Re: [ARSCLIST] Space Shuttle Recorder - Link to NY Times article

[Mike Richter <mrichter@xxxxxxx>]

At 06:06 PM 3/24/2003 -0800, Rod Stephens wrote:

> The overall problem is a common one, I would guess, since many of the
> machines used for archiving today are not in their first youths, and
> wear of the various transport parts will cause changes in speed.  I
> thought this should give others food for thought in checking out their
> analog decks, and I hope my experience will be of help to others.
>
> The irony here is that I'm sure the vendor from whom I purchased the
> Scully wasn't aware of any problem, since he was doing "in house"
> recordings, and, as long as the recordings were played back on the same
> deck, they would be relatively "perfect", playing at the same speed they
> were recorded.
>
> I have gone back and remastered the offending recordings.  I have also
> in some cases used my digital software to resample the .wav files using
> the "time/pitch" tools in Cool Edit Pro, since I have now found that
> some of the library's 1/4" tape masters were not recorded perfectly on
> pitch, either, due to the machines they were using back in the '50's and
> '60's.
>
> So, the moral of the story is: "Nothing Is Perfect".

Not knowing the design of the deck, I cannot guess what part of speed
regulation could have gone out to cause the problem. I want to note that
correcting pitch in software is a compromise that I prefer to avoid. There
is a simple alternative if you know the speed error when you begin and if
the error is constant over the reel. In that case, simply adjust the sample
rate.

On very rare occasion, I have a 78-rpm disc to digitize. To save space, I
long ago gave up my table and preamp for those, so if I know that the
recording is at 78 rpm, I simply record it with a suitable cartridge at 45
rpm and sample at 25442 sps - (45 x 44100) / 78. I use CDWAV for that
recording; most other programs restrict sample rates. I then change the
extension on the file so I can open it in CoolEdit as though recorded at
44100, clip off the header and correct the EQ.That is actually much simpler
than it sounds and avoids the delay and the imperfection of altering pitch
in the audio editor.


Mike
mrichter@xxxxxxx
http://www.mrichter.com/