On Sun, 04 Feb 2018 19:55:14 +1030
Tim <ignored_mailbox@xxxxxxxxxxxx> wrote:
> Allegedly, on or about 4 February 2018, Wolfgang Pfeiffer sent:
> > it's definitely true that with fine tools you can encode to mp3's
> > with a quality so high that for me at least it's difficult to find a
> > difference to the wav's they were encoded from, even with decent
> > stereo equipment. Also true, I'm old, so I might have ruined ears
> > enough to be unable to hear differences where they actually are.
> >
> > Easy test: try this in a dir with wav's, and the command below will
> > (should) code them to mp3's. With the resulting mp3's I'd bet
> > anyone will have difficulties to find a remarkable difference
> > between the wav's and the mp3's ...
> >
> > for f in *.wav; do ffmpeg -i "$f" -codec:a libmp3lame -qscale:a 0
> > "${f/%wav/mp3}"; done
>
> Well, if you're going to encode to an unusually high bit rate (that
> example did it at 320kB/s), I'm going to agree with you (that most
> people won't pick the difference).
Well, that's the only way i found to get decent mp's. ... ;)
>
> However, I find most people encode MP3s to a much lower bitrate, where
> I can hear burbles, squeaks and squealies, and the quieter nuances of
> some music disappears completely.
If you look at the files you seem to have encoded with the oneliner
from my prev. message (with "mediainfo" for example) you should see
that the mp3's have variable bit rate. That's, IINM, and hopefully, one
way to keep silent parts of a song existent and silent, and the louder
ones just as loud as they are.
> There's also a number of old, and
> not very good, codecs around, to which I notice that treble seems to be
> lacking. But it's the added noises that I particularly notice and
> dislike.
>
> One thing I notice with MP3 encoding that I can give it a wave with
> specific lead-in and lead-out time, and the encoded file is missing
> that (screwing up audio comprised of multiple files). Sometimes to the
> point where it's actually slightly cutting off the start of the audio.
> Whatever Audacity was doing behind the scenes tended to do that a lot.
I don't use Audacity for reencoding of existing files. I do that with
ffmpeg. Most of the time. And it just works - although it's hard at
times to find the right switches to get it done.
I added a few notes to my git repo how fading in/out of videos/audio
files, can be done with that tool. Look for "afade" in man ffmpeg-all
and in the notes I uploaded.
These notes are simply taken from what I write down often when finding
what worked on Linux, or Windows. They're mostly unedited, just
commands to use, with a few comments added - definitely no How-To ...
https://github.com/wlfgp/notes/blob/master/ffmpeg.txt
Clicking the "Raw' button on that page might give better
readability ...
HTH,
Regards
--
Wolfgang Pfeiffer
_______________________________________________
users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx
