[William Sommerwerck]

Have you considered carting them down to the end of Lonely Street? I hear
that Heartbreak Hotel has a lot of cheap rooms where you might store them.

[Trevor]

"Adrian Tuddenham" wrote in message
nvalid.invalid...
It is far less trouble to build the right pre-amp for the job in the
first place.

You forgot, *IN YOUR OPINION*.

Sorry, I forgot to add that it was only in my opinion - but my opinion
was based on 20 years of experience as a professional 78s transfer
engineer.

And there we have it, *IF* I was doing it professionally for 20 years I'd
want the proper tools for the job too, no argument!
And as I already said, the OP who mentioned 850 discs would be better off
with them too, *including* the right turntable/cartridge/stylii for the job.

Trevor.

[Trevor]

"Adrian Tuddenham" wrote in message
alid.invalid...
You are correct that I was thinking about doing it in analogue, but the
errors involved in doing it digitally are just as bad and less easy for
the user to control. Even generating a simple 6dB/octave slope in
digital is a more difficult process than generating it in analogue, and
less likely to be error-free.

Have you ever tried it? What tools did you use?

The digital process would require two slopes to undo RIAA and one to
re-equalise, then the product of all of them would be applied to the
signal.

And it could still be more accurate than a hardware EQ.

To generate the software for this, or to try to manipulate
exising software, would not be a trivial process.

For the DSP software engineers it's probably more trivial than building a
hardware EQ to do the job. Each to their own.

The equalisation
would also be applied to any errors resulting from the A/D conversion
and it would generate artifacts of its own due to truncation errors.

With 24 bit sampling, and 32 or 64 bit FP arithmatic of 78's sound quality,
not an actual problem.


Most of the equalisation required for 78s is intended to
reverse-engineer an analogue process and, to my mind, is much more
easily done in the analogue domain before digitisation. In the example
above, six moderately accurate components would be required. If the
equalisation were done at-source (without any intervening RIAA stage)
only two moderately accurate components would be needed.

So why is it necessary to spend $2k on the pre-amp as was suggested?


There is also the point raised elsewhe it is extremely difficult to
monitor the signal for errors if it is not available at the right speed
and correctly equalised during the actual transcription process.

You monitor for errors in the DAW while you do the transfer. Latency is not
an issue.
You may be happy doing it your way, which is fine as it may be even be the
best way, but you appear to have no idea about whether alternative methods
can work for others.

Trevor.

[Trevor]

"Scott Dorsey" wrote in message
...
$2k is about ten hours of my time.

Lucky you :-)

So, if I invest $2K in a preamp, is it
going to save ten hours by straightening out my workflow?
Probably not because I'm not doing a lot of shellac work these days.

Exactly. And for others the situation may be even less likely.


And since we know that there are many different "standards" for 78's, and
many non standard, an expensive pre-amp with infinite EQ adjustment buys
you
no more benefit than a DAW with infinite EQ adjustment. And both can
usually
save those settings.

Depends on how easy it is to adjust the DAW... when you spend money on
consoles and preamps, a lot of what you're spending money on is user
interface. A lot of the time dealing with unknown shellac is spent
fiddling
around and listening to changes and trying to decide which settings are
really correct.

Exactly, whatever method you use.


Money is a work-value equivalent... I pay someone else money so that I
don't
have to do some work in order to get something I want. Whether it is
worth
it or not depends on how much work I save. I could build everything in
the
studio from component level if I had to, but I'd rather spend money and
let
someone else do that.

Of course, but everyone has different requirements, income, budget, skills,
time etc.


snip many good observations
And then acoustic recordings are a whole different kettle of fish, and
that's where it becomes entirely a judgement call to equalize anything.

I can imagine, but isn't that what sound engineers do every day even with
modern recording?

Trevor.

[Trevor]

"Scott Dorsey" wrote in message
...
Doug McDonald wrote:
I don't really understand why one would want to record
most (electrical) 78s with a flat response preamp, then fix
in software.

Because the alternative is to record it with an adjustable preamp,
and making an adjustable preamp that is repeatable and accurate is not
all that easy.

OR record it with a standard RIAA pre-amp and adjust it in the digital
domain from there. Surely easier and better for many people?


Since all eq curves are very very roughly a 6 dB per octave
low pass filter at some frequency above 20 Hz, it would
seem to me that the "base" would be just that rather than flat.
It would bet much closer to putting all frequencies equally
well in range of the ADC.

Right, there are most often one or two single-pole filters, at
different frequencies. So you need to make the frequencies adjustable,
and the filters possible to disable. You may want some notch filtering
and a high-order rumble filter too.

All simple for digital processing.


There are adjustable preamps that do this, and you can build your own
or you can pay KAB to make one for you.

I still think they are for people who want to LISTEN to their 78's, not just
transfer a few to CD.


There is also a list I found on the web that gives
settings for recording with an RIAA preamp and
a 1 octave graphic equalizer (which are, of course,
not going to give a truly accurate result.)

It's way, way far off... even worse than you'd expect. I did that once
at a radio station where I had to do some rough dubs to tape and did not
have proper tools, and I was really surprised how poor it was.


Right, why use a one octave EQ when you can do it far better in the digital
domain these days.

Trevor.

[Scott Dorsey]

In article , Trevor wrote:
"Scott Dorsey" wrote in message
And then acoustic recordings are a whole different kettle of fish, and
that's where it becomes entirely a judgement call to equalize anything.

I can imagine, but isn't that what sound engineers do every day even with
modern recording?

That's actually a good analogy, because with an acoustic recording you
often have no reference, and don't really know what it _should_ sound
like, so you just have to work on it until it sounds good to you. Just
like folks do with rock music today.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

In article , Trevor wrote:
"Scott Dorsey" wrote in message

There are adjustable preamps that do this, and you can build your own
or you can pay KAB to make one for you.

