009 22:01:51 +0000 (17:01 -0500)]
cleanup

10 years agoframe size
Jean-Marc Valin [Thu, 5 Mar 2009 16:57:47 +0000 (11:57 -0500)]
frame size

10 years agoRemoved a few more inconsistancies
Jean-Marc Valin [Thu, 5 Mar 2009 16:42:05 +0000 (11:42 -0500)]
Removed a few more inconsistancies

10 years agoptime update
Jean-Marc Valin [Thu, 5 Mar 2009 15:03:38 +0000 (10:03 -0500)]
ptime update

10 years agoTrying to use ptime
Jean-Marc Valin [Wed, 4 Mar 2009 22:48:30 +0000 (17:48 -0500)]
Trying to use ptime

10 years agoA stereo example.
Gregory Maxwell [Wed, 4 Mar 2009 18:31:49 +0000 (13:31 -0500)]
A stereo example.

10 years agominor changes
Jean-Marc Valin [Tue, 3 Mar 2009 22:37:29 +0000 (17:37 -0500)]
minor changes

10 years agoUsing b=AS: for the bandwidth. Moved the example to the end of the SDP section
Jean-Marc Valin [Tue, 3 Mar 2009 19:02:02 +0000 (14:02 -0500)]
Using b=AS: for the bandwidth. Moved the example to the end of the SDP section

10 years agoTook Greg Maxwell's channel mapping and simplified it
Jean-Marc Valin [Tue, 3 Mar 2009 17:19:14 +0000 (12:19 -0500)]
Took Greg Maxwell's channel mapping and simplified it

10 years agoMultiple channels
Jean-Marc Valin [Mon, 2 Mar 2009 22:23:46 +0000 (17:23 -0500)]
Multiple channels

10 years agoRTP Draft updates; merge new security boiler-plate from draft-ietf-avt-rtp-howto...
Gregory Maxwell [Mon, 2 Mar 2009 19:34:15 +0000 (14:34 -0500)]
RTP Draft updates; merge new security boiler-plate from draft-ietf-avt-rtp-howto-06.txt

10 years agomisc changes to rtp draft
Jean-Marc Valin [Fri, 27 Feb 2009 22:12:09 +0000 (17:12 -0500)]
misc changes to rtp draft

10 years agoRemoved a few dozen warnings from the RTP XML draft.
Jean-Marc Valin [Fri, 27 Feb 2009 14:23:28 +0000 (09:23 -0500)]
Removed a few dozen warnings from the RTP XML draft.

10 years agoAdd version info to celtenc/celtdec output as suggested on the
Gregory Maxwell [Thu, 26 Feb 2009 22:05:37 +0000 (17:05 -0500)]
Add version info to celtenc/celtdec output as suggested on the
mailing list by Rob Til Freedmen. Also add a bitstream version
check to these applications.

10 years agoAPI documentation updates.
Gregory Maxwell [Tue, 17 Feb 2009 02:29:15 +0000 (21:29 -0500)]
API documentation updates.

10 years agoMiscellaneous comment, copyright notice, readme updates.
Gregory Maxwell [Mon, 16 Feb 2009 23:56:44 +0000 (18:56 -0500)]
Miscellaneous comment, copyright notice, readme updates.

10 years agoPreparing for version 0.5.2
Jean-Marc Valin [Mon, 16 Feb 2009 12:32:44 +0000 (07:32 -0500)]
Preparing for version 0.5.2

10 years agoFixed an overflow in the bit allocation table for large frame stereo
Jean-Marc Valin [Tue, 10 Feb 2009 04:54:52 +0000 (23:54 -0500)]
Fixed an overflow in the bit allocation table for large frame stereo
(need to make that a bit better)

10 years agoBetter allocation rounding as suggested by derf
Jean-Marc Valin [Tue, 10 Feb 2009 04:17:44 +0000 (23:17 -0500)]
Better allocation rounding as suggested by derf

10 years agoSimple check that application supplied packet length is not negative.
Gregory Maxwell [Mon, 9 Feb 2009 06:33:21 +0000 (01:33 -0500)]
Simple check that application supplied packet length is not negative.

10 years agoChanged the allocator to be smarter about the way it allocates
Jean-Marc Valin [Mon, 9 Feb 2009 05:45:48 +0000 (00:45 -0500)]
Changed the allocator to be smarter about the way it allocates
fine energy bits. Also, doing better rounding of the bits.

10 years agoUpdate comment in celt.h to reflect that frame sizes up to 512 are supported.
Gregory Maxwell [Fri, 6 Feb 2009 23:16:41 +0000 (18:16 -0500)]
Update comment in celt.h to reflect that frame sizes up to 512 are supported.

