The following are my formulae for various speeds. I use Dartpro 24 and the
speed adjustments I use are done through 'adjusting sampling rate'. For
later recordings, I would record at 44,100 and resample to 22,050 for
processing - the parameters of the program limit the top 'adjust sample
rate' in such a way as I can't adjust it high enough - i.e. beyond the range
of the program to correct the speed. Initial higher sample rate deals better
with transients and outlines the waveform with greater detail-just like in
Photoshop with pictures
As I am in the paractice of saving all stages of processing, I can always
take a step backward if i'm not satisfied with the results - even years
later. I have literally thousands of directories in my working sound folder
of as yet unfinished projects, using original serial number. I can search if
I don't remember whether I've recorded in the vocalist directory or in the
band's name. They are always recorded - both sides in one file so i don't
omit a side. Once I have listened and finished editing, I save under the
title's name in both the vocalist's directory and the band's name. Thus, it
is easy to assemble a compilation quickly.
Verticals do not seem to benefit from slow transfer - using the Pickering
M190 tone arm that I use as there is not enough mass in the cartridge
assembly to maintain a proper 'zero' and noise ratio is compromized
33 / 22050 = 75 / 49500 (early electrical Columbia-some Grey Gulls) 1925 -
ca. sometime in 1928 - check out 'Say Yes Today' by Ruth Etting and you'll
see what I mean
33 / 22050 = 76.6/50633 many acoustical and electrical Victors, brunswick
acoustics and some electrics
33 / 22050 = 78 / 51598 anything after 1928
33 / 22050 = 80 / 52880 acoustical American Columbias and electrical
British, Edisons etc.
Record at 78 and adjust to 75 rpm
78 / 22050 = 75 / 21201
Record at 78 and adjust to 76.6
78 / 22050 = 76.6 / 21654 78/44,100 = 43308
Edison / columbia Acoustic
78 / 22050 = 80 / 22615
A worn record will have inferior noise/signal ratio recorded slow, but
clarity and depth of sound will still be enhanced. It can be a compromise -
you'll have to check either way for a worn record.
One thing about recomended speed adjustments for some records - in the
crank-up days, sometimes you would let the turntable 'free-wheel' to adjust
the speed, but the drag of the sound box on the record would slow the speed
Follow me on Twitter
M.C.Productions Vintage Recordings
710 Westminster Ave. West
----- Original Message -----
From: "Tom Fine" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Saturday, December 21, 2013 9:49 AM
Subject: Re: [ARSCLIST] Record Equalization
> Hi Mickey:
> Do you know off-hand what is the pitch shift to get a 33 1/3 transfer to
> 78 pitch? Soundforge has a pitch-shift tool, so I could definitely try
> What you're saying makes some sense but if you're doing for-release
> transfers, why not own a various-curves preamp like the TDL Restoration
> Preamp or something more fancy? Or a mastering equalizer?
> I start with the TDL, then usually feed the output to an equalizer to nip
> and tuck here and there. I've found that if you nip and tuck the midrange
> to get the most natural tonal balance, the bass and treble will fall into
> line. Some discs need treble taming to get rid of noise, and if you use a
> parametric you can usually tune up above any significant musical
> information. I've heard claims of high treble content on 78s, but I have
> never seen it on a spectrum graph. The mics of the day weren't capable of
> gathering information at the frequencies claimed, and horn systems weren't
> sensitive enough to pick up any high-treble information.
> -- Tom Fine
> ----- Original Message -----
> From: "Mickey Clark" <[log in to unmask]>
> To: <[log in to unmask]>
> Sent: Saturday, December 21, 2013 12:18 PM
> Subject: Re: [ARSCLIST] Record Equalization
>>I get around the RIAA curve by recording slow. High frequencies on a 78
>>when played at 33 1/3 will be presented to the RIAA circuitry at just
>>under a third of their original frequency. i.e. 10,000 cycles at 78 will
>>be just over 3,000 cycles at 33 1/3 which means less suppression of the
>>high frequencies at playback.