I still think they are for people who want to LISTEN to their 78's, not just
transfer a few to CD.

Well... yes.... but... really, what's the sense of having records if you
can't just sit down and listen to them?
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Adrian Tuddenham[_2_]]

Trevor wrote:


For the DSP software engineers it's probably more trivial than building a
hardware EQ to do the job. Each to their own.

That is the key to the whole "A is easier than B" argument; it depends
on which camp the operator is already in. It is much more difficult to
resolve the "A is better than B" argument because very few people are
equally expert in both camps and able to give an unbiassed opinion.

The digital process would require two slopes to undo RIAA and one to
re-equalise, then the product of all of them would be applied to the
signal.

And it could still be more accurate than a hardware EQ.

It could be, but absolute accuracy is less important than other factors
such as artifacts. Most modern analogue components are far more
accurate than their counterparts were when 78s were recorded, so a
modern analogue equaliser can easily be more accurate than the original
recording characteristic. Greater accuracy than that is of no
consequence.


Most of the equalisation required for 78s is intended to
reverse-engineer an analogue process and, to my mind, is much more
easily done in the analogue domain before digitisation. In the example
above, six moderately accurate components would be required. If the
equalisation were done at-source (without any intervening RIAA stage)
only two moderately accurate components would be needed.

So why is it necessary to spend $2k on the pre-amp as was suggested?

I have no idea, you had better ask the vendor and purchaser. It works
out a lot cheaper if you build your own, but that doesn't include the
time costs during building. Those time costs can be easily made up when
you start using the equipment - it is a relatively quick process to just
change a few components in something you designed and built yourself, so
as to get the required characteristic or make it operate in a way which
suits your particular circumstances. The time and cost of having a
commercially-built piece of equipment customised can often mean it is
more economical to turn away the work. (That is how I pick up many of
my more interesting jobs, turned away by mainstream transcriptors.)


There is also the point raised elsewhe it is extremely difficult to
monitor the signal for errors if it is not available at the right speed
and correctly equalised during the actual transcription process.

You monitor for errors in the DAW while you do the transfer. Latency is not
an issue.
You may be happy doing it your way, which is fine as it may be even be the
best way, but you appear to have no idea about whether alternative methods
can work for others.

I do have some idea, but not a lot of experience.

--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[Arny Krueger[_4_]]

"Trevor" wrote in message
...

Frankly the idea of "de-EQing" from RIAA to flat, then re-EQing to a
somewhat similar although different curve, both having fairly large gain
ranges, is guaranteed to reduce quality.

There is a large selection of preamps with flat response on the market -
they are called mic preamps. The input impedances are too low, but that is
primarily set by a resistor that can be changed.

There is also this:

http://www.rfwilmut.clara.net/repro78/circuit.html

[Scott Dorsey]

Arny Krueger wrote:
"Trevor" wrote in message
...

Frankly the idea of "de-EQing" from RIAA to flat, then re-EQing to a
somewhat similar although different curve, both having fairly large gain
ranges, is guaranteed to reduce quality.

There is a large selection of preamps with flat response on the market -
they are called mic preamps. The input impedances are too low, but that is
primarily set by a resistor that can be changed.

Actually, not. It's really not a trivial job to make a mike preamp give
you such a high-impedance load. Also, the leakage capacitance of the
mike preamp is usually too low (because there's no reason to bother making
it low when your input impedance is so low).

In fact, if I were going to put a 47K source into a mike preamp, what I
would probably do is put a pair of jfets into a box as a Schmitt splitter,
and use the phantom power on the preamp to power them.

Hmm..... you think this would make a good magazine project?
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Don Pearce[_3_]]

On Sun, 8 Apr 2012 07:23:24 -0400, "Arny Krueger"
wrote:

"Trevor" wrote in message
...

Frankly the idea of "de-EQing" from RIAA to flat, then re-EQing to a
somewhat similar although different curve, both having fairly large gain
ranges, is guaranteed to reduce quality.

There is a large selection of preamps with flat response on the market -
they are called mic preamps. The input impedances are too low, but that is
primarily set by a resistor that can be changed.

There is also this:

http://www.rfwilmut.clara.net/repro78/circuit.html


You can't just change the resistor. The front end of the pre-amp is
designed to be noise matched to a low impedance - this is done by
choosing the right ratio of voltage noise to current noise. When you
switch to a higher impedance source, the op-amp has to change too, if
noise is not to become an issue.

I realize that for 78s, this is not likely, but I'm speaking
generally.

d

[Trevor]

"Scott Dorsey" wrote in message
...
In article , Trevor
wrote:
"Scott Dorsey" wrote in message

There are adjustable preamps that do this, and you can build your own
or you can pay KAB to make one for you.

I still think they are for people who want to LISTEN to their 78's, not
just
transfer a few to CD.

Well... yes.... but... really, what's the sense of having records if you
can't just sit down and listen to them?

Isn't that the whole point of transferring them to CD, to make it easier to
*listen* to them later, without any further handling/wear to the originals?

Trevor.

[Trevor]

"Adrian Tuddenham" wrote in message
alid.invalid...
Trevor wrote:
For the DSP software engineers it's probably more trivial than building a
hardware EQ to do the job. Each to their own.

That is the key to the whole "A is easier than B" argument; it depends
on which camp the operator is already in. It is much more difficult to
resolve the "A is better than B" argument because very few people are
equally expert in both camps and able to give an unbiassed opinion.

Right, something I stated and others seem to ignore.


The digital process would require two slopes to undo RIAA and one to
re-equalise, then the product of all of them would be applied to the
signal.

And it could still be more accurate than a hardware EQ.

It could be, but absolute accuracy is less important than other factors
such as artifacts. Most modern analogue components are far more
accurate than their counterparts were when 78s were recorded, so a
modern analogue equaliser can easily be more accurate than the original
recording characteristic. Greater accuracy than that is of no
consequence.

