Is lossy outdated in 2019 & onwards?

[digicidal]

Agreed. Also having control over the conversion- dither, input level etc- is important. Archive in the highest bitrate and depth. Streaming is always dependant on the weakest link in the signal chain and whether the DAC is resampling or not.

I personally like 24 bit 48khz PCM but I have been known to enjoy MP3 on ocasion

I don't have SACDs or DVD-A source material... but if I did, then sure. Can't do better than the source.

 

[Julf]

I personally like 24 bit 48khz PCM but I have been known to enjoy MP3 on ocasion

Have you found any commercial mainstream recordings that actually use more than 16 bits of dynamic range?

 

[Julf]

So is there an objective and universally accepted MP3 encoding bit rate under which the lossy MP3 compression is considered unacceptable compared with a 24-32 bit PCM stream with losless compression?

32 bits is pretty pointless considering the best DACs are capable of 21 or 22 bits, and pretty much all recording studios, commercial recordings and listening rooms have less than 16 bits of dynamic range.

 

[digicidal]

Have you found any commercial mainstream recordings that actually use more than 16 bits of dynamic range?

The number of commercial recordings that even require 16 bits of dynamic range can probably all fit on a single shelf - at least that's my guess based on the ones I've heard and or seen analyzed.

 

[Julf]

The number of commercial recordings that even require 16 bits of dynamic range can probably all fit on a single shelf - at least that's my guess based on the ones I've heard and or seen analyzed.

Indeed.

 

[Blumlein 88]

I've asked in a few threads, and the best I've come across are about 80 db SNR. This in only a couple of the very best. Anyone has a suggestion for a recording with more and I'm as they say, all ears.

Theoretically looking at microphone self noise, and venue noise it might be possible with all the stars aligned to do better. I don't think we'll ever exceed 16 bit for dynamic range (excepting totally electronic creations).

I've some very low noise microphones, and you are hard pressed to exercise what they are capable of in the end.

Most issues are FR issues.

 

[Soniclife]

Theoretically looking at microphone self noise, and venue noise it might be possible with all the stars aligned to do better. I don't think we'll ever exceed 16 bit for dynamic range (excepting totally electronic creations).

Anything that is mixed in a DAW could easily hit higher numbers, real or electronic in source, it would have no relationship to real sound, but that's sort of the point. The fact that given this power it's not used shows its not actually wanted by music creators, or listeners.

 

[Julf]

Anything that is mixed in a DAW could easily hit higher numbers, real or electronic in source, it would have no relationship to real sound, but that's sort of the point. The fact that given this power it's not used shows its not actually wanted by music creators, or listeners.

It would have to be electronically generated. I think Blumlein 88's point was that "looking at microphone self noise, and venue noise it might be possible with all the stars aligned to do better" than 80 dB, but exceeding a 16 bit range is going to be very hard.

 

[Soniclife]

It would have to be electronically generated. I think Blumlein 88's point was that "looking at microphone self noise, and venue noise it might be possible with all the stars aligned to do better" than 80 dB, but exceeding a 16 bit range is going to be very hard.

I don't follow. Using a multitrack recording in a 24 bit DAW you can make any track quieter, which lowers the noise as well as the desired signal. You could easily create a recording of real drums, and real acoustic guitar where the guitar is any amount desired quieter than the drums, or the other way around.

 

[Krunok]

Here is mine as an answer to a challenge a few years ago:

----

foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/19 19:45:33

File A: C:\Users\Amir\Music\Arnys Filter Test\keys jangling 16 44.wav
File B: C:\Users\Amir\Music\Arnys Filter Test\keys jangling 16 44_01.mp3

Heh, keys jangling is a kind of "music" which actually suits wav format much better than mp3 while Schubert String Quartet No14 might do mp3 a much better justice. If you choose to retest your hearing skills here is a download link.

 

[Blumlein 88]

Speaking of the sound quality, does anyone know how FM radio compares to CD? I’ve spent some time trying to dig it online but could not find any professionally looking explanation.

I was driving my friend yesterday and he was telling me about a new radio station that he had discovered. When I asked him whether they stream it online he said: “But they broadcast it on FM and surely in terms of audio quality that is superior to any streaming!” I was skeptical about his claim but I didn't know the frequency range and throughput of a typical FM station signal, hence the question. FM signal attenuation aside, is there a way to compare the quality of CD vs FM in ideal conditions? Thanks!

FM in the analog version is 30-15 khz. 19 khz pilot tone indicating stereo had to be filtered. My Nakamichi tape deck has a filter for it. I recorded many albums that way. SNR was probably 65-70 db generally, though some excellent tuners could do better. Tuners like from McIntosh or Magnumm Dynalab or Carver. Better than streaming, I'd say usually.

 

[Julf]

I don't follow. Using a multitrack recording in a 24 bit DAW you can make any track quieter, which lowers the noise as well as the desired signal. You could easily create a recording of real drums, and real acoustic guitar where the guitar is any amount desired quieter than the drums, or the other way around.

As you say, making a track quieter lowers both the noise and the desired signal, so the signal to noise ratio doesn't improve.

 

[Julf]

FM in the analog version is 30-15 khz. [...] SNR was probably 65-70 db generally.
Better than streaming, I'd say usually.

How do those two things match up? Streaming is definitely better than 30-15 kHz and 70 dB SNR.

 

[Blumlein 88]

How do those two things match up? Streaming is definitely better than 30-15 kHz and 70 dB SNR.

Error spectrum. Most music has little or nothing below 30 hz and above 15 khz. Streaming with its codecs if subtracted from the original signal would have more and more annoying residue than FM.

 

[Willem]

FM and lossy internet radio both have their limitations, and those are hard to compare. Low bit rate streaming is probably inferior, but bit rates are going up all the time, with BBC radio 3 now at 320 kbs. At that level internet radio is clearly superior, but also at somewhat lower bitrates. My hunch is that by now the two are already roughly equivalent for many stations, and streaming is only getting better because bitrates are going up. Now that we have roughly equivalent sound quality what clinched the deal for me was that with internet radio you get the world at your doorstep. I have ditched my Quad FM3 in favour of TuneIn streaming via the optical output of a CCA.

 

[Julf]

Streaming with its codecs if subtracted from the original signal would have more and more annoying residue than FM.

Your subjective personal opinion is noted.

 

[Willem]

A measure of subjectivity is inevitable here, because the two kinds of imperfections are so different: what do you find least objectionable? For me the answer was clear, so I went for streaming.

 

[solderdude]

Is MP3 just as obsolete, outdated, deprecated and superseded as vinyl ?

Will I still be able to play them in the coming decade ?

 

[Julf]

Is MP3 just as obsolete, outdated, deprecated and superseded as vinyl ?

Not yet, as there still are a lot more people listening to mp3 than to vinyl.

Will I still be able to play them in the coming decade ?

Yes. The source code for several of the decoders are open source.

 

[Soniclife]

As you say, making a track quieter lowers both the noise and the desired signal, so the signal to noise ratio doesn't improve.

Only the track you made quieter, the other track, say the drums is still at loud at it was, so the range increases.