24bit vs 16bit, the myth exploded!
Aug 17, 2021 at 4:03 PM Post #6,346 of 6,480

71 dB

Headphoneus Supremus
Joined
Sep 17, 2017
Posts
1,697
Likes
561
Location
Helsinki, Finland
Why do you think that Youtube sonic quality should be much inferior to Apple Music, for instance ?
They are both audio streaming other IP. They both can reach 192/24. Youtube uses a little compression, but this is not very audible. They have made a long way since
5 or 6 years ago, when they had a more or less trashy sound.
I don't know what the quality is, but I doubt it is CD quality. In 2013 or so it was something like 192 kbps AAC, am I wrong?
 
Aug 17, 2021 at 4:32 PM Post #6,347 of 6,480

audiokangaroo

Head-Fier
Joined
Jul 26, 2021
Posts
70
Likes
2
Location
France
I don't know what the quality is, but I doubt it is CD quality. In 2013 or so it was something like 192 kbps AAC, am I wrong?
Yes, in 2013, but 8 years are a long time for the Internet. Apple Music is now offering ALAC streaming up to 192/24, which is lossless compression. 8 years ago, Youtube sound was trashy but they have improved quite a lot. If you can listen to a good master on Youtube the sound quality is pretty good.
 
Aug 17, 2021 at 4:47 PM Post #6,348 of 6,480

vergesslich2

New Head-Fier
Joined
Jul 21, 2021
Posts
41
Likes
12
Location
world
[...] Youtube sound was trashy but they have improved quite a lot. If you can listen to a good master on Youtube the sound quality is pretty good.
Get youtube-dl and do "youtube-dl -F URL", without quotes, and you'll see. When you want to see which streams are playing in the browser, you can right click and click stats for nerds. It shows the stream numbers like (251). These numbers are also in the list (at the very left) that youtube-dl outputs, together with samplerates.
 
Last edited:
Aug 17, 2021 at 9:20 PM Post #6,351 of 6,480

KeithPhantom

100+ Head-Fier
Joined
Apr 3, 2020
Posts
385
Likes
203
Location
USA
Even through youtube transcoding, most DSD masters sound impressive and much better than 44.1/16 or even 96/24.
Assessing mono-level modulation by using multi-level modulation and using ears that are better than a spectrum analyzer or an APx555. If you also knew how the broadband spectrum looks for 1-bit noise-shaped modulation you will be scared to death of the amount of quantization noise there is.
 
Aug 17, 2021 at 10:24 PM Post #6,352 of 6,480

audiokangaroo

Head-Fier
Joined
Jul 26, 2021
Posts
70
Likes
2
Location
France
Assessing mono-level modulation by using multi-level modulation and using ears that are better than a spectrum analyzer or an APx555. If you also knew how the broadband spectrum looks for 1-bit noise-shaped modulation you will be scared to death of the amount of quantization noise there is.
I globally prefer PCM to DSD, but the latter can sound really could under some conditions.
I read a few articles about sigma delta modulation and noise shaping. Are you interested in talking about DSD ?
Did you study this subject ?
 
Aug 17, 2021 at 11:01 PM Post #6,356 of 6,480

audiokangaroo

Head-Fier
Joined
Jul 26, 2021
Posts
70
Likes
2
Location
France
This thread is about 16 vs 24-bit PCM, not DSD. For a DSD discussion, you need another thread.
I guess that you have understood that we have been talking about sampling rate rather than bit depth for a few pages.
As for 16 bit vs 24 bit, I think that 16 bit is enough for playback. Maybe 18 bit could be a little better, but 24 bit certainly don't make sense for audio playback.
In my opinion, sampling rate in the real issue as for achieving high fidelity with digital.
Discussing bit depth is not really worth., because going beyond 16 bit will not improve sound in a significant manner.
 