Right, so no reason to pay $2k if you can do it in software for free then
:-)

Trevor.

[Trevor]

"Arny Krueger" wrote in message
...
"Trevor" wrote in message
...

Frankly the idea of "de-EQing" from RIAA to flat, then re-EQing to a
somewhat similar although different curve, both having fairly large gain
ranges, is guaranteed to reduce quality.

I meant over just applying a single correction, but using modern ADC's and
software there is no reason why even going via a two step process *in the
digital domain* should be a problem. Using a flat pre-amp for recording OTOH
is a problem for increased noise, overload margins etc :-(

There is a large selection of preamps with flat response on the market -
they are called mic preamps. The input impedances are too low, but that is
primarily set by a resistor that can be changed.

I'm aware of that Arny, and also aware of why I *wouldn't* use one for vinyl
record transcription, as I also said in my other posts. Better to use a
standard RIAA pre-amp and re-EQ from that in the digital domain IMO. Others
can use a mic pre-amp if they want. I just don't see the point?

Trevor.

[Trevor]

"Scott Dorsey" wrote in message
...
Arny Krueger wrote:
There is a large selection of preamps with flat response on the market -
they are called mic preamps. The input impedances are too low, but that is
primarily set by a resistor that can be changed.

Actually, not. It's really not a trivial job to make a mike preamp give
you such a high-impedance load.


Actually not all mic's or mic pre-amps are LOW impedance anyway. I still
have some 50k mics floating around somewhere, so any pre-amp designed for
them should "work". I still wouldn't use it for 78 transcription myself!

Trevor.

[PStamler]

On Apr 8, 5:43*pm, "Trevor" wrote:
Using a flat pre-amp for recording OTOH
is a problem for increased noise, overload margins etc

With careful design it's actually not that hard. Overload is pretty
straightforward; according to some tests I ran, the fastest signal
likely to show up on a disc is about 100cm/sec (that would be a BIG
scratch). With a cartridge of typical sensitivity that generates 100mV
output. Amplify that by +34dB (50x), and you get 5V out, which a good
opamp with +/- 15V supplies can easily handle cleanly. +34dB gain will
usually get you a decent signal level for recording -- electrically-
recorded 78s tended to be cut hotter than LPs were.

Noise is a little trickier, mostly because of the inductive nature of
most phono cartridges. This doesn't affect the generation of Johnson
noise, but with bipolar transistors you may run into problems with
current noise. Cartridges like Grados are lower-inductance, and an
NE5534a or NE5532a will be adequately quiet; higher-inductance
cartridges like Stantons will be noisier. I like Grados for unwarped
discs (they track well), but keep a Stanton on hand for warped ones
(they don't wobble like Grados do). They are noisier -- I can see the
meter on my DAW jump up a few dB when I plug in the Stanton. Still,
we're dealing with **78s** here, where the noise level is way higher
than any preamp noise would be, so I don't sweat it.

A flat preamp for 78s isn't that hard to do, at least in its basic
functions. Refinements make it a little harder (RFI proofing, supply
regulation, all the little stuff that makes a finished design) but no
harder than any other good circuit design.

Peace,
Paul

[Adrian Tuddenham[_2_]]

Trevor wrote:

"Adrian Tuddenham" wrote in message
alid.invalid...
Trevor wrote:
For the DSP software engineers it's probably more trivial than building a
hardware EQ to do the job. Each to their own.

That is the key to the whole "A is easier than B" argument; it depends
on which camp the operator is already in. It is much more difficult to
resolve the "A is better than B" argument because very few people are
equally expert in both camps and able to give an unbiassed opinion.

Right, something I stated and others seem to ignore.


The digital process would require two slopes to undo RIAA and one to
re-equalise, then the product of all of them would be applied to the
signal.

And it could still be more accurate than a hardware EQ.

It could be, but absolute accuracy is less important than other factors
such as artifacts. Most modern analogue components are far more
accurate than their counterparts were when 78s were recorded, so a
modern analogue equaliser can easily be more accurate than the original
recording characteristic. Greater accuracy than that is of no
consequence.

Right, so no reason to pay $2k if you can do it in software for free then
:-)

You will still need a pre-amp. If you accept that you have to have a
pre-amp, why not fit it with the correct components, instead of using
the wrong ones and then mangling the sound afterwards in software? It
actually needs fewer components to equalise a 78 (which has no top boost
in its recording characteristic) than it does to equalise RIAA.

The reason RIAA pre-amps are cheap and '78s' pre-amps are expensive is
mainly down to the difference in sales numbers and consequent production
costs. If you can pick up a reasonable secondhand RIAA pre-amp (one
that uses printed circuits, not surface mount) the cost of conversion is
trivial as long as you don't mind it finishing up with a home-made
appearance.

Assuming I am already using a computer* for something else, so the
additional cost of using it for audio is minimal - and assuming it
already is equipped with A/D converters of the necessary standard -
where do I get the free software which does objective equalisation in
real time? (By 'objective' I mean something which comes with sufficient
documentation and on-screen control that the user reliably knows what
response he is getting at all times.)

* Mac G3 running OS 8.6 with on-board A/D converters as my first choice
or an Intel iMac running OS X 4 with outboard converters at a pinch.



--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[Adrian Tuddenham[_2_]]

Trevor wrote:

... Using a flat pre-amp for recording OTOH
is a problem for increased noise, overload margins etc :-(

It should not be.

The only things which might overload a pre-amp are the clicks and pops
from record damage; they can be over 20dB above the maximum recorded
amplitude. It is perfetly possible to design a pre-amp with 40dB
overload margin and a noise floor which is comparable to the thermal
noise of the cartridge resistance. I have done this on different
occasions using discrete transistors, op-amps and, more recently,
all-valve technology.

