when it comes to roll-offs on a graphic equalizer my highest frequency is 16khz, i know there is nothing there but noise on older 78s so i cut it the max -15 db.
Next is 12.5khz, again nothing but noise so that gets cut max, followed by the 10khz...
8khz could contain some modulation but not likely enough to keep the surface noise along with it so i roll it off 12db.
Most real musical content becomes apparent with the next band 6.3khz. at this frequency i will roll off to -9 db and gradual slope till im flat at 1khz..... this is where your ear plays more of a roll than any technical graph.
i may stay flat out to 3-4 khz depending on the content and a sharper roll off depending on the noise level. there are times when i might even do a slight boost between 800hz and 2.5khz knowing ill loose of it if im going to do any digital noise reduction if this enhances the final musical content.
Something as small as 3 db change can make a lot of difference.
--- On Mon, 3/7/11, Jan Myren <[log in to unmask]> wrote:
> From: Jan Myren <[log in to unmask]>
> Subject: [ARSCLIST] SV: [ARSCLIST] Turnover and rolloff curves for correct playback of 78 rpm records!
> To: [log in to unmask]
> Date: Monday, March 7, 2011, 1:08 PM
> Does this mean that most of you about never use the FLAT
> position on the
> May it be that you use -5 instead just to "soften" the
> noise a little bit;
> then do the rest by reducing the very top a little bit
> using a parametric or
> graphic EQ?
> All the best
> -----Opprinnelig melding-----
> Fra: Association for Recorded Sound Discussion List
> [mailto:[log in to unmask]]
> På vegne av Robert Cham
> Sendt: 6. mars 2011 19:29
> Til: [log in to unmask]
> Emne: Re: [ARSCLIST] Turnover and rolloff curves for
> correct playback of 78
> rpm records!
> I used to use a parametric equalizer for that. Any of
> them would do
> fine, but this one happened to be in a McCurdy
> console. The
> technique was the same as yours.
> The advantage of a good parametric equalizer is that you
> can vary the
> "Q" which is in effect the width of the notch. Too
> narrow of a
> notch, which graphic equalizers frequently are can causes
> you can sort of accomplish the same thing with a graphic by
> the curve more gradual using the adjacent controls.
> The Mc Curdys were great equalizers fo 78s, especially
> combined with
> the preamp controls on the old EMT turntables that we used
> to use.
> Bob Cham
> >It's not just the top and bottom, it also is the
> middle. All
> >acoustical recordings have horn resonances, and if you
> use a
> >parametric equalizer you boost the midband and sweep it
> up and down
> >till you really hear a BIG boost -- and then you cut it
> a at that
> >frequency a few dBs below the rest to even out that
> resonance. With
> >a graphic eq you can smooth out several
> resonances. Tom Owen was
> >the first to really address that issue when he
> introduced the Owl I.
> >The Urei Multifilter is also useful. You will
> also find resonant
> >peaks in electrical recordings. I am surprised
> that this has not
> >been discussed much here.
> >Mike Biel [log in to unmask]
> >On 3/6/2011 7:04 AM, Tom Fine wrote:
> >>One man's opinions/experiences ...
> >>You need to use your ears. These are guidelines,
> not rules. I've
> >>never liked the sound of anything with flat
> rolloff, it's always
> >>too noisy. The early stuff has very limited
> high-frequency content
> >>so you can rolloff quite a bit without effecting
> any recorded sound
> >>that made it into the shellac pressing you are
> playing. It gets
> >>more interesting in the electronic era because
> there comes a point
> >>-- different times for different record companies,
> when there
> >>actually is important audible content in the
> "treble" range and
> >>it's loudness is effected by both turnover and
> rolloff since you
> >>are both moving and bending the curve. Start with
> the "common
> >>wisdom" and then season to taste. There are many,
> many listings
> >>online of different curves recommended for
> different record
> >>companies in different eras. I have found the
> "common wisdom" on
> >>RCA, USA Decca, Capitol and Mercury to be my
> preferred settings in
> >>most cases. The "common wisdom" on pre-electric
> anything, Columbia,
> >>Okeh, Vocalion and British HMV does not suit my
> ears, I usually
> >>tweak those to sound best to the content (most
> content, least
> >>noise, to over-simplify it). Almost uniformly, my
> experience is
> >>that the later-era the 78, the easier it is to pull
> the content out
> >>of the grooves. Early stuff is a challenge and
> there's only so much
> >>content there.
> >>Also note that not all recordings were made to
> allegedly "standard"
> >>curves, not all recordings sound good to begin with
> and not all
> >>pressings are uniformly good. So, use the options
> of variable EQ to
> >>bring out maximum audibility and clarity.
> Basically, you'll find a
> >>pair of settings that sounds "best" for every
> record. Sometimes it
> >>sounds clearly "best" compared to all other
> settings (instruments
> >>sound natural, balance of the ensemble sounds
> right, background
> >>noise does not distract), sometimes it sounds
> "least bad" because
> >>the recording was bad or the pressing is bad, etc.
> If your preamp
> >>has high-pass and low-pass filters, also experiment
> with them. You
> >>can further bring out the program content and
> reduce the background
> >>noise sometimes.
> >>I also try to go for the best sound quality at each
> stage of analog
> >>playback so you can then use a minimal number of
> stages. If I can
> >>get a "best" sound right at the phono preamp, then
> I don't need to
> >>insert EQ into the signal chain. In the case of
> 78's, if I can play
> >>it back crisp, clear and relatively quiet, then I
> don't need to
> >>mess with DSP noise reduction or EQ in the
> computer. Also, using
> >>the combination of turnover and rolloff to both
> bend and move the
> >>curve sometimes produces much better-sounding
> results than blind
> >>adherence to "common wisdom" with after-the-fact
> reliance on
> >>heavy-handed EQ to "quiet things down." This method
> is widely used
> >>on low-cost reissues of 78's, they are usually
> chopped off
> >>somewhere around 10k and "NR'd" to death so all the
> >>content is removed and it's like listening to a
> low-grade AM
> >>Always rely on your own ears and taste rather than
> >>Documentation and "common wisdom" is the place to
> start, not the
> >>place to end.
> >>-- Tom Fine
> >>----- Original Message ----- From: "Jan Myren"
> <[log in to unmask]>
> >>To: <[log in to unmask]>
> >>Sent: Sunday, March 06, 2011 5:33 AM
> >>Subject: [ARSCLIST] Turnover and rolloff curves for
> >>playback of 78 rpm records!
> >>>I am about to learn to find the general correct
> turnover and rolloff
> >>>for US and European 78 rpm records.
> >>>So far I have learned:
> >>>US 78 rpm records:
> >>>Pre. 1938: turnover 500 (RIIA), rolloff FLAT
> >>>After 1938: turnover 500 (RIIA), rolloff -5
> >>>European 78 rpm records:
> >>>Pre 1938: turnover 300, rolloff FLAT
> >>>After 1938, turnover 300, rolloff -5
> >>>Can anybody share their experiences within this
> matter, please?
> >>>All the best
> >>>Jan Myren
> Bob Cham
> KTRU FM