Content-Type: text/plain; charset=US-ASCII
>>>>> "Chris" == Chris Zwilling <chris@cloudnet.com> writes:
Chris> There are many problems with tape. Stretching and speed
Chris> drift have been touched on. Tape alignment is another one.
Chris> How about wow and flutter? Then we have abrasion - taking
Chris> the emulsion right off the tape. If we limited ourselves to
Chris> one channel on a stereo deck, we could have RAID-1 type
Chris> redundancy. ;)
Chris> Of course we could try a simple experiment. Take a sound
Chris> sampled from CD. Record it to the tape deck with the sound
Chris> card (dd if=sample.raw of=/dev/dsp bs=512
Chris> count=128). Resample from the tape and compare the results
Chris> with the original sample (dd if=/dev/dsp of=sample2.raw
Chris> bs=512 count=128). My guess is that although the files will
Chris> sound similar the byte differences will be quite large. You
Chris> can generate some random data by catting /dev/random to a
Chris> file.
There's also the problem that the ADCs found on most soundcards are
pretty ordinary. A friend was working on a new sound-compression
project, and as part of his work sampled the audio from the CD-ROM
drive. He then compared the result to the raw digital data on the CD
and found that when using 16-bit sampling the bottom 5 bits were
"random". This means that either the CD-ROM DAC was crap, the audio
cable was noisy, the sound-card's ADC was crap, or all of the
above. When you consider the likely price of the ADC used on the
average sound-card and compare it to "scientific" ADCs, you can
understand that the sound-card's ADC must leave a lot to be desired.
This could put a bit of a damper on what is achievable, but then again
this might not be a limiting factor depending on the encoding scheme
used. :)
- Andrew
--
#!/usr/bin/env python
print(lambda s:s+"("+`s`+")")\
('#!/usr/bin/env python\012print(lambda s:s+"("+`s`+")")\\\012')
print(lambda x:x%`x`)('print(lambda x:x%%`x`)(%s)')
--pgp-sign-Multipart_Tue_Sep_15_16:18:03_1998-1
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit
-----BEGIN PGP MESSAGE-----
Version: 2.6.3i
Comment: If you don't know what this is, you can safely ignore it.
iQCVAwUBNf4Gnj9oumhqYnjxAQEnGQQAyCwtkH2LCDSvmdGOA5d7OQGK/bA2SQhq
ro1ec0z0mK8qe3IxeyJh43D3ukxSj0GpPKEEootrmWOW7LQ8ilUk4CZxJR5DKZ4o
TxPsFrJR0p1wCkoS/FM1cPaej2XyO3T/x+a5byHSRzrxb0nkFbDh44SDwFwNFld+
L8LFFFMbgX8=
=w6cv
-----END PGP MESSAGE-----
--pgp-sign-Multipart_Tue_Sep_15_16:18:03_1998-1--
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/