The problems (if any) arise if the re-recording medium does not have
enough headroom. Even then, if an A/D converter overloads by just
clippping, the damage to the waveform will be minimal (but there will be
low frequency artifacts audible from the 'flat-topping'). My own
preference is to use an analogue computer to knock out the clicks and
pops before re-recording, but that is a whole new can of worms.


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[Trevor]

"Adrian Tuddenham" wrote in message
valid.invalid...
Right, so no reason to pay $2k if you can do it in software for free then
:-)

You will still need a pre-amp. If you accept that you have to have a
pre-amp, why not fit it with the correct components, instead of using
the wrong ones and then mangling the sound afterwards in software? It
actually needs fewer components to equalise a 78 (which has no top boost
in its recording characteristic) than it does to equalise RIAA.

My theory is that most people who have a turntable already have an RIAA
pre-amp. My issue is the idea that they should spend $2k on another pre-amp
when they probably won't even spend that on a turntable. My other issue is
why you'd want to buy/build a flat pre-amp instead when digitising 78's. For
those who can easily modify their RIAA pre-amp for the proper EQ's required,
that may be a better option. That's obviously not everybody however.


The reason RIAA pre-amps are cheap and '78s' pre-amps are expensive is
mainly down to the difference in sales numbers and consequent production
costs. If you can pick up a reasonable secondhand RIAA pre-amp (one
that uses printed circuits, not surface mount) the cost of conversion is
trivial as long as you don't mind it finishing up with a home-made
appearance.

Right, I know the reason why very low volume items can be expensive, and
that's why they are poor value for many people who just want to copy a few
old 78's.
I also mentioned the fact that you can modify most old RIAA pre-amps for the
job in a previous post, if you'd rather do that than re-EQ in software. Not
everyone can though, (and some can't do either I guess :-)


Assuming I am already using a computer* for something else, so the
additional cost of using it for audio is minimal - and assuming it
already is equipped with A/D converters of the necessary standard -
where do I get the free software which does objective equalisation in
real time? (By 'objective' I mean something which comes with sufficient
documentation and on-screen control that the user reliably knows what
response he is getting at all times.)

I'm sure you can always come up with a reason why you don't like what's
currently available. Your choice. Others are entitled to theirs.
I'm certainly NOT, and never have suggested everyone must use the same
methods. But I certainly can't see the point in a $2k pre-amp for most
people, nor how a flat pre-amp is a better alternative to using an RIAA
pre-amp and re-EQing in software. Modifying an RIAA one is certainly a good
option if you have a reasonable number of 78's to do, and you can do it of
course.

Buying a proper turntable, cartdidge & stylii options would still be my
first priority if I had *many* 78's to do however, which was my main point.
For those *seriously* into 78 replay I'm sure they have already decided what
works best for their needs and budget, and they don't want my opinions. I'm
fine with that! :-)

Trevor.

[Trevor]

"Adrian Tuddenham" wrote in message
valid.invalid...
Trevor wrote:

... Using a flat pre-amp for recording OTOH
is a problem for increased noise, overload margins etc :-(

It should not be.

The only things which might overload a pre-amp are the clicks and pops
from record damage; they can be over 20dB above the maximum recorded
amplitude. It is perfetly possible to design a pre-amp with 40dB
overload margin and a noise floor which is comparable to the thermal
noise of the cartridge resistance. I have done this on different
occasions using discrete transistors, op-amps and, more recently,
all-valve technology.

Of course some can, and I can, (although I doubt I could design a decent
tube amp! Fortunately I'd never want to! :-) but many people cannot. The
question is how much *they* should pay for one, and whether that's their
best option for their desired budget.


The problems (if any) arise if the re-recording medium does not have
enough headroom.

You're kidding right? With many converters capable of 110dB+ DNR, that's the
LEAST of your problems.


Even then, if an A/D converter overloads by just
clippping,

Use a pad, build your pre-amp with less gain, just get the voltage at the
right level for your converter in the first place!


the damage to the waveform will be minimal

What rubish, there is *NO* need for clipping when copying a low DNR signal
to ANY modern ADC, and whether it's minimal or not depends on the amount
obviously!


(but there will be low frequency artifacts audible from the
'flat-topping').

Maybe you need to learn how to avoid doing that then.


My own
preference is to use an analogue computer to knock out the clicks and
pops before re-recording, but that is a whole new can of worms.

Wow, everyone else realised the disadvantage of declicking in the analog
domain compared to the digital domain decades ago. You are certainly going
to affect the sound far more by doing that, but you're welcome to your
choices.

Trevor.

[Adrian Tuddenham[_2_]]

Trevor wrote:

"Adrian Tuddenham" wrote in message
valid.invalid...
Trevor wrote:

... Using a flat pre-amp for recording OTOH
is a problem for increased noise, overload margins etc :-(

It should not be.

The only things which might overload a pre-amp are the clicks and pops
from record damage; they can be over 20dB above the maximum recorded
amplitude. It is perfetly possible to design a pre-amp with 40dB
overload margin and a noise floor which is comparable to the thermal
noise of the cartridge resistance. I have done this on different
occasions using discrete transistors, op-amps and, more recently,
all-valve technology.

Of course some can, and I can, (although I doubt I could design a decent
tube amp! Fortunately I'd never want to! :-)

I was paid to do it. :-)

...but many people cannot. The
question is how much *they* should pay for one, and whether that's their
best option for their desired budget.

If someone wants to modify a pre-amp to make the response flat they
could probably pick up an old RIAA one on Freecycle and modify it with a
penknife - as long as they picked the original type correctly, it ought
to work well when modified.

If they want to pay someone to make them a flat pre-amp or buy a special
audiophile job with all the bells and whistles, then they must expect to
pay more.

If they don't want to do any of those things and don't want to learn
anything about what they are doing or how to do it, they cannot expect a
magic piece of software to sort it out it for them when all the other
tools are wrong.

