Skip to main content

Recent Posts

2
This is  a list of formats available for this video: https://www.youtube.com/watch?v=2w5B1Ne-fy4
It's new, from a channel that gets viewers by the millions so it's uploads get transcoded to all available formats

Spoiler (click to show/hide)

So this is the best all videos can get transcoded to: opus @160kbps, or aac@192kbps.... all of them vbr, so the actual bitrate changes from one video to another.

Although you can upload flac (and pcm) in hi-res, you don't get hi-res when playing the video, and the original upload is not available to anybody else but google.

And this is what you get on a 480p video with less than 10 views on a really small channel:
Code: [Select]
format code  extension  resolution note
139          m4a        audio only DASH audio   51k , m4a_dash container, mp4a.40.5@ 48k (22050Hz)
140          m4a        audio only DASH audio  130k , m4a_dash container, mp4a.40.2@128k (44100Hz)
160          mp4        192x144    DASH video  108k , avc1.4d400b, 30fps, video only
133          mp4        320x240    DASH video  242k , avc1.4d400c, 30fps, video only
134          mp4        480x360    DASH video  554k , avc1.4d401e, 30fps, video only
135          mp4        640x480    DASH video 1025k , avc1.4d401e, 30fps, video only
17           3gp        176x144    small , mp4v.20.3, mp4a.40.2@ 24k
36           3gp        320x240    small , mp4v.20.3, mp4a.40.2
18           mp4        480x360    medium , avc1.42001E, mp4a.40.2@ 96k
43           webm       640x360    medium , vp8.0, vorbis@128k (best)
128kbps AAC and Vorbis.
3
Thanks for the Reply.

Is the LAME tag a field in or part of the ID3v1, ID3v2, or APE tag system or is it a completely different tag?

If it is different, is there any software that can read it? An easy way to transfer the data generated into a text file or something that I can then add to the replaygain fields in an ID3 tag via mp3tag.

Or is just using MP3gain (APE) the better more supported way to go?

Thx
4
The consequences of this sort of thing to the high rez audio and other snake oil demonstrations seems pretty grim. There ain't no such thing as real 24/96 audio on YouTube, it seems. If someone seems to claim otherwise, they are either ignorant or claiming falsely, or lying.
This is  a list of formats available for this video: https://www.youtube.com/watch?v=2w5B1Ne-fy4
It's new, from a channel that gets viewers by the millions so it's uploads get transcoded to all available formats

Code: [Select]
[info] Available formats for 2w5B1Ne-fy4:
format code  extension  resolution note
139          m4a        audio only DASH audio   49k , m4a_dash container, mp4a.40.5@ 48k (22050Hz), 3.46MiB
249          webm       audio only DASH audio   57k , opus @ 50k, 3.75MiB
250          webm       audio only DASH audio   77k , opus @ 70k, 5.34MiB
171          webm       audio only DASH audio  121k , vorbis@128k, 8.22MiB
140          m4a        audio only DASH audio  128k , m4a_dash container, mp4a.40.2@128k (44100Hz), 9.23MiB
251          webm       audio only DASH audio  143k , opus @160k, 9.52MiB
17           3gp        176x144    small , mp4v.20.3, mp4a.40.2@ 24k
36           3gp        320x180    small , mp4v.20.3, mp4a.40.2
43           webm       640x360    medium , vp8.0, vorbis@128k
18           mp4        640x360    medium , avc1.42001E, mp4a.40.2@ 96k
22           mp4        1280x720   hd720 , avc1.64001F, mp4a.40.2@192k (best)

So this is the best all videos can get transcoded to: opus @160kbps, or aac@192kbps.... all of them vbr, so the actual bitrate changes from one video to another.

Although you can upload flac (and pcm) in hi-res, you don't get hi-res when playing the video, and the original upload is not available to anybody else but google.

5
--replaygain-accurate differs from the default settings by computing replaygain levels slightly more accurately (and slowly) and also I think calculating clip levels. However, these replaygain values are still stored in the (essentially obsolete) LAME tags, not the now-universal REPLAYGAIN tags. That's why you thought that nothing happened.
6
Using Lame Front End with Lame 3.99.5 and the option --replaygain-accurate. Am wondering if it actually does anything?

To test this I encoded the same .wav file to an mp3 with no replaygain added (--noreplaygain) three times and also to a flac with replaygain added.

I then used mp3gain to replaygain one of the mp3s to 92dB and one to 89dB.

Then I checked all the tags in mp3tag to see if there was replaygain info. There was in the FLAC tag under REPLAY_GAIN_etc and the two mp3gained mp3s in the APEv2 tag. However neither the lame encoded mp3s had replaygain info in them.

Then I played all the files in foobar toggling the playback settings for track replaygain on and off. The wav file seems to have been recorded around 95-96dB so it is loud. The FLAC file is hugely different when toggling the replay on and off. The lame replaygain-accurate encoded mp3 and the --noreplaygain encoded file sounded no different from the wav file with the toggle on or off...hence it appears that NOTHING ACTUALLY HAPPENED!

Am I right about this does anyone know? Is this just an artifact of a setting that used to work?

PS-FYI:
When I played the 4 mp3s in foobar with replaygain OFF (playback), the two as I just mentioned above sounded exactly the same and the 92dB (mp3gain) was less loud than those two and the 89dB was the least loud as expected.

I played them with replaygain playback off because I think if there is any replaygain info at all foobar automatically sets the loudness to 89dB so the 89dB and 92dB sounded the same as each other in foobar with this playback setting...although they were both equally less loud than the other two.
7
I think youtube will never provide untranscoded sources for free and anonymous download. The medium bitrate policy should be a copyright concern.
8
My first thought is that I don't care what they accept, I care about what they deliver.

If YouTube or similar services lose quality, where does it happen and is any of it under the viewer/listener's control?

My test system uses a M-Audio 24/192 running at 96 KHz sampling looped back to itself. The Audio Rightmark program finds about 100 dB dynamc range and response within 0.1 dB over the normal range, and well above it.  It can definitely tell the difference between high sample rates, high dynamic range, and vin ordinaire.

I've done one set of tests where I looked at the web browser as a potential source of quality loss. I was not disappointed, in that my 24/96 test files were obviously downsampled to more like 16/44 by the current release of Chrome running under Win 7/32.

The next easiest thing was to search YouTube for uploaded test files, things with page names that suggest pure tones. I  downloaded them to see how they look. Again I was not disappointed in that while frequency response wasn't too badly smashed below 20 KHz, dynamic range was way less than RedBook, on the order of 70 dB in most cases.

The consequences of this sort of thing to the high rez audio and other snake oil demonstrations seems pretty grim. There ain't no such thing as real 24/96 audio on YouTube, it seems. If someone seems to claim otherwise, they are either ignorant or claiming falsely, or lying.

Any other evidence or insights will be welcomed.
10
General - (fb2k) / Re: auto update playlists
Last post by lvqcl -
I do a search of my library using the syntax - (%GENRE% HAS POP)  AND NOT (%GENRE% HAS CHRISTMAS) AND NOT (%GENRE% HAS DANCE), which produces a playlist. I save that playlist using the ...create autoplaylist.
This autoplaylist should be automatically updated when you add a new song (if you didn't disable library monitoring and foobar2000 is running when you add new song).

All playlist files are also stored in D:\All My Music, example : All Pop.fpl
This file is just a regular playlist file, it cannot be auto-updated.