>> The result is a fabulous kick of bass and depth , and clarity and
>> transparency for the high frequencies. I have digitized thousands or
>> recordings this way and am satisfied that it works for most recordings -
>> but not all. e.g. Columbias - notably 'Gentlemen Prefer Blondes' is one
>> of these exceptions.
>> Once you play a 78 through an RIAA preamp at full speed, you kill much of
>> the high frequency information and you cannot equalize it to get it
>> back-Anyone interested in a sample to illustrate this, please advise me
>> and I can send an example. I have covered everything from symphonic,
>> opera, blues, jazz and spoken word and have samples I can send as mp3's.
>> The link below is a slow transfer of the Gigli/Caniglia version of
>> Verdi's Requiem. I did adjust the bass eq slightly to smooth it out, as
>> there was a bit of a hump in the curve before eq. the trble is kept
>> intact - all the best to the group and a Merry Christmas to All-Mickey
>> ----- Original Message -----
>> From: "Doug Pomeroy" <[log in to unmask]>
>> To: <[log in to unmask]>
>> Sent: Friday, December 20, 2013 1:52 PM
>> Subject: Re: [ARSCLIST] Record Equalization
>>> The Radiotron Handbook entry shows only one curve "used by R.C.A. Victor
>>> for 78 r.p.m. shellac discs, 45 r.p.m. and 33 1/3 r.p.m. fine groove"
>>> discs. As there was no RCA Victor before 1929, this does not tell the
>>> whole Victor story and does not answer the specific question Steve
>>> Smolian asks (about electrical Victors made in the short period before
>>> November of 1925).
>>> Maxfield and Harrison's 1926 paper describing the Western Electric
>>> system clearly indicates a bass turnover of 200 Hz and pre-emphasis
>>> ("constant acceleration") above "approximately 4000 Hz". This was before
>>> electrical record players were widely available and this curve was
>>> developed in consideration of playback by acoustic record players.
>>> As has been mentioned, Victor raised the bass turnover over the years
>>> from 200 Hz to 300 Hz and finally to 500 Hz. Nick Bergh knows the
>>> cutting equipment used and speaks with authority on the subject of EQ.
>>> I showed him a Victor ledger sheet from December 1926 which includes a
>>> column marked "Eqlzr." and it only shows the words "on" or "off" which
>>> doesn't tell us what we'd really like to know! It's safe to say that at
>>> Victor, experiments with various cutting equalizations were being done
>>> on an ongoing basis.
>>> Doug Pomeroy
>>> Audio Restoration and Mastering Services
>>> 193 Baltic St
>>> Brooklyn, NY 11201-6173
>>> (718) 855-2650
>>> [log in to unmask]
>>>> Date: Thu, 19 Dec 2013 18:36:58 +0000
>>>> From: "Gray, Mike" <[log in to unmask]>
>>>> Subject: Re: Record equalization
>>>> From Radiotron Designer's Handbook, 4th Ed. 1953, p. 728, item 17.5:
>>>> "There is no 'optimum' cross-over frequency because the choice is
>>>> necessarily a compromise. Where distortion is the principal criterion,
>>>> a low cros-over frequency from 250 to 350 c/s will be adopted for
>>>> standard groove 78 r.p.m. Where needle scratch is troublesome with 78
>>>> r.p.m. a high cross-over frequency of say 500 c/s may be adopted."
>>>> On page 730, under Practical recording characteristics:
>>>> "There does not appear to be any generally accepted definition of
>>>> published recording characteristics."
>>>> In the two figures on this page, European characteristics show several
>>>> cross-over points, ranging from 300 Hz (Decca ffrr)
>>>> to 600 Hz for EMI 78 rpm. Note that there is no pre-emphasis at all for
>>>> the EMI discs, which are 'flat' up to 10 Khz.
>>>> American 78s, however, show a boost of ca. 12 db at 10 KHz ...
>>>> Mike Gray
>> This email is free from viruses and malware because avast! Antivirus
>> protection is active.
This email is free from viruses and malware because avast! Antivirus protection is active.