By all means let's encourage people who have never experienced the
wonders of listening to musical genres of their grandparents or earlier
to get started by making intelligent use of the tools they already have;
but also let them know that with a bit of understanding and better
tools, the experience will be so much better. You have already gone a
long way down the software road and it has obviously worked for you -
but a lot can equally well be done in hardware with a limited financial
outlay and a bit of lateral thinking.

You don't need a vast budget to get the basics right, but you do need to
use your budget wisely. Spend the bulk of the money on a good basic
cartridge and a range of styli, pick up a cheap solid secondhand
turntable with a wide range of variable speed (such as a Lenco). That
applies regardless of whether you are going to follow-up with analogue
or digital processing.

The rest of the hardware can be made for almost no extra cost. If there
is a suitable existing arm, you can just go ahead and use it. If you
find the arm is fine for L.P.s but inadequate for 78s becaue it needs to
be longer, higher, more robust or able to allow the cartridge to swivel,
you can extend it or make another from some old television aerial tubing
opened out at the end and flattened with a hammer.

Good bearings can be salvaged from a scrap computer hard drive. Attach
the cartridge with an improvised swivel or even Blu-Tack. If you
cannot get proper pickup wire (it seems to have vanished off the
market), use the thinnest screened cable you can get and allow several
generous loops underneath the deck to allow it to bend. Alternatively
strip thin braiding off some cheap co-ax and make up your own screened
cable by drawing-in cores wound by twisting enamelled wire off a scrap
relay with button thread, using a hand drill. If connectors are a
problem, don't use them; solder the wires straight in (except on the
back of the cartridge, where slide connectors must always be used).

THEN....

If you want to go analogue, pick up a secondhand pre-amp and modify it
with an extra switch and some selected capacitors, then you will have a
good range of equalisation available. It doesn't matter if you have to
make a hole in the side or the back of the box to accommodate the switch
- the results are what matter.

If you want to go digital, pick up a secondhand pre-amp and modify it by
removing the RIAA capacitors, then you will have a good flat signal
ready to equalise in your computer.


Either way, don't spend lots of money on esoteric ready-made kit, just
make your own and have fun!



My own
preference is to use an analogue computer to knock out the clicks and
pops before re-recording, but that is a whole new can of worms.

Wow, everyone else realised the disadvantage of declicking in the analog
domain compared to the digital domain decades ago.

I said it was a can of worms. Please don't believe everything the
software people tell you. There are almost as many things that can go
wrong with 78s as there are 78s, so no digital program is going to
recognise them all and put them right.

I would be interested to know which free or cheap programs (or plug-ins)
make a good job of de-clicking 78s. Many seem to work well with the odd
click or two per rev of an L.P. but I have yet to find anything at low
cost which can handle the hundreds of clicks per second of a 78 without
generating dreadful and very audible artifacts. Even a professional
program like CEDAR has its problems when used alone - and it is more
expensive than an analogue de-clicker.

Both analogue and digital declicking have their place in the
professional 78s transfer world. I have had long experience of both -
and of previously of doing it the hard way with tape. Most of the
professionals I know use both systems in tandem: as a general rule the
analogue declicker comes first in the chain because it needs information
from both groove walls and a wide bandwidth pre-equalisation. After
analogue equalisation the recording is tidied-up manually with a digital
editing program and then followed by a very light pass with an automatic
digital de-hissing program if necessary.

...You are certainly going
to affect the sound far more by doing that, but you're welcome to your
choices.

For archival copies no de-clicking is allowed, regardless of the method
employed.


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[PStamler]

On Apr 9, 2:14*am, (Adrian
Tuddenham) wrote:

tt actually needs fewer components to equalise a 78 (which has no top boost
in its recording characteristic) than it does to equalise RIAA.

True of many European recordings, but many North American recordings
used top boost.

A few more points made in later posts (not by Mr. Tuddenham):

1. I don't know of any free software that's particularly useful for
working on 78s, but DC EIGHT is reasonably affordable, as is Audition.

2. It's a good rule of thumb that, all other things being equal, the
less you do to a signal, the less you damage it. Hence my preference
for starting with flat, then adding the EQ, rather than starting with
RIAA, de-equalizing, then re-equalizing.

3. It's really easy to modify a Hagerman preamp, to take an example,
for flat EQ. And it doesn't cost $2k; more like $200, plus a few bucks
for a box to put it in and an hour at the drill press. A commercial
unit with multiple switched EQ curves could easily hit $2k, minimum,
when you count the labor costs, the manufacturer's markup, the
distributor's markup, etc..

Peace,
Paul

[Scott Dorsey]

Adrian Tuddenham wrote:
The basic recording characteristic of most electrical 78s is equivalent
to just a single filter. The only ones which appear to have two bass
roll-off filters are British H.M.V.s in the very early days of the W.E.
recording system. The U.S. version of the equipment was set up with a
single 3dB point at about 500 c/s but the U.K. version used 300 c/s and
sometimes a second roll-off.

RCA Red Seals also had two poles, and most importantly a lot of transcription
discs. Transcription discs are still decent business today.

At a guess, I would think this was because the changeover from acoustic
to electrical reproduction took place much more rapidly in the U.S. -
and a 500c/s characteristic would sound satisfactory on an electrical
reproducer. In the U.K. a little more mid-bass was needed to make the
recordings sound good on acoustic machines, but the 300c/s
characteristic caused tracking and wear problems with low bass. A
second time constant was switched into the microphone circuit to remove
low bass when necessary. This was documented in the studio log sheets,
because if a recording failed the 'wear test' it would have to be
re-made with a more severe bass cut setting.

I can buy that. 78s lived on in the UK years after US companies stopped
pressing them too.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Trevor]

"Adrian Tuddenham" wrote in message
lid.invalid...
If someone wants to modify a pre-amp to make the response flat they
could probably pick up an old RIAA one on Freecycle and modify it with a
penknife - as long as they picked the original type correctly, it ought
to work well when modified.

IF they have the skills to do so.


