summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* mips C equivalent: Using simple rounding rather than the original biasexp_mips_altJean-Marc Valin2013-11-301-3/+3
| | | | | The bias was only needed because the original code rounds towards -inf on every multiplication.
* Merge branch 'master' into exp_mips_altJean-Marc Valin2013-11-291-1/+1
|\
| * Whoa! This fixes a pretty bad fixed-point stereo regressionJean-Marc Valin2013-11-291-1/+1
| |
* | Fixes the mips_alt buildJean-Marc Valin2013-11-292-3/+3
| |
* | Merge branch 'master' into exp_mips_altJean-Marc Valin2013-11-2933-471/+1003
|\ \ | |/ | | | | | | | | Conflicts: celt/fixed_generic.h celt/pitch.h
| * Fixes OPUS_SET_GAIN (was a bug in MULT16_32_P16())Jean-Marc Valin2013-11-291-1/+1
| |
| * Minor speedups to celt_pitch_xcorr_edsp().v1.1-rc2Timothy B. Terriberry2013-11-271-8/+8
| | | | | | | | This moves some stuff out of the main loop for the unaligned case.
| * Drop the use of --always to git-describeRon2013-11-282-3/+2
| | | | | | | | | | | | | | This is a partial revert of c435f06b308e52d7a5d2aad9eec98a1541832b47, since it didn't actually fix the original problem with shallow clones. It returns just a hash, not something we (currently) can extract as a version string in the form things are expecting.
| * Remove the celt_pitch_xcorr() test code.Timothy B. Terriberry2013-11-261-25/+0
| | | | | | | | Revision control will remember it for us forever.
| * Actually use my NEON code.Timothy B. Terriberry2013-11-261-347/+194
| | | | | | | | | | I'd commented out mine and stuck azanelli's code in for testing, and then accidentally committed it like that.
| * Make celt_pitch_xcorr_edsp() work on ARMv5TE.Timothy B. Terriberry2013-11-263-32/+138
| | | | | | | | | | | | We were assuming that LDR, LDRD, and STRD could be used on unaligned addresses, but this turns out not to be true on really old hardware.
| * Clean up celt_pitch_xcorr() test code.Timothy B. Terriberry2013-11-261-5/+6
| | | | | | | | | | | | | | | | This will be removed shortly, but at least it should be correct before it goes. Makes the code get used unconditionally (when uncommented), and actually frees the buffer it allocates (now using the libopus allocation functions).
| * Silences a warning in fixed-pointv1.1-rcJean-Marc Valin2013-11-251-0/+2
| |
| * Merges surround and stereo intensity stereo thresholds again.Jean-Marc Valin2013-11-251-6/+3
| | | | | | | | The different results for surround were mostly due to a bug in squishyball.
| * Makes SILK surround masking a bit more aggressive in both directionsJean-Marc Valin2013-11-251-4/+8
| | | | | | | | | | This also splits the rate adjustment between SILK and CELT (previous code only changed SILK)
| * Intensity stereo update, partially based on Monty's listeningJean-Marc Valin2013-11-251-12/+8
| | | | | | | | | | | | | | Intensity stereo is now based on equiv_rate and should no longer oscillate in hybrid mode. Also, there is now a separate set of thresholds for surround tuned by Monty. The non-surround thresholds were also updated at low bitrate to be closer to what Monty found for surround.
| * Put custom_encoder_init decl behind CUSTOM_MODES. More softclip tests.Gregory Maxwell2013-11-242-4/+5
| | | | | | | | | | This removes an unused definition warning when compiling without custom modes.
