cjacek said:
I, frankly, shudder at the mere thought of capturing sounds as a series of non-linear, mathematically representated and approximated "snapshots" of the original. Doesn't sit well with me at all.
Daniel, I respect you for candidly saying what your problem is with digital recording, which seems to be at the conceptual level. Can I respectfully try to speak to this point?
Just like digital, analog audio recording is also only approximate. So it's only a question of HOW approximate which is important.
Both analog and digital only have to be accurate enough so that we cant tell the difference. Any more than that is a waste of tape and machine design (with analog) and a waste of processing and storage power (with digital).
Both formats are capable of recording with an accuracy waaayy beyond what is needed, although (you are going to hate this) analog tape can only record a limited number of octaves in one hit without level anomolies. But in normal circumstances it's not too much of a problem, as we know. It was a huge problem with trying to use analog tape for home video recording, but they eventually got around the problem with some very clever design, but that is another story.
If you want analog to go way above audible frequencies and still have a good bottom end, you cant do it. There is a tradeoff that has to be made. Other techniques like FM recording are needed,(as in hifi home videotapes) but that takes it beyond normal analog as we know it..
But with digital, there is no such tradeoff. You really can record from the lowest bass up to ultrasonic frequencies and not have to compromise.
That's why it's not right to speak of "approximate" as some kind of valid criticism. It's all approximate. The question is what is accurate enough for the purpose. If it's too approximate we just make it more precise. If it's too precise it's a waste of resources.
The whole world of science is based on measurement
within a certain level of accuracy.
Just like the specs on an analog machine, there should be a tolerance level stated, like + or - 2db etc. That tells you HOW accurate, or inaccurate it is. If you accept the approximations of analog tape, why not accept the approximations of digital? Nothing is perfect. But things can be made to be more than adequate for the purpose. Here as in everything else in life.
If your argument is about the supersonic frequencies that we cant actually hear but in some way sense, then sure, CD limits these sharply. But if you're really fussy, with a higher sample rate, as with SACD or DVD Audio you get your supersonics back. Digital can go supersonic even better than analog tape AND not sacrifice the bass response. (BTW SACD has been a poor seller because most people dont feel the need for the extra fidelity. We are talking about very minor improvements at best anyway)
The final proof is the listening but it's good to understand that digital's "approximations" can be made so very close that no one, but no one will be able to tell the difference, except maybe bats, cats and dogs, and most of those I've known personally dont buy a lot of recorded music anyway.
Stay with analog for all sorts of good reasons but in a final shootout, the best analog recording will not trump the best digital recording. That's reality.
Regards, Tim G
, nevermind.