If they want to pay someone to make them a flat pre-amp or buy a special
audiophile job with all the bells and whistles, then they must expect to
pay more.

Sure

If they don't want to do any of those things and don't want to learn
anything about what they are doing or how to do it, they cannot expect a
magic piece of software to sort it out it for them when all the other
tools are wrong.

But they can, it is a viable option for some. It's up to them to decide
which is best for their needs and budget.


By all means let's encourage people who have never experienced the
wonders of listening to musical genres of their grandparents or earlier
to get started by making intelligent use of the tools they already have;
but also let them know that with a bit of understanding and better
tools, the experience will be so much better.

And I'd encourage them to try the options before dismissing them, without
knowing "how much better" each may be. What I *wouldn't* do is encourage
them to start by buying a $2k pre-amp!!!


You have already gone a
long way down the software road and it has obviously worked for you -
but a lot can equally well be done in hardware with a limited financial
outlay and a bit of lateral thinking.

Of course. Once again if you know the options and your abilities, you can
choose what's best for your purposes and budget.


You don't need a vast budget to get the basics right, but you do need to
use your budget wisely. Spend the bulk of the money on a good basic
cartridge and a range of styli, pick up a cheap solid secondhand
turntable with a wide range of variable speed (such as a Lenco). That
applies regardless of whether you are going to follow-up with analogue
or digital processing.

Which is what I said already many times.


The rest of the hardware can be made for almost no extra cost. If there
is a suitable existing arm, you can just go ahead and use it. If you
find the arm is fine for L.P.s but inadequate for 78s becaue it needs to
be longer, higher, more robust or able to allow the cartridge to swivel,
you can extend it or make another from some old television aerial tubing
opened out at the end and flattened with a hammer.

Good bearings can be salvaged from a scrap computer hard drive. Attach
the cartridge with an improvised swivel or even Blu-Tack. If you
cannot get proper pickup wire (it seems to have vanished off the
market), use the thinnest screened cable you can get and allow several
generous loops underneath the deck to allow it to bend. Alternatively
strip thin braiding off some cheap co-ax and make up your own screened
cable by drawing-in cores wound by twisting enamelled wire off a scrap
relay with button thread, using a hand drill. If connectors are a
problem, don't use them; solder the wires straight in (except on the
back of the cartridge, where slide connectors must always be used).
THEN....
If you want to go analogue, pick up a secondhand pre-amp and modify it
with an extra switch and some selected capacitors, then you will have a
good range of equalisation available. It doesn't matter if you have to
make a hole in the side or the back of the box to accommodate the switch
- the results are what matter.

If you want to go digital, pick up a secondhand pre-amp and modify it by
removing the RIAA capacitors, then you will have a good flat signal
ready to equalise in your computer.

All fine and dandy for those with the skills, unfortunately I often deal
with people who have no idea how to hook up the cables with the gear they
buy, let alone make or modify anything.


Either way, don't spend lots of money on esoteric ready-made kit, just
make your own and have fun!

For those who can, no argument. They obviously don't need the $2k pre-amp
suggested.


My own
preference is to use an analogue computer to knock out the clicks and
pops before re-recording, but that is a whole new can of worms.

Wow, everyone else realised the disadvantage of declicking in the analog
domain compared to the digital domain decades ago.

I said it was a can of worms. Please don't believe everything the
software people tell you. There are almost as many things that can go
wrong with 78s as there are 78s, so no digital program is going to
recognise them all and put them right.

Of course not, nor any analog hardware unit!

I would be interested to know which free or cheap programs (or plug-ins)
make a good job of de-clicking 78s. Many seem to work well with the odd
click or two per rev of an L.P. but I have yet to find anything at low
cost which can handle the hundreds of clicks per second of a 78 without
generating dreadful and very audible artifacts. Even a professional
program like CEDAR has its problems when used alone - and it is more
expensive than an analogue de-clicker.

Sure, the problem is expecting anything to be automatic and good. But you do
have far more options in the digital domain, including manual editing.


Both analogue and digital declicking have their place in the
professional 78s transfer world. I have had long experience of both -
and of previously of doing it the hard way with tape. Most of the
professionals I know use both systems in tandem: as a general rule the
analogue declicker comes first in the chain because it needs information
from both groove walls and a wide bandwidth pre-equalisation. After
analogue equalisation the recording is tidied-up manually with a digital
editing program and then followed by a very light pass with an automatic
digital de-hissing program if necessary.

Yep, can of worms is right. Everyone has their own ideas of what is better.
As long as you and your customers are happy.


...You are certainly going
to affect the sound far more by doing that, but you're welcome to your
choices.

For archival copies no de-clicking is allowed, regardless of the method
employed.

As it should be for real archives. I simply save both pre and post processed
files for any transfer. Which rules out running an analog declicker during
recording anyway. (fortunately I don't want to :-)

Trevor.

[Trevor]

"PStamler" wrote in message
...
2. It's a good rule of thumb that, all other things being equal, the
less you do to a signal, the less you damage it. Hence my preference
for starting with flat, then adding the EQ, rather than starting with
RIAA, de-equalizing, then re-equalizing.

Rules of thumb are not always right. Each method may be better or worse
depending on the actual equipment used, and the operators skill.


3. It's really easy to modify a Hagerman preamp,

Not for everyone. As has been said already, it's really easy to modify many
RIAA pre-amps, for those who can.

Trevor.

[Marc Wielage[_2_]]

On Mon, 9 Apr 2012 00:14:03 -0700, Adrian Tuddenham wrote
(in article d.invalid):

You will still need a pre-amp. If you accept that you have to have a
pre-amp, why not fit it with the correct components, instead of using
the wrong ones and then mangling the sound afterwards in software? It
actually needs fewer components to equalise a 78 (which has no top boost
in its recording characteristic) than it does to equalise RIAA.
------------------------------snip------------------------------

Normally, I'd agree with you. But there is this company, Pure Vinyl, which
is pitching an all-digital RIAA compensation curve for the signal path:

http://www.channld.com/purevinyl/

Their idea is: hose the signal directly from the (unamplified) turntable and
into the computer input. The company does make a *flat* preamp just to bump
the signal up, and I think the concept could work: at least this would lift
the signal above the noise floor.

