[Jim Lesurf[_3_]]

In article , John Williamson
wrote:

High bit rate mp3 files are almost indistinguishable from PCM audio of
the same bit depth and sample rate, even by professionals on decent
playback systems.

Yes and no. :-)

e.g. Some years ago I got some 'AVRO' issued 256k mp3 files made using a
decent pro encoder. These were for one of the celebration events for the
Concergebauw or Haitink IIRC. In general, they sound quite good, albeit
that I have no source LPCM to compare with.

However in some quiet passages the result became quite 'ragged'. I finally
realised that the encoder judgement rule settings was treating a lot of the
low level detail as 'noise floor' simply because it wasn't fully taking
into account the overall low sound level.

So not all mp3 files are equal, even when at the same mp3 rate.

I use them happily for playback when not listening critically. For
critical work, then 24 bit uncompressed or losslessly compressed files
are what is needed, and as I tend to be editing as well, that rules out
native use of FLAC and similar codecs

As per earlier: Given the current cheapness and ubiquity of storage I long
ago decided there was no point in mp3. You can store quite a lot of flac
onto something like a = 4TB NAS, or even onto a couple of = 256 GB SD
card.

So to me nowdays, mp3 seems about as useful than making new 78 rpm discs
of shellac when you can choose alternatives.

Jim

--
Electronics https://www.st-andrews.ac.uk/~www_pa...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
biog http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html
Audio Misc http://www.audiomisc.co.uk/index.html

[John Williamson]

On 06/01/2020 15:26, Jim Lesurf wrote:
In article , John Williamson
wrote:

High bit rate mp3 files are almost indistinguishable from PCM audio of
the same bit depth and sample rate, even by professionals on decent
playback systems.

Yes and no. :-)

e.g. Some years ago I got some 'AVRO' issued 256k mp3 files made using a
decent pro encoder. These were for one of the celebration events for the
Concergebauw or Haitink IIRC. In general, they sound quite good, albeit
that I have no source LPCM to compare with.

I did a few experiments with recordings I had made, starting with 24 bit
PCM, and working down the quality range, and most listeners didn't
notice until I got down to 128 kbps mp3 files.

What did amuse me was an expressed preference by some listeners for the
128 kbs mp3 versions.

Admitted, the band concerned were playing 60s covers and I did take some
care with the settings, but still....
--
Tciao for Now!

John.

[Woody[_7_]]

On Tue 07/01/2020 13:58, John Williamson wrote:
On 06/01/2020 15:26, Jim Lesurf wrote:
In article , John Williamson
wrote:

High bit rate mp3 files are almost indistinguishable from PCM audio of
the same bit depth and sample rate, even by professionals on decent
playback systems.

Yes and no. :-)

e.g. Some years ago I got some 'AVRO' issued 256k mp3 files made using a
decent pro encoder. These were for one of the celebration events for the
Concergebauw or Haitink IIRC. In general, they sound quite good, albeit
that I have no source LPCM to compare with.

I did a few experiments with recordings I had made, starting with 24 bit
PCM, and working down the quality range, and most listeners didn't
notice until I got down to 128 kbps mp3 files.

What did amuse me was an expressed preference by some listeners for the
128 kbs mp3 versions.

Admitted, the band concerned were playing 60s covers and I did take some
care with the settings, but still....

I did similar. I downloaded a recording from Linn Records in flac, m4a,
and mp3. Playing one after another on decent kit and also headphones
there was very little noticable difference between flac and m4a, but
there was a quite sharp drop in HF details with mp3.

--
Woody

harrogate three at ntlworld dot com

[Jim Lesurf[_3_]]

In article , John Williamson
wrote:
I did a few experiments with recordings I had made, starting with 24 bit
PCM, and working down the quality range, and most listeners didn't
notice until I got down to 128 kbps mp3 files.

What did amuse me was an expressed preference by some listeners for the
128 kbs mp3 versions.

Admitted, the band concerned were playing 60s covers and I did take some
care with the settings, but still....

IIRC I read similar reports some years ago somewhere like in JAES. Perhaps
due to becoming habituated to the sound of 'popular' music via mp3. So in
effect, coming to regard the alterations as a 'part of the music'.

I got a similar impression some years ago wrt peak compression on R3 FM
which seemed to make something like a piano sound 'warmer' with more
sustain than via iplayer.

Jim

--
Electronics https://www.st-andrews.ac.uk/~www_pa...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
biog http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html
Audio Misc http://www.audiomisc.co.uk/index.html

[[email protected]]

Jim Lesurf wrote:
IIRC I read similar reports some years ago somewhere like in JAES. Perhaps
due to becoming habituated to the sound of 'popular' music via mp3. So in
effect, coming to regard the alterations as a 'part of the music'.

