h a l f b a k e r yNaturally, seismology provides the answer.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
Not to promote reverting to retro technology, but it occurred to me that the tape days could have been much better than they were.
This method of recording uses two tracks per audio channel using a format which is akin to modern bit-stacking, except it doesn't have to stick to powers of two. For
recording:
(1) Apply Dolby noise reduction.
(2) Determine the highest number of discrete levels which can be reliably stored and reproduced with the tape medium to get a quantization factor.
(3) Convert the signal to 16-bit (or more) digital samples.
(4) Divide the signal by the quantization factor, and subtract the (integer) result from a copy of the signal.
(5) Record the original signal on one track.
(6) Amplify the signal resulting from the subtraction and record that in an adjacent track.
(7) One margin of the tape can be used to record a steady frequency to synchronize the tape speed during playback (using a phase-locked loop).
Couple my idea with this one and you've got a winner!
Sound_20_22Clarifier_22
[doctorremulac3, Jan 29 2017]
Digital LP
A much simpler solution [pocmloc, Feb 03 2017]
[link]
|
| |
Use one track to record the sound, and the other to record the noise. On playback, preamplify both, then subtract the noise from the sound before the main amp. |
|
| |
If any group back in the 80s had had the presence of minds to have one of the backing vocalists hiss all through each track, they'd have solved the problem. |
|
| |
Incidentally, there must be a market for an audio component that can inject hiss into digital music, for nostalgia. |
|
| |
//Use one track to record the sound, and the other to record the noise.// |
|
| |
The worrying thing is that I'm not sure whether [8th] is smart enough to suggest that facetiously, or dumb enough to suggest it in earnest. |
|
| |
Us ? Facetious ? Perish the thought ... |
|
| |
hmm... so the difference between doing this and simply recording then playing an analog track is ... ? Unless you're missing |
|
| |
6.5) record the differential into digital. |
|
| |
@Ian Tindale,
If there was no noise floor to deal with, then there would be infinite headroom, assuming that even infinitesimal signals to the record head could cause magnetization. Limited headroom was what I had always heard to be the major reason for moving to digital audio.
@8th of 7,
//Use one track to record the sound, and the other to record the noise.//
The only noise you'd be able to isolate from the eventual playback signal, is the tape hiss. I actually considered that option first. The difficulty being that you'd need to play the sound immediately after recording it, so you could do the noise subtraction. This would require use of a delay line, the length of which would depend on how closely a playback head could follow a record head (hopefully just a few milliseconds). I would worry that the fidelity might suffer. |
|
| |
// I actually considered that option first. // |
|
| |
That is, in its way, rather worrying. |
|
| |
but then if I play it back faster and slower as it gets caught in
the pin thingy in the middle, how do I reproduce the mickey-
mouse effect? |
|
| |
Why not, as an alternative improvement, instead of directly
recording each channel to a track on the tape, record the
average and the difference of the left and right channels?
The difference values would often be less than the average
values, which I think would allow a narrower track. Then
the average track could be wider, meaning higher SNR,
right? Or the tape could be narrower overall, or you could
put more tracks on it. This is similar to the compression
that is commonly done for video where there's one channel
for how bright the image is at a particular spot onscreen (in
analog, a point along a scanline) and there are two channels
for the color, which have less bandwidth, all of which adds
up to less bandwidth than separate red, green, and blue
channels would need for the same quality. |
|
| |