I haven't tried Pure Vinyl, but it's an interesting idea, and it's gotten
pretty good reviews from STEREOPHILE, for what it's worth.

--MFW

[Trevor]

"Marc Wielage" wrote in message
.com...
On Mon, 9 Apr 2012 00:14:03 -0700, Adrian Tuddenham wrote
(in article d.invalid):

You will still need a pre-amp. If you accept that you have to have a
pre-amp, why not fit it with the correct components, instead of using
the wrong ones and then mangling the sound afterwards in software? It
actually needs fewer components to equalise a 78 (which has no top boost
in its recording characteristic) than it does to equalise RIAA.
------------------------------snip------------------------------

Normally, I'd agree with you. But there is this company, Pure Vinyl,
which
is pitching an all-digital RIAA compensation curve for the signal path:

http://www.channld.com/purevinyl/

Their idea is: hose the signal directly from the (unamplified) turntable
and
into the computer input. The company does make a *flat* preamp just to
bump
the signal up, and I think the concept could work: at least this would
lift
the signal above the noise floor.

Exactly, they want you to buy their pre-amp. So you're back to the option,
why buy their flat one when you can buy, build or modify one that's right
for the job. Or just use the RIAA one you probably already have. If not they
can be had for no more than a flat pre-amp.
And plugging the signal straight from the cartridge into a sound card,
applying bass boost in software, and then normalising the gain up to that
required, will be pretty dire! :-(


I haven't tried Pure Vinyl, but it's an interesting idea, and it's gotten
pretty good reviews from STEREOPHILE,
***for what it's worth.***

Exactly, probably not much.
(not that I've tried it, or want to :-)

Trevor.

[Adrian Tuddenham[_2_]]

Scott Dorsey wrote:

Adrian Tuddenham wrote:
The basic recording characteristic of most electrical 78s is equivalent
to just a single filter. The only ones which appear to have two bass
roll-off filters are British H.M.V.s in the very early days of the W.E.
recording system. The U.S. version of the equipment was set up with a
single 3dB point at about 500 c/s but the U.K. version used 300 c/s and
sometimes a second roll-off.

RCA Red Seals also had two poles, and most importantly a lot of transcription
discs. Transcription discs are still decent business today.

That's interesting. We don't get many Red Seals this side of the pond,
but it would be helpful if you could let me know the turnover
frequencies (or time constants). There have been several attempts to
draw up definitive tables of turnovers for different makes at different
times, but they always seem to be incomplete, in dispute or inaccurate.



At a guess, I would think this was because the changeover from acoustic
to electrical reproduction took place much more rapidly in the U.S. -
and a 500c/s characteristic would sound satisfactory on an electrical
reproducer. In the U.K. a little more mid-bass was needed to make the
recordings sound good on acoustic machines, but the 300c/s
characteristic caused tracking and wear problems with low bass. A
second time constant was switched into the microphone circuit to remove
low bass when necessary. This was documented in the studio log sheets,
because if a recording failed the 'wear test' it would have to be
re-made with a more severe bass cut setting.

I can buy that. 78s lived on in the UK years after US companies stopped
pressing them too.

....and even longer in India. I have heard that The Beatles were issued
on 78s in India.

--scott


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[Adrian Tuddenham[_2_]]

Trevor wrote:

[...]
I also mentioned the fact that you can modify most old RIAA pre-amps for the
job in a previous post, if you'd rather do that than re-EQ in software. Not
everyone can though, (and some can't do either I guess :-)

You have just hit the nail squarely on the head. Some people cannot be
bothered to find out how to do the job properly, they have two options:

1) Buy cheap 'magic' software and allow themselves to be convinced that
that's what 78s are meant to sound like.

2) Don't bother.

They won't have much success with a decent pre-amp and equaliser because
they won't learn how to make it or use it. They won't have much success
with good quality software because they won't learn how to install it or
use it. I would be reluctant to advise them.


[...]
...But I certainly can't see the point in a $2k pre-amp for most
people,...

I am completely in agreement with you on that. For a straightforward
pre-amp I can see no good reason to spend that amount of money. A
couple of NE5532s in a box will do the basic job; add a a few more and a
good quality pot or two and you have as good an equaliser as most
professionals would ever need. It has been pointed out by others that
the cost of making and selling such a device will be far higher than
just the cost of the components; but it shouldn't be as high as $2k.

There are a number of analogue processors which are in a high price
bracket, but they do much more than just pre-amplification and
equalisation. They justify their price in the professional field, where
they are in use all day and every day, saving time and earning their
keep.


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[Adrian Tuddenham[_2_]]

Trevor wrote:

"Adrian Tuddenham" wrote in message
lid.invalid...
[...]

... Even a professional
program like CEDAR has its problems when used alone - and it is more
expensive than an analogue de-clicker.

Sure, the problem is expecting anything to be automatic and good. But you do
have far more options in the digital domain, including manual editing.

I do not object to manual editing in the digital domain, it is something
I do all the time.

The other point where digital scores over analogue is 'time shifting'.
where a waveform can be analysed over a period of time instead of
instant-by-instant. That carries with it the disadvantage that the
signal cannot be monitored in real time, which can sometimes be
important and sometimes doesn't matter.

Whilst the best quality digital processing can do almost as good a job
of equalisation as analogue, I have yet to see a program which allows
the user to 'dial up' the required time constants (or frequencies) and
slopes in real time. This is particularly important when dealing with
unknown recording characteristics, as it allows the operator to assess
the required correction by ear whilst it is changing. This is not to
say that digital cannot do it, only that it does not do it in a way
which is suitable for professional use at present.

