2010-06-09 Jean-Marc ValinRemove unnecessary free of global stack on error
2010-06-07 Jean-Marc ValinCompletely disabling the tf code at low bitrate
2010-06-06 Jean-Marc ValinEncoding the tf parameters after coarse energy
2010-06-05 Jean-Marc ValinDoing the log2Amp() just in one place
2010-06-04 Jean-Marc Valinnothing to see here
2010-06-04 Jean-Marc ValinCompletely removed the old allocation table.
2010-06-04 Jean-Marc ValinBase everything on the "standard allocation table"
2010-06-03 Jean-Marc ValinUsing allocation table for "standard" frame sizes
2010-06-03 Jean-Marc ValinAllocation table now in bits/sample
2010-06-03 Jean-Marc ValinChanging the allocator resolution to 1/8 bit
2010-06-01 Jean-Marc ValinFix for PLC crash when using large frames
2010-06-01 Jean-Marc ValinTrying not to crash on bit errors
2010-06-01 Jean-Marc Valinremoved stupid code in transient_analysis()
2010-05-31 Jean-Marc ValinFixed stupid bug for single-sample bands
2010-05-30 Timothy B.... Change ec_{enc|dec}_bit_prob to take probabilities...
2010-05-30 Timothy B.... Provide direct implementations ec_{enc|dec}_bit_prob...
2010-05-29 Jean-Marc ValinCleaning up VBR to be frame-size independent
2010-05-29 Jean-Marc ValinFolding tuning
2010-05-29 Jean-Marc ValinRD optimisation now takes bitrate into account
2010-05-28 Jean-Marc ValinSimplification to the transient detection code
2010-05-28 Jean-Marc ValinMore simplifications to the Viterbi code
2010-05-28 Jean-Marc ValinSimplifications to the tf_res RD code
2010-05-28 Jean-Marc ValinViterbi-based RD optimization of the tf_res decisions
2010-05-28 Jean-Marc Valinchanging probability of first tf resolution symbol
2010-05-28 Jean-Marc ValinAdaptive time-frequency resolution
2010-05-27 Jean-Marc ValinAdds a range coder call to encode a single bit with...
2010-05-26 Jean-Marc ValinFixed a long-standing rare mismatch
2010-05-26 Jean-Marc ValinRe-organised the special case for N==1
2010-05-26 Jean-Marc ValinBunch of fixes for frames of 2.5 ms.
2010-05-26 Jean-Marc ValinMore recombining "infrastructure"
2010-05-26 Jean-Marc ValinCode for recombining bands (not enabled yet)
2010-05-24 Jean-Marc ValinRemoving the rest of the >32-bit PVQ code
2010-05-21 Jean-Marc ValinMinor code simplifications
2010-05-21 Jean-Marc ValinDoing folding from the highest freq possible
2010-05-21 Jean-Marc ValinIncreasing the temporal resolution of transients
2010-05-21 Jean-Marc ValinImproving the transient detection
2010-05-21 Jean-Marc ValinSplitting transients in time domain
2010-05-21 Jean-Marc ValinMoving code to quant_band()
2010-05-21 Jean-Marc ValinTuning/cleanup of angle quantization
2010-05-21 Jean-Marc ValinGetting rid of PVQ-level split
2010-05-20 Jean-Marc ValinMost 5ms bands now have even width
2010-05-19 Jean-Marc ValinTuned the 5ms mode to have a band at 4 kHz and one...
2010-05-19 Jean-Marc ValinImplemented CELT_SET_START_BAND
2010-05-19 Jean-Marc ValinSupport for passing your own ec state to the decoder
2010-05-19 Jean-Marc ValinSupport for passing your own ec state to the encoder
2010-05-19 Jean-Marc ValinSaving ~1 kB by using chars for the alloc table
2010-05-19 Jean-Marc ValinSharing more code between encode and decode (bis)
2010-05-15 Jean-Marc ValinMerged encoding/decoding of mono/stereo
2010-05-14 Jean-Marc ValinTrying to prevent PVQ-level splits
2010-05-14 Jean-Marc ValinEntropy-coding the new split parameter.