| * Using a more reasonably SILK surround calibration valueJean-Marc Valin2013-11-241-1/+1
| | | | | | | | Still needs more tuning
| * Reducing the thresholds for enabling stereo and higher audio bandwidthsJean-Marc Valin2013-11-241-14/+14
| |
| * Surround low-passing thresholds set to much lower bitratesJean-Marc Valin2013-11-241-3/+3
| |
| * Minor opus_pcm_soft_clip API hardening and tests.Gregory Maxwell2013-11-222-1/+49
| |
| * Add some multistream resets to the tests.Gregory Maxwell2013-11-221-1/+30
| |
| * Merge branch 'silk_pitch_fix'Jean-Marc Valin2013-11-2112-6/+76
| |\
| | * Makes pitch gain control less aggressivesilk_pitch_fixJean-Marc Valin2013-11-201-2/+7
| | |
| | * Using the maximum frequency response of pitch taps instead of maxabsJean-Marc Valin2013-11-205-5/+32
| | |
| | * Constrains accumulated pitch gain to avoid potential instability.Koen Vos2013-11-2010-6/+44
| | | | | | | | | | | | Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
| * | Applies padding to all CBR packets, including PLC and CELT silenceJean-Marc Valin2013-11-212-2/+2
| | |
| * | Adds trivial_example.c to make distJean-Marc Valin2013-11-211-1/+2
| | |
| * | Adds a simple example of Opus encoding+decodingJean-Marc Valin2013-11-211-0/+160
| | |
| * | opus_custom_encoder_init() now only needed for custom modes.Jean-Marc Valin2013-11-201-1/+3
| | |
| * | Add copyright header to test_opus_padding.c.Ralph Giles2013-11-211-1/+27
| |/
| * Additional opus_multistream_packet_(un)pad tests.Gregory Maxwell2013-11-191-0/+36
| |
| * Validate merge argument in repacketizer_demo.Gregory Maxwell2013-11-191-0/+5
| |
| * Validate channel count in opus_demo.Gregory Maxwell2013-11-191-7/+15
| |
| * Add opus_packet_(un)pad tests.Gregory Maxwell2013-11-182-3/+63
| |
| * More make dist fixesRon2013-11-191-1/+2
| | | | | | | | | | Don't include the generated celt/arm/celt_pitch_xcorr_arm-gnu.S in the dist tarball, but do include its celt_pitch_xcorr_arm.s source file.
| * Fix 'make distclean'.Ralph Giles2013-11-191-0/+3
| | | | | | | | | | | | | | | | | | Automake's dependency tracking seems to be confused by our asm generation double-indirection. It's sufficient to have just CLEANFILES = $(CELT_AM_SOURCES_ARM_ASM:.s.in=-gnu.S) to fix the error about 'armopts-gnu.S' not being cleaned, but celt_pitch_xcorr_arm-gnu.S is also not cleaned without raising a warning. Therefore we explicitly clean both sets of ARM_ASM.
| * Don't attenuate hybrid high-band for surroundJean-Marc Valin2013-11-181-5/+8
| |
| * Fixes unpad for multi-streamJean-Marc Valin2013-11-181-4/+9
| |
| * Unpad assertsTimothy B. Terriberry2013-11-181-1/+2
| | | | | | | | Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
| * Fixes previous commitMark Harris2013-11-181-2/+2
| | | | | | | | Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
| * Adds functions for multistream padding/unpadding and single-stream unpaddingJean-Marc Valin2013-11-182-2/+148
| | | | | | | | These are all completely untested.
* | Simulates the non-bitexact changes in exp_mips_opt (not for merging)Rhishikesh Agashe2013-11-299-147/+538
|/ | | | Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* ASM build fixTimothy B. Terriberry2013-11-181-1/+1
| | | | Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Fixes permissions on arm2gnu.plJean-Marc Valin2013-11-181-0/+0
|
* Adds Neon assembly for correlation/convolutionTimothy B. Terriberry2013-11-1843-145/+1515
| | | | | | | Optimizing celt_pitch_xcorr()/xcorr_kernel() which also speeds up FIRs, IIRs and auto-correlations Signed-off-by: Jean-Marc Valin <jmvalin@jmvalin.ca>
* Adds SATURATE16() to the fixed-point debug build tooJean-Marc Valin2013-11-171-0/+10
|
* Fixes SILK surround calibration while fixing other MSVC warningsJean-Marc Valin2013-11-172-3/+3
|
* Add some basic testing for OPUS_{GET|SET}_PREDICTION_DISABLED.Gregory Maxwell2013-11-172-1/+13
|
* oops, fix previous commitJean-Marc Valin2013-11-161-2/+2
|
View Lorry Controller Status