Commits · 37942649cc889816d53e54fd98f4bebc279efd63 · Alexander Traud / Opus · GitLab

Snippets Groups Projects

Mar 01, 2008
- Saved 4 kB of stack usage in find_spectral_pitch() by doing the FFT in-place · 37942649
  Jean-Marc Valin authored 17 years ago
  
  (sort of)
  37942649
- Making it easier to debug laplace testcase · 7cdc5a34
  Jean-Marc Valin authored 17 years ago
  
  7cdc5a34
- Stereo pitch search no longer requires twice the memory/complexity. · a556694e
  Jean-Marc Valin authored 17 years ago
  
  a556694e
Feb 29, 2008
- oops. Another 16-bit fix · 86cb7e86
  Jean-Marc Valin authored 17 years ago
  
  86cb7e86
- 16-bit fix · 9e006b6d
  Jean-Marc Valin authored 17 years ago
  
  9e006b6d
- Another C90-fying pass. Fixed some warnings in the process. · d748cd55
  Jean-Marc Valin authored 17 years ago
  
  d748cd55
- Moved mdct state and window to the mode so it can be shared between multiple · 81b38c22
  Jean-Marc Valin authored 17 years ago
  
  encoders and decoders.
  81b38c22
- Just realised half the twiddles in the real FFT weren't even used. Removing · 37131e3a
  Jean-Marc Valin authored 17 years ago
  
  them saves another 1kB for fixed-point (2kB for float or full precision)
  37131e3a
- Saved on stack usage by changing the order of the allocation · 8d4ac155
  Jean-Marc Valin authored 17 years ago
  
  8d4ac155
- Bit of memory reduction in the pitch search. More needed. · 9b9e9862
  Jean-Marc Valin authored 17 years ago
  
  9b9e9862
- Fixed temp arrays that were allocated too large (no change in peak mem) · cac91ec4
  Jean-Marc Valin authored 17 years ago
  
  cac91ec4
- fixed-point: converted intra prediction and folding, unb0rked mixed-precision · 877b1975
  Jean-Marc Valin authored 17 years ago
  
  877b1975
- fixed-point: overflow debugging now works again. · 2aaa0fee
  Jean-Marc Valin authored 17 years ago
  
  2aaa0fee
- Initial support for a managed stack/scratchpad. Still needs some work. · 8600f69f
  Jean-Marc Valin authored 17 years ago
  
  8600f69f
Feb 28, 2008
- fixed-point: energy quantisation is entirely converted now. · 8b2a5923
  Jean-Marc Valin authored 17 years ago
  
  8b2a5923
- fixed-point: more energy quantisation stuff. · adf87e26
  Jean-Marc Valin authored 17 years ago
  
  adf87e26
- fixed-point: a few more bits converted from the energy quantisation. Almost done · 47bf175f
  Jean-Marc Valin authored 17 years ago
  
  47bf175f
- fixed-point: copied the exp2 implementation from Speex, using it for dB2Amp() · 68b02b10
  Jean-Marc Valin authored 17 years ago
  
  68b02b10
- fixed-point: log approximation · 2eaba8ac
  Jean-Marc Valin authored 17 years ago
  
  2eaba8ac
- fixed-point: unquant_energy_mono() has received the fixed-point code from · 98193470
  Jean-Marc Valin authored 17 years ago
  
  quant_energy_mono()
  98193470
- fixed-point: second check-point on quant_energy_mono() · d0316201
  Jean-Marc Valin authored 17 years ago
  
  d0316201
- fixed-point: first check-point on quant_energy_mono() conversion · 43b3537d
  Jean-Marc Valin authored 17 years ago
  
  43b3537d
- fixed-point: alg_quant() entirely converter -- may require revisiting... · 4e1328b0
  Jean-Marc Valin authored 17 years ago
  
  4e1328b0
- fixed-point: (hopefully) last check-point for alg_quant() conversion · 8b158f56
  Jean-Marc Valin authored 17 years ago
  
  8b158f56
- fixed-point: The cross-products in alg_quant() are now all converted. · 642ff944
  Jean-Marc Valin authored 17 years ago
  
  642ff944
- fixed-point: third alg_quant() checkpoint · c9d606f5
  Jean-Marc Valin authored 17 years ago
  
  c9d606f5
- fixed-point: second check-point in the conversion of alg_quant() · ed9e4236
  Jean-Marc Valin authored 17 years ago
  
  ed9e4236
- fixed-point: First check-point in alg_quant() conversion · f675adce
  Jean-Marc Valin authored 17 years ago
  
  f675adce
- fixed-point: Added a ROUND() operator, no real change to the code · a02ca1ee
  Jean-Marc Valin authored 17 years ago
  
  a02ca1ee
Feb 27, 2008
- Oops. Fixed a missing \ in a Makefile.am · 6ba69782
  Jean-Marc Valin authored 17 years ago
  
  6ba69782
- Just in case sizeof(char)!=1 · 79fa8b6c
  Jean-Marc Valin authored 17 years ago
  
  79fa8b6c
- Removed some unused parameters and fixed a couple -W warnings · 5e171490
  Jean-Marc Valin authored 17 years ago
  
  5e171490
- fixed-point: Moved sqrt and cos approximations to mathops.h · 3ca9b1d2
  Jean-Marc Valin authored 17 years ago
  
  3ca9b1d2
- fixed-point: done with mix_pitch_and_residual() though a bit of cleaning up · a847b775
  Jean-Marc Valin authored 17 years ago
  
  wouldn't hurt
  a847b775
- fixed-point: mix_pitch_and_residual() check-point #3 · 1ca07225
  Jean-Marc Valin authored 17 years ago
  
  1ca07225
- fixed-point: another mix_pitch_and_residual() check-point · b50c541d
  Jean-Marc Valin authored 17 years ago
  
  b50c541d
- fixed-point: half-way converting mix_pitch_and_residual() -- just check-pointing · d17edd35
  Jean-Marc Valin authored 17 years ago
  
  d17edd35
- fixed-point: compression factor (alpha) now a 16-bit value (still internally · 9d8d9b3f
  Jean-Marc Valin authored 17 years ago
  
  converted to float though)
  9d8d9b3f
- fixed-point: converted pitch_quant_bands() -- that one was an easy one-liner · 9fbb56a5
  Jean-Marc Valin authored 17 years ago
  
  9fbb56a5
- fixed-point: converted compute_pitch_gain() and removed the energy-based · ff74e396
  Jean-Marc Valin authored 17 years ago
  
  weighting that didn't seem to help anyway.
  ff74e396