Page 5 of 5
Re: Saving (exporting) to non-Wav is needlessly tedious
Posted: Fri Dec 14, 2018 3:07 pm
by tcmullet
Hmm... We're getting closer... I will be doing "amplify" and "normalize", but nothing else other than cutting. (Input is 16-bit; output is 16-bit.) So, in that context, what dithering should I set the high-quality (for use by Export) to? Shaped or Off?
Re: Saving (exporting) to non-Wav is needlessly tedious
Posted: Fri Dec 14, 2018 3:19 pm
by steve
tcmullet wrote: ↑Fri Dec 14, 2018 3:07 pm
Hmm... We're getting closer... I will be doing "amplify" and "normalize", but nothing else other than cutting. (Input is 16-bit; output is 16-bit.) So, in that context, what dithering should I set
Same as the defaults (as shown in the screenshot:
https://manual.audacityteam.org/man/qua ... ences.html) "best quality" + "shaped".
Re: Saving (exporting) to non-Wav is needlessly tedious
Posted: Fri Dec 14, 2018 3:38 pm
by tcmullet
So yes, my "amplify" and "normalize" are "gain changes". Therefore will leave it set to "shape". BUT, if I was to do cuts-only (no volume tampering), then turn dithering off. Got it. Thanks!
But it does raise the question as to whether there would be any point to proceed with a resample-export SOLELY doing the resample (with dither off), then open the new file (which is now in my target 44.1k), do my gains, and re-export changing the dither to Shape. Would these extra steps be useless?
Re: Saving (exporting) to non-Wav is needlessly tedious
Posted: Fri Dec 14, 2018 3:50 pm
by steve
tcmullet wrote: ↑Fri Dec 14, 2018 3:38 pm
So yes, my "amplify" and "normalize" are "gain changes". Therefore will leave it set to "shape". BUT, if I was to do cuts-only (no volume tampering), then turn dithering off. Got it. Thanks!
That's correct.
tcmullet wrote: ↑Fri Dec 14, 2018 3:38 pm
whether there would be any point to proceed with a resample-export SOLELY doing the resample (with dither off), then open the new file (which is now in my target 44.1k), do my gains, and re-export changing the dither to Shape.
Assuming that you want to change the sample rate and change the gain, then it would probably be less effort doing both at the same time - no benefit in adding an extra export / import step.
Ideally you would do ALL editing + processing + resampling + whatever else, all in 32-bit float format, then export to the final 16-bit with dither as the very last step.
Re: Saving (exporting) to non-Wav is needlessly tedious
Posted: Fri Dec 14, 2018 3:57 pm
by tcmullet
I appreciate your attention to detail and a consciousness of quality. May I ask one more question here? (Maybe a new thread would be better, but I'll try here.) When I try to open an audio file with an extension of .ac3, Audacity reports "---.ac3 is a Dolby Digital audio file. Audacity cannot currently open this type of file. You need to convert it to a supported audio format, such as WAV..." It appears that just as resampling involves parameters that can affect the output quality, AC3-to-WAV conversions also have parameters (I am told). Can you recommend a way that gets the best quality 16-bit wav output? (I respect your advice.)
Re: Saving (exporting) to non-Wav is needlessly tedious
Posted: Fri Dec 14, 2018 4:08 pm
by steve
tcmullet wrote: ↑Fri Dec 14, 2018 3:57 pm
Can you recommend a way that gets the best quality 16-bit wav output?
I'd just use Foobar2000 (if I was using Windows). I expect that will give as good results as anything else.