You're welcome, Ben. Taking Paul's math, which looks correct, the
adjustment you are looking for will be to raise the pitch 2.2% (.022). To
raise it a half step, that would be 5.6% (.056). So the adjustment you
want to make is a little less than a quarter tone, but you definitely want
to make it, as a 2.2% change in pitch and tempo (and yes, you want to keep
them hooked together, as Richard said) can be very noticeable in most
music, particularly if there is a vocal. The human voice really wants to
be reproduced right on pitch, and small changes one way or the other can
make a voice sound shrill or thick. But the tone of many instruments
suffers as well. A quarter tone is a lot in this context.
I don't know how reliable the speed is for Edisons, but for most other 78's
you cannot count on the speed being exactly 78.26, the nominal standard.
In fact commercial 78's ranged all over the place. That is why I said it
would be best to pitch each record individually.
There is an easy way to do this, which I use almost every day. I recently
bought a Yamaha electronic keyboard (only $100 on sale), and I play on it
all the time along with the music to compare the pitch. Even just a few
chords will tell you instantly where the pitch needs to go to be right. By
my Korg tuner, which I also use, the Yamaha keyboard is exactly
top-dead-center on pitch (unlike my Steinway piano). Getting the pitch
right becomes a snap.
When it comes to restoration work, getting the pitch right might be the
very most important thing. I am always astonished how many transfers of
both tapes and records, even commercial ones, are done off pitch, sometimes
wildly off pitch. It is just as critical an issue when working from tape
sources, as it can seem like no two tape recorders in the old days ever
played at exactly the same speed.
Sometimes, where large changes are necessary, you need to get it close in
the analog playback domain, and not just on the computer, because the
operation of phono EQ or tape EQ is impacted. Those "built in" EQ systems
are expecting to "see" recordings on pitch and will make the wrong
adjustments for a recording that is way off pitch. This will be especially
true for the bass turnover point for records.
Best,
John Haley
On Mon, Apr 27, 2015 at 10:27 PM, Paul Stamler <[log in to unmask]> wrote:
> On 4/27/2015 6:14 PM, Richard L. Hess wrote:
>
>>
>> Of course 96 is also good for accurately reproducing the scratches to
>> make removal easier.
>>
>
> Unless you're using the descratching algorithm in DC-EIGHT or its
> predecessors, which for some unaccountable reason is optimized for 44.1kHz
> to the extent that it basically doesn't work at higher sample rates.
>
> I usually work at 44.1kHz, so I don't have a problem with this, but I
> really don't like the limitation.
>
> And yes, it's vital to work at 24-bit resolution; speed changes and EQ
> work a helluva lot better on 24-bit files, and so (in my experience) does
> descratching.
>
>
> Peace,
> Paul
>
> ---
> This email has been checked for viruses by Avast antivirus software.
> http://www.avast.com
>
|