Jump to content

Stop Making CD'S - Lets Use MP3'S Instead.


audioicon

Recommended Posts

  • Members

Yah... I mighta got a few rungs up on my high horse, there, but, really, the cassette was fine to fill its niche as an informal recording format and utterly awful as a delivery format. It was so bad in that capacity (as it worked out in the real world of commerce) that, when I needed a cassette version of an album for some reason or other, I would buy the LP and a good blank cassette and dub the record off myself -- even thought that would add an extra couple of bucks to the cost. (Of course, that would leave a yawning void of 90 minutes or so on side II just crying out to be filled by... something... ;) )

 

Anyhow, like the cassette, I think it's safe to say that 128 kbps MP3s and other, slightly hi-er fi/more 'efficient' codecs at that bitrate were a product that fit a now faded niche... the first DSL broadband hookups seemed to just barely service media at around 128 kbps and it made sense in the late 90s.

 

I definitely do not think 128 kbps Mp3s, AACs, and WMAs are any longer appropriate for distribution at the closing end of this decade... but I still think a well-encoded 128 Mp3, in most cases, can give a cassette recording a very good run for its money. And at, say 160 kbps and up for the advanced formats (AAC, WMA, Ogg) or 190 kbps up for Mp3, there's pretty much no contest.

Link to comment
Share on other sites

  • Replies 132
  • Created
  • Last Reply
  • Members


It wasn't good.
:D

 

Now they did have some hidden benefits - that only become apparent later

 

for instance

 

I had a bashable "walkman" that played a few % fast -- so when I put together a tape indexed to a certain ride (you were at point X in the ride at point Y in this or that tune)...if I wanted to get a performance hit, I'd just use the "faster player" to pace myself.

 

 

Hand rewinding those things on a pencil might help with your envelopments and disengages if you fence (keep that blade circle tight and even kids)

 

think about the hours of frustainment you gained with 10 meters of naked tape hanging out as you hunted the elusive twist!

Link to comment
Share on other sites

  • Members

Regardless, audio data compression does strange stuff to music that I can hear.

For that matter, so does digital in general. Sorry you guys can't hear the two dimensional quality of audio data compression...perhaps you guys need to hit the doob again.

 

It's also common knowledge that most common knowledge is bull{censored}. I'll take wow and flutter, and 45db of dynamic range over plasticized audio. Imagine if music represented by a wav file was a cube, now imagine that it was a plane.

 

And when was the last time I listened to cassette? I'm not sure. When was the last time I listened to an mp3? Probably an hour ago. I just love these cassette vs mp3 threads.

Link to comment
Share on other sites

  • Members

 

It would be my contention that a cassette (even suffering from limited dynamic range, etc) contains far more harmonic content, and just far more content period than an mp3.

 

 

might be good to distinguish between content and information there

 

after all, the white noise between stations on the ole AM dial contains more harmonic content than the program material from the transmitters...it's just not information

Link to comment
Share on other sites

  • Members

you might be thinking information compression as opposed to data compression


(data compression doesn't have to have loss associated with it)

 

Thanks for the correction. What I do know is letting some software deconstruct an audio waveform into a small percentage of it's original and the performing a reconstruction based on a fraction of the original is blasphemy. ;)

Link to comment
Share on other sites

  • Members

 

might be good to distinguish between content and information there


after all, the white noise between stations on the ole AM dial contains more harmonic content than the program material from the transmitters...it's just not information

 

 

On the am dial, it probably isn't "information", but in music harmonic content is.

Link to comment
Share on other sites

  • Members

 

Well, sure... that's... uh... a foregone conclusion. ;)

 

. I'll take wow and flutter, and 45db of dynamic range over plasticized audio. Imagine if music represented by a wav file was a cube, now imagine that it was a plane.


And when was the last time I listened to cassette? I'm not sure. When was the last time I listened to an mp3? Probably an hour ago. I just love these cassette vs mp3 threads

Well, I just transcribed a cassette within the last week (on my 'thousand dollar' deck) so the memory is pretty fresh for me. I have -- quite literally -- thousands of cassettes, live music as well as, er, transcriptions of commercial releases. (I didn't get scruples until the possibility of making near-perfect copies digitally reared its pretty, seductive head. At the point where you could put up a reasonable copy of an album and distribute it online to scores if not hundreds of people an hour, the implications gang-jumped me and I slid over into the IP-sympathetic column.)

 

 

Having some difficulty negotiating the wave file/cube metaphor. Many folks misunderstand how digital audio works and think that the analog wave form created by digital audio manifests quantization steps at a resolution discernible by the human ear. This is not, however, the case with regards to properly recorded and processed digital audio in contemporary, professional formats.

 

A frequent problem when discussing these issues is that many people try to compare analog transcriptions of audio with digital transcriptions -- rather than comparing either to the actual, live, untranscribed source. Of course, digital audio does not sound like the audio coming off a tape machine. The sound coming off the tape machine does not sound like the sound coming into the control room live from the studio floor. At least not in my experience.

Link to comment
Share on other sites

  • Members

Thanks for the correction.

 

Hey no problem! (really more a point/request for clarification than a "correction")

 

It's a distinction in information science that's easy to gloss right over (but can have some big ramifications)

 

 

What I do know is letting some software deconstruct an audio waveform into a small percentage of it's original and the performing a reconstruction based on a fraction of the original is blasphemy. ;)

 

Well I should hope it would be blasphemous -- I mean if "it's also common knowledge that most common knowledge is bullshit" then, as blasphemy, it would be in keeping with that sentiment

Link to comment
Share on other sites

  • Members

 

On the am dial, it probably isn't "information", but
in music
harmonic content is.

 

 

This is why it is important to define the difference between the content and the information

 

If the harmonic content is PART of the music -- if, we have, for instance, various types of distortion (or even just thermal noise)related to the audio, then we can have harmonic content that may or may not be information

 

[hmm, I guess that's actually one of the central issues behind (lossy) information compression - be it in the analog or digital domain]

 

(we can can even have some areas that might require tighter definition with respect to information- I believe it was B2B that brought up "euphonic distortion" -- harmonic content, but not part of the original information...the question there can be If and when and by who's definition is it now part of the musical information stream)

Link to comment
Share on other sites

  • Members

 

It would be my contention that a cassette (even suffering from limited dynamic range, etc) contains far more harmonic content, and just far more content period than an mp3.


Why does a wave file sound better than an mp3?

 

 

I was going to suggest a practical 'experiment' you could perform yourself:

 

Make a high VBR Mp3 recording of a properly mastered, high fidelity CD.

 

Make a VBR recording at the same rate of a well-recorded cassette version of the same material.

 

Compare the data size of the two files...

 

... but I realized that the cassette would likely have a great deal of hiss that would require more data to transcribe into the Mp3 format.

 

Shucks... I thought I was on to something there.

Link to comment
Share on other sites

  • Members

 

My concern is that these reduction schemes are be leaving out vital information inherent in the original waveform. Information too complex for these schemes to be able to reproduce.

 

 

That's pretty much the central concern with "information compression" (though I suppose it's really more truncation or other loss ) in general [be it analog or digital] as we are having to make decisions about significance

 

which is why it's really important to define terms and parameters (and those could even vary from use to use )

Link to comment
Share on other sites

  • Members

 

Having some difficulty negotiating the wave file/cube metaphor. Many folks misunderstand how digital audio works and think that the analog wave form created by digital audio manifests quantization
steps
at a resolution discernible by the human ear. This is not, however, the case with regards to properly recorded and processed digital audio in contemporary, professional formats.

 

 

Yes the steps are very small and decent filters will smooth them out.

 

Dan

Link to comment
Share on other sites

  • Members

My concern is that these reduction schemes are be leaving out vital information inherent in the original waveform. Information too complex for these schemes to be able to reproduce.

 

It's not "too complex" to reproduce -- reproducing it is much 'easier' than removing it in such a way as to leave the data-compressed version sounding psychoacoustically similar.

 

It's simply a matter of where one is willing to draw the line and make the trade-offs.

 

As has been oft-noted here and elsewhere, double blind testing tends to reveal that even golden ear types (especially golden ear types? :D ) have great difficulty or find it impossible to consistently differentiate high rate Mp3s [let's say 256kbps and above just to be arbitrary] and other lossy codecs from the un-data-compressed versions.

Link to comment
Share on other sites

  • Members

So, you agree that cassette sounds better?
;)

 

No! I'd agree that cassettes aren't even fit for reproducing various tonalities of my anus after a day at BurritoFest.

 

If you are concerned with "missing information", you should be extremely wary of a format that "misses" everything above and below audible frequency ranges, and then adds a ridiculous amount of tape hiss to mask the things that it can reproduce.

 

But if you can rock with the cassette, then by all means, my wayward son: rock on. :thu:

Link to comment
Share on other sites

  • Members

 


... but I realized that the cassette would likely have a great deal of hiss that would require more data to transcribe into the Mp3 format.

 

 

ya know it's kinda funny audible type 1 (which is really a type of extremely compressed Mp3 -- that's one area where Mp3 has actually done pretty well, I mean were talking programs that are sometimes in excess of 30 hrs being xfered over the interwebs) ran into a problem there as a lot of that stuff was digitized from tape...so you ran into very opaque noise gating type effects

 

The tapes tended to have a lot of hiss, but at this level of compression, silences are RLE'd -- so during the speaker's talking parts, there would be an underlying hiss, during silences, the hiss would be squelched as the pauses were RLE'd as zero

 

it was really a quite annoying emergent condition caused by the stacking of two kinds of data limitations

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.


×
×
  • Create New...