Cyberserf wrote:
I've had a bunch (850ish) 78rpm disc dropped off today, and I'd like to
digitize them. I've had good success with 33 and 45rmp record and various
post processing (dehiss, declick, depop, compress, mash and torture), but my
only player that can handle 78s is...well...old and does not have any
outputs from the stylus (really a needle...seriously)except those to the
cone. My question is pretty straightforward: Has anyone had luck recording a
78 at 45 rpm digitally and speeding it up using a software solution, and if
so, what was the software used? I figure if someone can make Britney sound
ok, this must be feasible and, regardless, I'm gonna try, so if someone can
save me the experimentation, I'd be grateful.
It depends what sort of result you consider acceptable. If you just
want to hear what kind of music is on the discs, you will find lots of
software that claims to change speed. If you want to do the job
properly there are several things you need to know:
1) Speed:
Not all "78s" run at 78 rpm, there were many other speeds both
intentional and unintentional. The most popular alternative speed was
80 rpm which was standardised by the Columbia Graphophone Co. on all its
labels until about 1929. Pathés can run up to over 100 rpm.
2) Equalisation:
The only thing you can guarantee about equalisation is that RIAA will be
wrong for any 78. There were lots of standards which different
companies followed at different times and you need a wide range of
settings to cope with them. Also bear in mind that if you equalise
before pitch changing, your equalisation setting will need to change in
proportion to the pitch change.
3) Stylus:
There were many different groove profiles and you will need a range of
styli to cope with them. Often a groove which is badly worn at one
depth can be tracked successfully with a slightly larger or smaller
stylus which rides higher or lower and avoids the damaged areas.
4) "Azimuth":
The recording engineers sometimes rotated the cutter to direct the swarf
towards the rim or the centre of the disc. This means that the
waveforms on the two groove walls will be out of step. A conical
playback stylus is not affected by this, but a truncated elliptical
stylus will have to be swivelled to get the best results. Parallel
tracking is essential to keep the azimuth correct across the whole
playing surface, unless you are prepared to keep stopping to tweek the
angle of a radial-tracking arm.
5) Archival practice:
It is good archival practice to undo the various recording effects in
the reverse order from that in which they originally occurred. e.g. the
distortion from a badly-fitting stylus should be tackled during
playback, as there is no satisfactory way of removing it further down
the chain. In general, it is assumed that digital processes will
generate artifacts which cannot be later identified and removed, it is
therefore not acceptable to process the signal in the digital domain.
This means that your speed, stylus size, stylus pressure, azimuth and
equalisation must all be right before the analogue signal is digitised.
This is rigidly applied to the Archive copy unless there is no
alternative; other practices are sometimes accepted for the Playback
copy. This is a contentious area and I am sure someone will vehemently
tell me I am wrong, but those are the rules I have to work to when I
produce transfers for various archives.
As I said at the start, it all depends on what sort of sound quality you
consider acceptable and how much trouble you are prepared to go to in
order to get it. The azimuth adjustment and archival practices will only
be needed for top quality work, but the rest is fairly fundamental.
If you want to find out more about the subject, have a look at:
http://www.bl.uk/reshelp/findhelpres...loguesoundrest
oration.pdf
--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk