Quality of digital recordings

[Pano]

I started doing digital recordings in the 1980s. It was Sony PCM to video tape. Nothing wrong with those. Now I can do them with a hand held Tascam gadget that cost $75.

IMO, it's not the digital recording, it's what happens after that's the problem. Really.

[BeaufortRalph]

to me the satellite radio sounds very bad, even in my automobile.

[Pano]

I agree. The talk stations are awful - and that not just the content.

[aphocus]

Originally Posted by Johnny2Bad

The cutoff frequency is determined by the codec ... There is nothing an encoder can do about it. If the codec and bitrate call for 15k it will be no more than 15k (for example).

LAME, uses a low pass filter to remove high frequency sounds, less sound means less entropy for the encoder to encode and generally sounds better than encoding the entire frequency range sure the mp3's are technically still 44.1kHz, but the list of Bandpass frequencies I had before I didn't just make up, see below for a spectrogram

Originally Posted by Johnny2Bad

Maybe a nitpick but for clarity it seems you use "mp3" where "codec" should be. An mp3 is an mp3 ... Its a patent-protected codec that has not changed since the 90's when it was introduced; devepment goes back earlier. Psycho-acoustic modeling and blind listening were instrumental in its formulation ... Its not new.

The spec defines how to decode the file, it doesn't like all modern codecs define what data specifically you should put in it, this allows the encoder to have flexibility via the psycho-acoustic(or visual) models, and why LAME sounds better than every other mp3 encoder because it has a more accurate psycho-acoustic model compared to the others.

Just like Xvid looks better than Dvix (Both MPEG-4 Part 2 Encoders), iTunes AAC is better than Nero AAC and even better than FAAC (which I assume Youtube uses, and is banned by The Scene, and questionably worse than LAME) and libvoacc is so bad that no one even bothers blind testing at 320kbps because it's that audibly terrible.

The encoder very much does matter, and with the example of libvoacc clearly shows how much more it matters over bitrate.

I can provide you with some samples if you would like to hear the difference?

[KaffiMann]

Originally Posted by BeaufortRalph

to me the satellite radio sounds very bad, even in my automobile.

Yes, they use compression with focus on sibilants (most radio and tv stations do this). You can clearly hear the distorted s when they talk, the excuse often used is they want to make the speech perfectly clear and understandable, even at low volume. The h sound can sometimes be low if there is little or no compression. And they claim it helps on cheap radios and tv's with top end roll off.

Somewhere in my mind I'm sure it's just marketing BS, somewhere there is a salesman thinking "Hah! I sure managed to fool those numpties into buying all those useless gizmo's I had laying about (insert cheeky grin)"

Edit:
It's nothing wrong with digital recording, it's all about how techniques and equipment are applied in the process. 2L use digital recording, they do a splendid job of it in my opinion.

Edit2:
http://www.2l.no/hires/index.html
Nice, I did not know they had freebies :-)

Edit3:
Bridge Over Troubled Water played by Ivar Kleive is awesome! It is one of my very favorite test tracks, very nice low end.

[Charles Darwin]

Originally Posted by Pano

I started doing digital recordings in the 1980s. It was Sony PCM to video tape. Nothing wrong with those. Now I can do them with a hand held Tascam gadget that cost $75.

IMO, it's not the digital recording, it's what happens after that's the problem. Really.

I very much agree with that.
The problem is excessive post-production and it does not make a blind bit of difference if its done digitally or analogue.
It's just that digital is a lot cheaper, easier and these days allows one to fiddle with aspects of the recording which used to be impossible to fiddle with like timing and pitch.
The bane of lifeless records began in the '80s when they started comping one 'perfect' take from 30 or more actual takes during big money productions.

I firmly believe that it is those tiny variations and imperfections that make music come to life.
Pretty much all my favourite records have been recorded as life rather than independently tracked one at a time.

[KaffiMann]

Originally Posted by Charles Darwin

I very much agree with that.
The problem is excessive post-production and it does not make a blind bit of difference if its done digitally or analogue.
It's just that digital is a lot cheaper, easier and these days allows one to fiddle with aspects of the recording which used to be impossible to fiddle with like timing and pitch.
The bane of lifeless records began in the '80s when they started comping one 'perfect' take from 30 or more actual takes during big money productions.

I firmly believe that it is those tiny variations and imperfections that make music come to life.
Pretty much all my favourite records have been recorded as life rather than independently tracked one at a time.

You do not know of any midi I can analyze for this?

If anyone could provide me with midi of a "melody" and the same melody as "music", I will see if I can incorporate those "flaws" into a tune I am playing with. It would mean a lot. Thanks.

Ps.
If needed I can provide a base melody or a more advanced one, if someone skilled at either bass, guitar, piano/synth/organ or harp/xylophone would mind playing the melody with flaws and record it as sound or midi, it would be a tremendous help in my analysis. Collaboration project?

Edit:
Midi is very preferable because it takes a long time to write midi from listening, and it is very much easier to analyze the midi to see which strings or keys are missed, slipped or unintentionally pressed or touched.