I got a similar impression some years ago wrt peak compression on R3 FM
which seemed to make something like a piano sound 'warmer' with more
sustain than via iplayer.


Just remember that the MP3 format does not, in any audible way, affect dynamic range compression or
loudness processing. Two factors led to that mass public mis-perception:

1. The timing of MP3 becoming a viable consumer digital format coinciding with the advent of the
digital-era Loudness Wars.

2. That digital audio as a subject itself contains many words with two meanings. IE 'compression'. It is
both something done to level differences in music, and, is a convenient term for the data-reduction
performed in the creation of lossy formats such as MP3.

[[email protected]]

Jim Lesurf:

To follow up:

So what youngsters are showing preference for is not the "sound of MP3",
but the sound of music, and of certain instruments, as modern mastering
techniques make them sound.

[John Williamson]

On 08/01/2020 17:00, wrote:
Jim Lesurf:

To follow up:

So what youngsters are showing preference for is not the "sound of MP3",
but the sound of music, and of certain instruments, as modern mastering
techniques make them sound.

In the case if my experiment, the only processing done to the files was
to alter the file format.

--
Tciao for Now!

John.

[Tobiah]

On 1/8/20 8:53 AM, wrote:
Jim Lesurf wrote:
IIRC I read similar reports some years ago somewhere like in JAES. Perhaps
due to becoming habituated to the sound of 'popular' music via mp3. So in
effect, coming to regard the alterations as a 'part of the music'.

I got a similar impression some years ago wrt peak compression on R3 FM
which seemed to make something like a piano sound 'warmer' with more
sustain than via iplayer.


Just remember that the MP3 format does not, in any audible way, affect dynamic range compression or
loudness processing. Two factors led to that mass public mis-perception:

1. The timing of MP3 becoming a viable consumer digital format coinciding with the advent of the
digital-era Loudness Wars.

2. That digital audio as a subject itself contains many words with two meanings. IE 'compression'. It is
both something done to level differences in music, and, is a convenient term for the data-reduction
performed in the creation of lossy formats such as MP3.

Great points.

[Scott Dorsey]

Jim Lesurf wrote:
In article , John Williamson

e.g. Some years ago I got some 'AVRO' issued 256k mp3 files made using a
decent pro encoder. These were for one of the celebration events for the
Concergebauw or Haitink IIRC. In general, they sound quite good, albeit
that I have no source LPCM to compare with.

However in some quiet passages the result became quite 'ragged'. I finally
realised that the encoder judgement rule settings was treating a lot of the
low level detail as 'noise floor' simply because it wasn't fully taking
into account the overall low sound level.

This is adjustable! You can set some of those thresholds down manually
when necessary.. And yes, for classical music it's necessary.

So not all mp3 files are equal, even when at the same mp3 rate.

This is why we pay mastering engineers. It's just like cutting LPs, you are
trying to pack 10 gallons of music in a 5 gallon container. Something has to
be thrown away. It's better for a person with ears to help make the decision
about what instead of letting the computer do it.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Jim Lesurf[_3_]]

In article , Scott Dorsey
wrote:
However in some quiet passages the result became quite 'ragged'. I
finally realised that the encoder judgement rule settings was treating
a lot of the low level detail as 'noise floor' simply because it wasn't
fully taking into account the overall low sound level.

This is adjustable! You can set some of those thresholds down manually
when necessary.. And yes, for classical music it's necessary.

Understood and agreed. The surprise was that it actually was allowed to
happen when the encoding was being done by professionals.

So not all mp3 files are equal, even when at the same mp3 rate.

This is why we pay mastering engineers. It's just like cutting LPs, you
are trying to pack 10 gallons of music in a 5 gallon container.
Something has to be thrown away. It's better for a person with ears to
help make the decision about what instead of letting the computer do it.
--scott

Again, agreed. But again, a reason why some LPs sound rather poorer than
others, even when coming from professionals.

The advantage of LPCM - flac is that no such judgements are needed.

Jim

--
Electronics https://www.st-andrews.ac.uk/~www_pa...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
biog http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html
Audio Misc http://www.audiomisc.co.uk/index.html

[Trevor]

On 8/01/2020 4:04 am, Jim Lesurf wrote:

Again, agreed. But again, a reason why some LPs sound rather poorer than
others, even when coming from professionals.

So you are saying not all professionals are equal. What a radical
thought! :-) :-)

[Jim Lesurf[_3_]]

In article , Trevor wrote:
On 8/01/2020 4:04 am, Jim Lesurf wrote:

Again, agreed. But again, a reason why some LPs sound rather poorer
than others, even when coming from professionals.

So you are saying not all professionals are equal. What a radical
thought! :-) :-)

Well, I've only ever been a 'professional' wrt the 'user end' of the flow.
Thus I can only judge those 'upstream' on the diversity of the results we
get. :-)

Jim

--
Electronics https://www.st-andrews.ac.uk/~www_pa...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
biog http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html
Audio Misc http://www.audiomisc.co.uk/index.html

[Don Pearce[_3_]]

On Tue, 07 Jan 2020 17:04:46 +0000 (GMT), Jim Lesurf
wrote:

In article , Scott Dorsey
wrote:
However in some quiet passages the result became quite 'ragged'. I
finally realised that the encoder judgement rule settings was treating
a lot of the low level detail as 'noise floor' simply because it wasn't
fully taking into account the overall low sound level.

This is adjustable! You can set some of those thresholds down manually
when necessary.. And yes, for classical music it's necessary.

Understood and agreed. The surprise was that it actually was allowed to
happen when the encoding was being done by professionals.

So not all mp3 files are equal, even when at the same mp3 rate.

This is why we pay mastering engineers. It's just like cutting LPs, you
are trying to pack 10 gallons of music in a 5 gallon container.
Something has to be thrown away. It's better for a person with ears to
help make the decision about what instead of letting the computer do it.
--scott

Again, agreed. But again, a reason why some LPs sound rather poorer than
others, even when coming from professionals.

The advantage of LPCM - flac is that no such judgements are needed.

Jim

Well, I've just been having fun with MP3s. I drive a lot, so I keep
myself occupied with audio books. In my new car I found that some
played and others wouldn't. Long story short - there was a maximum
amount of metadata my car player could deal with before it could no
longer read the file. So I just re-recorded everything minus metadata.

And - almost forgot - this only applied to sticks in the USB socket.
MP3s on CD ROM play with all the metadata present.

d

[Scott Dorsey]

Jim Lesurf wrote:
In article , Scott Dorsey
This is why we pay mastering engineers. It's just like cutting LPs, you
are trying to pack 10 gallons of music in a 5 gallon container.
Something has to be thrown away. It's better for a person with ears to
help make the decision about what instead of letting the computer do it.

Again, agreed. But again, a reason why some LPs sound rather poorer than
others, even when coming from professionals.

When I was starting out, I was cutting 45s... and I was cutting 25 to 30
sides a day... so you can believe that not a lot of care and attention was
being paid on each one. With something like lacquer cutting where you
really are having to make real compromises, it matters a lot about how
much time and care and listening is done. Professionals are sometimes
too expensive to have the time to do it right.

The advantage of LPCM - flac is that no such judgements are needed.

I don't really see a need for flac. For local storage, disks have become
so incredibly cheap that just keeping everything around as PCM files is
no problem. On the other hand, if you want to stream over the network,
nobody has clients to stream flac and you are stuck using the formats
that people can read (namely MP3 and RA).
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Jim Lesurf[_3_]]

In article , Scott Dorsey
wrote:
The advantage of LPCM - flac is that no such judgements are needed.

I don't really see a need for flac. For local storage, disks have
become so incredibly cheap that just keeping everything around as PCM
files is no problem. On the other hand, if you want to stream over the
network, nobody has clients to stream flac and you are stuck using the
formats that people can read (namely MP3 and RA).

Well, as a domestic user I find that my NAS is well over half full even
using flac (and aac for BBC material). And flac also helps when I use a DAP
that only has a couple of mini sd cards[1]. So it seems useful for me. No
doubt my next NAS will be larger... :-)

If I were working in a professional context, I'd probably agree with you,
though.

I 'stream' over my network using a net filing system. Works fine with the
players, etc, on all my Linux and RISC OS boxes. IIUC No need for a media
'client' in the sense you give above, just 'everything is a file'. :-) I'd
probably do the same in a pro situation *if* that was convenient. But
presume this would depend on the context of who else was doing what, and
what others required, etc.

However as per my earlier comment, apart from a few years designing home
audio equipment, my interest tends to be home/domestic audio.

Jim

[1] IIRC one card c400GB and the other 256GB.

--
Electronics https://www.st-andrews.ac.uk/~www_pa...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
biog http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html
Audio Misc http://www.audiomisc.co.uk/index.html