10 years agoLicense for the kiss-fft headers
Jean-Marc Valin [Wed, 4 Feb 2009 04:27:49 +0000 (23:27 -0500)]
License for the kiss-fft headers

10 years agoSome additional documentaion in the celtenc UI.
Gregory Maxwell [Wed, 4 Feb 2009 02:03:04 +0000 (21:03 -0500)]
Some additional documentaion in the celtenc UI.

10 years agoAdditional copyright notices.
Gregory Maxwell [Wed, 4 Feb 2009 01:46:53 +0000 (20:46 -0500)]
Additional copyright notices.

10 years agoInclude missing copyright information in various files, for signoff.
Gregory Maxwell [Wed, 4 Feb 2009 01:36:57 +0000 (20:36 -0500)]
Include missing copyright information in various files, for signoff.

Signed-off-by: Timothy B. Terriberry <tterribe@xiph.org>
10 years agoMinor enhancements to celtenc to avoid silently capping bitrate and allow an
Gregory Maxwell [Tue, 3 Feb 2009 23:22:50 +0000 (18:22 -0500)]
Minor enhancements to celtenc to avoid silently capping bitrate and allow an
adjustable frame size. Indentation fix in cwrs.c.

10 years agoEnlarge CELT pseudo-stack for floating point unless an architecture specific value
Gregory Maxwell [Tue, 3 Feb 2009 21:07:01 +0000 (16:07 -0500)]
Enlarge CELT pseudo-stack for floating point unless an architecture specific value
has been set. This fixes a crash bug for some platforms with 512 sample frames and stereo.

10 years agoThis fixes a budget overrun and segfault for unreasonably low bitrates.
Gregory Maxwell [Tue, 3 Feb 2009 19:35:13 +0000 (14:35 -0500)]
This fixes a budget overrun and segfault for unreasonably low bitrates.

10 years agoThis fixes a couple of issues with celtclient.c and makes it easier to
Gregory Maxwell [Tue, 20 Jan 2009 02:13:37 +0000 (21:13 -0500)]
This fixes a couple of issues with celtclient.c and makes it easier to
change the channel count and deactivate the AEC.

10 years agoPitch now quantised at the band level, got rid of all the VQ code.
Jean-Marc Valin [Wed, 14 Jan 2009 04:04:12 +0000 (23:04 -0500)]
Pitch now quantised at the band level, got rid of all the VQ code.

10 years agoMoved the application of the pitch gain to (un)quant_bands(). This doesn't
Jean-Marc Valin [Wed, 14 Jan 2009 01:48:30 +0000 (20:48 -0500)]
Moved the application of the pitch gain to (un)quant_bands(). This doesn't
change anything to the codec, but will make the next changes easier.

10 years agoSpec file
Christopher Montgomery [Tue, 13 Jan 2009 18:39:53 +0000 (13:39 -0500)]
Spec file

10 years agomode parameters in ietf draft
Jean-Marc Valin [Fri, 9 Jan 2009 22:44:22 +0000 (17:44 -0500)]
mode parameters in ietf draft

10 years agocross-referencing the source code
Jean-Marc Valin [Fri, 9 Jan 2009 16:28:26 +0000 (11:28 -0500)]
cross-referencing the source code

10 years agonothing to see here
Jean-Marc Valin [Fri, 9 Jan 2009 05:23:56 +0000 (00:23 -0500)]
nothing to see here

10 years agoDetails on the reference implementation
Jean-Marc Valin [Thu, 8 Jan 2009 23:00:50 +0000 (18:00 -0500)]
Details on the reference implementation

10 years agoAdded all the source files to the draft
Jean-Marc Valin [Thu, 8 Jan 2009 22:37:22 +0000 (17:37 -0500)]
Added all the source files to the draft

10 years agoMore decoder details
Jean-Marc Valin [Thu, 8 Jan 2009 21:38:01 +0000 (16:38 -0500)]
More decoder details

10 years agoLaplace details
Jean-Marc Valin [Tue, 6 Jan 2009 22:58:58 +0000 (17:58 -0500)]
Laplace details

10 years agoMore text in the IETF draft
Jean-Marc Valin [Tue, 23 Dec 2008 22:19:29 +0000 (17:19 -0500)]
More text in the IETF draft

10 years agoSome details on the MDCT, fixed a bunch of warnings
Jean-Marc Valin [Tue, 23 Dec 2008 19:48:27 +0000 (14:48 -0500)]
Some details on the MDCT, fixed a bunch of warnings

10 years agoTest whether the compiler supports -fvisibility=hidden
Jean-Marc Valin [Tue, 23 Dec 2008 14:53:36 +0000 (09:53 -0500)]
Test whether the compiler supports -fvisibility=hidden

10 years agoUnit tests now #include the relevant source files to prevent symbol visibility
Jean-Marc Valin [Tue, 23 Dec 2008 14:31:39 +0000 (09:31 -0500)]
Unit tests now #include the relevant source files to prevent symbol visibility
problems when building shared libraries only.