Last edited:
Aug 18, 2021 at 4:04 AM Post #6,357 of 6,480

71 dB

Headphoneus Supremus
Joined
Sep 17, 2017
Posts
1,697
Likes
561
Location
Helsinki, Finland
I guess that you have understood that we have been talking about sampling rate rather than bit depth for a few pages.
As for 16 bit vs 24 bit, I think that 16 bit is enough for playback. Maybe 18 bit could be a little better, but 24 bit certainly don't make sense for audio playback.
In my opinion, sampling rate in the real issue as for achieving high fidelity with digital.
Discussing bit depth is not really worth., because going beyond 16 bit will not improve sound in a significant manner.
Well nice to see that you at least understand bit depth and its effect on sound quality, allthou I disagree with the "maybe 18 bit could be a little better" part. 16 bit already can provide more dynamic range than needed in consumer audio. I think 13 bits is enough in consumer audio. That is almost 80 dB of dynamic range (~20 dB more than vinyl at best) even without dither and with shaped dither perceptual dynamic range can be 90-100 dB. So, 16 bit digital audio is already overkill by 3 bits.

This thread is about 16 vs 24-bit PCM.
Ironically we haven't been discussing much about bit depth here lately. Bit depth clearly isn't as controversial topic as 44.1/48 kHz vs. 96/192 kHz.
 
Aug 18, 2021 at 6:45 AM Post #6,358 of 6,480

gregorio

Headphoneus Supremus
Joined
Feb 14, 2008
Posts
4,127
Likes
2,245
Do the experience. It may look surprising, but in spite of the youtube transcoding and compression I can clearly hear the difference between DSD and 96/24, at least
when the original analogue master was not or only a little processed. Processing can make things at little blurred.
By the way, I believe that Youtube can go up to 192/24 and their compression process is very clean.
No it can't. If you are a Premium Member, the very highest quality that can be selected is AAC 256 kps. For anything uploaded by someone who isn't a premium member playback is restricted to either AAC 128 or Opus or Vorbis no higher than about 160 kbps, sample rate is typically 44.1kHz and never higher than 48kHz. YouTube doesn't support lossless playback, not even of 44/16.

With regard to the sample rate when uploading music videos, YouTube states: "44.1kHz recommended. Higher sample rates are accepted but not required (for example, 48kHz or 96kHz)". - This implies the playback sample rate of music videos will be 44.1kHz, although Youtube doesn't specifically say that. In any event, even with Premium Membership, the playback sample rate cannot be not be higher than 96kHz because that's the highest sample rate supported by the AAC format.

So, what is your experience, that you can clearly hear the difference between a 44.1kHz Opus 160 vs a 44.1kHz Opus 160? Even with the highest quality possible as a Premium Member and even if YouTube supports playback at the highest possible sample rate, the DSD master would have exactly the same sample rate as the 96/24 master!

My experience, from what you've just posted, is that if something says "DSD" and something else says 96/24, then some people will "clearly hear the difference" in these hi-res formats when in fact they are both exactly the same (low-res, lossy) format!!

Youtube uses a little compression, but this is not very audible.
So, you can "clearly hear" a sample rate difference when the sample rates are in fact exactly the same but the relatively huge difference from quite heavy lossy compression is "not very audible"?

G
 
Aug 18, 2021 at 7:59 AM Post #6,359 of 6,480

71 dB

Headphoneus Supremus
Joined
Sep 17, 2017
Posts
1,697
Likes
561
Location
Helsinki, Finland
My experience, from what you've just posted, is that if something says "DSD" and something else says 96/24, then some people will "clearly hear the difference" in these hi-res formats when in fact they are both exactly the same (low-res, lossy) format!!

G
Placebo effect in full action! :smile_phones:
 
Aug 18, 2021 at 8:03 AM Post #6,360 of 6,480

audiokangaroo

Head-Fier
Joined
Jul 26, 2021
Posts
70
Likes
2
Location
France
Well nice to see that you at least understand bit depth and its effect on sound quality, allthou I disagree with the "maybe 18 bit could be a little better" part. 16 bit already can provide more dynamic range than needed in consumer audio. I think 13 bits is enough in consumer audio. That is almost 80 dB of dynamic range (~20 dB more than vinyl at best) even without dither and with shaped dither perceptual dynamic range can be 90-100 dB. So, 16 bit digital audio is already overkill by 3 bits.


Ironically we haven't been discussing much about bit depth here lately. Bit depth clearly isn't as controversial topic as 44.1/48 kHz vs. 96/192 kHz.
I read that Philips wanted tu use 14 bit for the CD format and that Sony advocated for 16 bit.
It was not only for dynamic but also for linearity reasons. 16 bit files are enhanced with a dithering process to improve linearity.
Did you actually experiment with 13 bit to hear how it sounds ?
 

Users who are viewing this thread

Top