2010-05-13 Jean-Marc ValinNew band splitting method based on the stereo code
2010-05-12 Jean-Marc ValinStereo quantization also made more generic
2010-05-12 Jean-Marc ValinMaking stereo code a bit more generic
2010-05-12 Jean-Marc ValinTrivial code cleanup
2010-05-11 Jean-Marc ValinProper use of logN and logM for the offsets
2010-05-11 Jean-Marc ValinMaking stereo code use [un]quant_bands() as well
2010-05-11 Jean-Marc ValinAllocation adjustment code in quand_band().
2010-05-09 Jean-Marc ValinCleaning up the folding code.
2010-05-08 Jean-Marc ValinAPI change: optional resynthesis
2010-05-08 Jean-Marc ValinFixed pitchEnd wrt to variable frame size. It should...
2010-05-08 Jean-Marc ValinMore work on variable frame size (getting rid of FRAMES...
2010-05-08 Jean-Marc ValinChanging the encoder API to add the frame size
2010-05-08 Jean-Marc ValinOops! Fixed the fixed-point build
2010-05-08 Jean-Marc ValinFix for very short frame size (M=1)
2010-05-08 Jean-Marc ValinYet another step towards variable frame size
2010-05-08 Jean-Marc ValinMore preparation work for variable frame size
2010-05-08 Jean-Marc Valinreplaced instances of mode->nbShortMdcts by the dynamic...
2010-05-08 Jean-Marc ValinAllocation table stored with BITRES accuracy based...
2010-05-08 Jean-Marc ValinshortWindow hadn't been used for a while
2010-05-08 Jean-Marc ValinShrinking the MDCT's table by separately rotating by...
2010-04-26 Jean-Marc ValinMaking the band definition the same at all frame sizes.
2010-04-21 Jean-Marc ValinMinor tweak to the band layout offset
2010-04-21 Jean-Marc ValinRe-enabled special stereo code for N=2
2010-04-21 Jean-Marc ValinBands are now (again) a multiple of the number of short...
2010-04-20 Jean-Marc ValinSkipping transient shape renormalisation in the encoder...
2010-04-20 Jean-Marc ValinBump bitstream version now that we broke everything
2010-04-20 Jean-Marc ValinTuning the amount of spreading based on Koen's feedback
2010-04-19 Jean-Marc ValinImproved transient_analysis() by adding one frame of...
2010-04-19 Jean-Marc ValinRemoved unnecessary calls to log2Amp()
2010-04-19 Jean-Marc ValinDisabling resynthesis when not needed (need to remove...
2010-04-18 Jean-Marc ValinRe-introducing the successive spreading rotations,...
2010-04-17 Jean-Marc ValinCorrected some non-sensical code
2010-04-17 Jean-Marc ValinConverted a few double-precision constants to single...
2010-04-14 Jean-Marc ValinMaking sure compute_allocation_table() doesn't overrun...
2010-04-14 Jean-Marc Valinversion string set to 0.8.0-git
2010-04-14 Jean-Marc ValinChanged compute_allocation_table() so it handles ebands...
2010-04-07 Jean-Marc ValinFix small mode leak (forgot to free the new logN array).
2010-04-07 Jean-Marc ValinCaching log2_frac(N, BITRES) in the more data to save...
2010-04-07 Jean-Marc ValinA few minor optimisations (compute_allocation, denormal...
2010-04-07 Jean-Marc ValinA bit more tuning on the pseudo-frac-Hadamard. Also...
2010-04-03 Jean-Marc ValinDoing the spreading with a "pseudo-fractional-Hadamard...
2010-03-20 Jean-Marc Valinfixed-point: improving accuracy of the energy prediction
2010-03-20 Jean-Marc Valinfixed-point: slightly increasing the decoder accuracy...
2010-03-16 Timothy B.... Update the log2 approximation to accomodate the increas...
2010-03-16 Jean-Marc Valinfixed-point: Further reduction in mismatch
2010-03-16 Jean-Marc Valinfixed-point: reducing the mismatch in the folded part
2010-03-13 Jean-Marc ValinIncreasing log energy resolution
2010-03-13 Jean-Marc ValinUsing the mean energy even when start>0
2010-03-12 Jean-Marc ValinImproves fixed-point precision for quant_coarse_energy
2010-03-01 Jean-Marc ValinFix quant_bands() for start != 0