10 years agoMiscellaneous compiler warning cleanups.
Gregory Maxwell [Mon, 22 Dec 2008 23:15:42 +0000 (18:15 -0500)]
Miscellaneous compiler warning cleanups.

10 years agoIETF draft for CELT itself
Jean-Marc Valin [Mon, 22 Dec 2008 21:37:45 +0000 (16:37 -0500)]
IETF draft for CELT itself

10 years agoFix ectest to not check a case which isn't guaranteed to work, and which we don't...
Timothy B. Terriberry [Sun, 21 Dec 2008 03:12:18 +0000 (22:12 -0500)]
Fix ectest to not check a case which isn't guaranteed to work, and which we don't use.

When I removed the special case for EC_ILOG(0) in commit
 06390d082dcdfa8addb3dde337543bc0f0ebae44, it broke ec_dec_uint() with _ft=1
 (which should encode the value 0 using 0 bits).
This feature was tested by ectest.c, but not actually used by libcelt.
An assert has been added to ec_dec_uint() to ensure that we don't try to use
 this feature by accident.
ec_enc_uint() was actually correct, but support for this feature has been
 removed and the assert put in its place.

10 years agoBump version to 0.5.1
Jean-Marc Valin [Thu, 18 Dec 2008 01:00:35 +0000 (20:00 -0500)]
Bump version to 0.5.1

10 years agoEnsure that log2_frac() is _really_ an upper bound.
Timothy B. Terriberry [Sun, 14 Dec 2008 19:40:34 +0000 (14:40 -0500)]
Ensure that log2_frac() is _really_ an upper bound.

This version has actually been tested for all 32-bit inputs.

10 years agoFix log2_frac() to return an upper bound, not a lower bound.
Timothy B. Terriberry [Sun, 14 Dec 2008 06:42:37 +0000 (01:42 -0500)]
Fix log2_frac() to return an upper bound, not a lower bound.

10 years agoCompute the real maximum required bits for a split.
Timothy B. Terriberry [Sun, 14 Dec 2008 06:04:24 +0000 (01:04 -0500)]
Compute the real maximum required bits for a split.

10 years agoAdd a SET_LTP call to allow on the fly activation/deactivation of the long term
Gregory Maxwell [Sun, 14 Dec 2008 01:42:03 +0000 (20:42 -0500)]
Add a SET_LTP call to allow on the fly activation/deactivation of the long term
predictor even in high complexity mode (LTP off is more loss robust), fix the
case where turning on LTP on a running stream can cause prediction against
bogus data by delaying the usage of the LTP until enough data has been encoded.

10 years agoMinor pitch handling cleanups.
Gregory Maxwell [Sat, 13 Dec 2008 06:31:32 +0000 (01:31 -0500)]
Minor pitch handling cleanups.

10 years agoNew VQ search is now enabled by default after fixing the last remaining issues:
Jean-Marc Valin [Sat, 13 Dec 2008 01:50:56 +0000 (20:50 -0500)]
New VQ search is now enabled by default after fixing the last remaining issues:
1) the case of the null vector and 2) getting rid of the 32-bit division

10 years agoFix optional synthesis bug in fixed point mode.
Gregory Maxwell [Fri, 12 Dec 2008 21:54:25 +0000 (16:54 -0500)]
Fix optional synthesis bug in fixed point mode.

10 years agoFixes two bugs in get_required_bits() for the split cwrs case.
Jean-Marc Valin [Sun, 7 Dec 2008 19:39:42 +0000 (14:39 -0500)]
Fixes two bugs in get_required_bits() for the split cwrs case.

10 years agoFix a case where the new search can leave us with no pulse left
Jean-Marc Valin [Sun, 7 Dec 2008 04:44:55 +0000 (23:44 -0500)]
Fix a case where the new search can leave us with no pulse left

10 years agoCleaned up some junk left from the last allocator change
Jean-Marc Valin [Sat, 6 Dec 2008 12:52:48 +0000 (07:52 -0500)]
Cleaned up some junk left from the last allocator change

10 years agoNew VQ search nearly fixed-point ready
Jean-Marc Valin [Fri, 5 Dec 2008 05:28:28 +0000 (00:28 -0500)]
New VQ search nearly fixed-point ready

10 years agoNew VQ search works in fixed-point even though there's still some float ops left
Jean-Marc Valin [Fri, 5 Dec 2008 04:26:32 +0000 (23:26 -0500)]
New VQ search works in fixed-point even though there's still some float ops left