Despite all my computers being destroyed three days ago except an
ancient one,
I am relieved still to be able to announce another round of release
candidates.
14.4.4 has more bug fixes to the 14.4.2 we all know and love:
o Fix "silence" not to add a random selection of fragments at the end
o Fix coreaudio device name truncation on MacOS X
o Fix delay buffer full flag assigned during drain
o Read mono 8-bit MAUD files with an odd number of samples
o Multi-channel LADSPA plugins now work
o Finish processing the audio data if the process feeding a pipe dies
o Detect and report read errors from short audio files
o Detect and report write errors to audio files
o When an fatal error is detected, stop and exit non-zero in all cases
o Revert "fix" to "Floating point exception in startwrite()" which caused
playing with -t ao to play only the first part of an audio file.
o Make it work on Haiku and with the ucrt and Sun C libraries
o Correct the manual and make its mcompand example work
o Compile successfully on yet more different Unix systems
14.5.1 instead also has fixes to new features in 14.5.0,
a more extensive revision of the manual, and some of the
less intrusive novelties to come:
New features:
o On Windows, support Unicode filenames and command-line arguments
o Automatically read Ogg FLAC (.oga) files with ffmpeg
o Remove echo(s)' limits on number/length of delays and range of
gains/decays
o Remove the limit on the number of echo/echos stages (it was 7)
o Remove "channels"' limit on the number of channels (it was 4)
o Raise "chorus"' limit on the number of stages from 7 to 256
o Remove limits on arguments' ranges for chorus echo echos flanger phaser
o Raise "spectrogram"'s size limit to 1Mx1M and make it need much
less RAM
o Standardize the format of error messages
o Make "--help-effect" show the range and meaning of all parameters
and add diagrams for chorus compand earwax echo echos flanger phaser
Bug fixes:
o Revise the manual, explain the meaning of all parameters, document
undocumented options and change the "mcompand" example to one that
works
o Change "stats" Bit-depth values to reflect what they really means
o Revise all effects' usage messages to a standard format
o Several effects now show their flow diagram in their usage output
o Make "echos" and "chorus" work (they never have)
o Make "bend"'s logarithmic frequency curve more logarithmic
o Stop "silence" from appending a random selection of audio fragments
o Make "fitfit" understand frequencies like "10k"
o Detect and report read errors from truncated audio files
o Detect and report errors when writing audio files
o Check that "tempo" has at least one parameter instead of looping
forever
o Remove the "experimental" and "deprecated" status of formats and
effects
o No longer autodetect "ffmpeg"'s ADX format (too many false positives)
o Reject parameter values of "NaN" instead of dumping core
o Finish processing the audio data if the process feeding a pipe dies
o The LD_LIBRARY_PATH change has been reverted (it bombed on Solaris)
https://codeberg.org/sox_ng/sox_ng/releases/tag/sox_ng-14.4.4-rc1
https://codeberg.org/sox_ng/sox_ng/releases/tag/sox_ng-14.5.1-rc1
On Windows, you still need to "set AUDIODRIVER=waveaudio" beforehand
to be able to use the audio devices; please report any other problems.
Thanks to the many people who have reported bugs, fixed bugs,
tested interim versions and supplied info and samples,
to the historic SoX developers who are still with us and,
as usual, to the admins of the GCC Compile Farm.
The next release is "minor" with more new features, scheduled for May 18th.
M
_______________________________________________
Sox-users mailing list
Sox-users@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/sox-users
