advertisement


YouTube quality selection

steve_1979 said:
http://www.h3xed.com/web-and-internet/youtube-audio-quality-bitrate-240p-360p-480p-720p-1080p
Click to expand...
This doesn't seem to be generally true.

First, there is not one audio track, there are multiple to choose from - however this heavily depends on the video (ID). For videos with one audio track, it's true as there's no other track to pick.

Choosing some acoustic music, Jdy3PbutQLQ (not linking directly due to forum rules):
Code: 
format code  extension  resolution note
249          webm       audio only DASH audio   61k , opus @ 50k, 1.53MiB
250          webm       audio only DASH audio   92k , opus @ 70k, 2.14MiB
140          m4a        audio only DASH audio  129k , m4a_dash container, mp4a.40.2@128k (44100Hz), 4.04MiB
171          webm       audio only DASH audio  143k , vorbis@128k (44100Hz), 3.42MiB
251          webm       audio only DASH audio  171k , opus @160k, 4.03MiB
141          m4a        audio only DASH audio  255k , m4a_dash container, mp4a.40.2@256k (44100Hz), 8.01MiB
160          mp4        256x144    DASH video  111k , avc1.4d400c, 15fps, video only, 3.20MiB
278          webm       256x144    DASH video  130k , webm container, vp9, 15fps, video only, 2.52MiB
242          webm       426x240    DASH video  259k , vp9, 30fps, video only, 4.45MiB
133          mp4        426x240    DASH video  306k , avc1.4d4015, 30fps, video only, 7.31MiB
243          webm       640x360    DASH video  470k , vp9, 30fps, video only, 8.24MiB
134          mp4        640x360    DASH video  603k , avc1.4d401e, 30fps, video only, 8.17MiB
244          webm       854x480    DASH video  867k , vp9, 30fps, video only, 14.18MiB
135          mp4        854x480    DASH video 1103k , avc1.4d401f, 30fps, video only, 16.29MiB
247          webm       1280x720   DASH video 1667k , vp9, 30fps, video only, 27.75MiB
136          mp4        1280x720   DASH video 2205k , avc1.4d401f, 30fps, video only, 31.32MiB
248          webm       1920x1080  DASH video 2802k , vp9, 30fps, video only, 50.84MiB
137          mp4        1920x1080  DASH video 4131k , avc1.640028, 30fps, video only, 58.15MiB
264          mp4        2560x1440  DASH video 9903k , avc1.640032, 30fps, video only, 143.83MiB
271          webm       2560x1440  DASH video 11903k , vp9, 30fps, video only, 136.98MiB
266          mp4        3840x2160  DASH video 22090k , avc1.640033, 30fps, video only, 343.20MiB
313          webm       3840x2160  DASH video 22189k , vp9, 30fps, video only, 382.94MiB
138          mp4        3840x2160  DASH video 25670k , avc1.640033, 30fps, video only, 333.95MiB
36           3gp        320x?      small , mp4v.20.3,  mp4a.40.2
17           3gp        176x144    small , mp4v.20.3,  mp4a.40.2@ 24k
5            flv        400x240    small , h263, mp3  @ 64k
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)
we see 6 audio only tracks (not counting formats with audio bundled) and at least the flash player in my firefox seems to pick the lowest bitrate when streaming 144p video and the highest one for 4K video.

However I can see (hear) why people have problems identifying them - opus is very competitive even at lower bitrates and much better than whatever was there before. I downloaded the worst and best bitrate audio and got 9/10 correctly blind, but it was pretty tough.

That being said, what audio track gets picked depends on the player and, by extension, the browser, so different people will get different results depending on browser, browser versions, operating systems (Safari might prefer AAC, etc.) and who knows what else.

I'm not a web developer, so I don't know the preference specifics behind the HTML5 file API, but I can objectively prove that there is still an audio quality difference between low and high resolution video on at least some setups with videos which offer multiple audio track bitrates.

PS: The tool mentioned on the site seems to hide some tracks, I was able to download everything the API listed (see the output above).
PPS: Measured/detected via sslmitm (youtube forces https) and tcpdump/wireshark, look for mime type in the actual GET.
PPPS: The advertised .MPD file doesn't have all the bitrates - these can be found in the html page itself (in javascript), look for "ytplayer.config =" JSON.
 
jirij - Thanks for your monumental effort in trying to answer this question. At 4K resolution, I can't listen to the Vivaldi recording you site, because my laptop's CPU goes into overdrive when I try to play it at more than 720p (it just stutters).

So, do I hear a difference between 144p and 720p? Sorry, it's Vivaldi's Four Season's and I'm not on a phone ... I'll try comparing some other 4K videos at 720p and lower.
 
BobMaximus said:
jirij - Thanks for your monumental effort in trying to answer this question. At 4K resolution, I can't listen to the Vivaldi recording you site, because my laptop's CPU goes into overdrive when I try to play it at more than 720p (it just stutters).

So, do I hear a difference between 144p and 720p? Sorry, it's Vivaldi's Four Season's and I'm not on a phone ... I'll try comparing some other 4K videos at 720p and lower.
Click to expand...
If there's a difference for you, it will be MUCH smaller than in the past - you might have a hard time noticing it, either because it's low quality (ie. 128K) vs low quality (ie. 256K) :) ... or because it's "that good" in doing lossy compression. Many people don't want to believe it, but lossy compression isn't where it was 10 years ago.

Just use 720p and happily enjoy the music. Or anything lower - chances are that, without deliberate side-by-side comparison, you won't notice a difference. Especially considering that most of the music on youtube isn't really "challenging" for the compression algorithm.

As always, I'm happy to be proven wrong.
 
jirij said:
It definitely affects audio bitrate as well, I had a very nice example where one could clearly hear 240p vs 360p vs 480p vs 720p, but the video has since been taken down (copyright violation, I presume). The difference between 720p and 1080p is very small (might be nonexistent, I haven't checked the bitrates), but ie. 480p vs 720p is huge on some HQ-audio videos.
Click to expand...

Yes agree 100%, Low Res Videos also have very poor sound quality...
 
You must log in or register to reply here.


advertisement


Back
Top