Another feature which I have always felt would be useful and easy to do
in the digital domain is graphically-controlled speed and frequency
sweep (rather like the Oramics machine). I have had a number of
recordings* with pitch 'wows' which were very difficult to correct by
analogue means; when I tried to correct them in the digital domain I
found that the program would only apply a fixed speed change to a
selected block of samples. I need to be able to draw a graph of the
required speed change pattern and then apply it to a longer section of
the recording. Do you know of any software that will do it?


Both analogue and digital declicking have their place in the
professional 78s transfer world. I have had long experience of both -
and of previously of doing it the hard way with tape. Most of the
professionals I know use both systems in tandem: as a general rule the
analogue declicker comes first in the chain because it needs information
from both groove walls and a wide bandwidth pre-equalisation. After
analogue equalisation the recording is tidied-up manually with a digital
editing program and then followed by a very light pass with an automatic
digital de-hissing program if necessary.

Yep, can of worms is right. Everyone has their own ideas of what is better.
As long as you and your customers are happy.

A few years ago I took part in a trial organised by a major archive to
compare different digital and analogue de-clickers. I was happy that my
own design of analogue de-clicker came out top on the grounds of
performance alone; the best digital system came second on performance
but was rated easier to use by a less-skilled operator, so they were
happy too.



* Example of 'wow' on a commercial recording:

http://www.poppyrecords.co.uk/other/.../Motoring3.mp3

Columbia recorded this with a portable machine in a London theatre and I
think the 'wows' were caused by the hawser of the weight-motor being
incorrectly wound when they set it up.


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[PStamler]

On Apr 10, 4:00*am, (Adrian
Tuddenham) wrote:
...But I certainly can't see the point in a $2k pre-amp for most
people,...

I am completely in agreement with you on that. *For a straightforward
pre-amp I can see no good reason to spend that amount of money. *A
couple of NE5532s in a box will do the basic job; add a a few more and a
good quality pot or two and you have as good an equaliser as most
professionals would ever need. *It has been pointed out by others that
the cost of making and selling such a device will be far higher than
just the cost of the components; but it shouldn't be as high as $2k.

Meanwhile I've finally dug up the link to Tracertek's flat preamp:

http://www.tracertek.com/flat-phono-preamps

No idea what they use for amplification, but if it's anything like an
NE5532 or an OPA2132, the results should be fine.

Peace,
Paul

[Steve King]

"Adrian Tuddenham" wrote in message
valid.invalid...
Trevor wrote:

"Adrian Tuddenham" wrote in message
lid.invalid...
[...]

SNIP

Another feature which I have always felt would be useful and easy to do
in the digital domain is graphically-controlled speed and frequency
sweep (rather like the Oramics machine). I have had a number of
recordings* with pitch 'wows' which were very difficult to correct by
analogue means; when I tried to correct them in the digital domain I
found that the program would only apply a fixed speed change to a
selected block of samples. I need to be able to draw a graph of the
required speed change pattern and then apply it to a longer section of
the recording. Do you know of any software that will do it?


Many video editing programs have the capability to introduce various effects
over time. They do this by the use of key frames, specific points in time
in the project, which define the beginning or end of an effect. For
instance, at this point in time I want the speed to be nominal, then at this
point in time I want the speed to be 1 + X, then at this point in time I
want the speed to revert to nominal. As many key-frames as necessary can be
created to cause a specific effect. In the case of 'wow', a relatively
regular speed change, three key frames might be assigned for the beginning
of speed up, the maximum speed, and the slow down to nominal. That sequence
of speed changes can then be looped as many times as necessary to cover wow
over many speed change occurances. Could this be controlled by a
graphically represented pattern? I suppose. I'm more familiar with the
principle than the GUI in specific software. I believe that Vegas, which is
both a video editing program and a powerful audio program as well, has the
capability if not the convenient interface.

Steve King

[MarkK]

Another feature which I have always felt would be useful and easy to do
in the digital domain is graphically-controlled speed and frequency
sweep (rather like the Oramics machine).

SNIP

Do you know of any software that will do it?

A program called Capstan

I have not used it but the You Tube Ad is impressive

http://www.youtube.com/watch?v=qqK6wgsh3QA

I think the audio forensics restoration of old recordings is a fascinating
intersection of art and science......

Mark

[PStamler]

On Apr 10, 9:17*pm, "MarkK" wrote:
Another feature which I have always felt would be useful and easy to do
in the digital domain is graphically-controlled speed and frequency
sweep (rather like the Oramics machine).

SNIP

Do you know of any software that will do it?

A program called Capstan

I have not used it but the You Tube Ad is impressive

http://www.youtube.com/watch?v=qqK6wgsh3QA

Unfortunately, the price is equally impressive. Otherwise, I'd be
using it.

Peace,
Paul

[Adrian Tuddenham[_2_]]

PStamler wrote:

On Apr 10, 9:17*pm, "MarkK" wrote:
Another feature which I have always felt would be useful and easy to do
in the digital domain is graphically-controlled speed and frequency
sweep (rather like the Oramics machine).

SNIP

Do you know of any software that will do it?

A program called Capstan

I have not used it but the You Tube Ad is impressive

http://www.youtube.com/watch?v=qqK6wgsh3QA

Unfortunately, the price is equally impressive. ...

Hells Bells! That costs more than my entire kit.

It certainly looks like exactly what is needed - I must find a rich
customer who is prepared to cover the cost.


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[Luxey]

On Wednesday, 11 April 2012 09:25:51 UTC+2, Adrian Tuddenham wrote:


Hells Bells! That costs more than my entire kit.

It certainly looks like exactly what is needed - I must find a rich
customer who is prepared to cover the cost.

Tried Vari Pitch, within Cubase?

[Luxey]

It certainly looks like exactly what is needed - I must find a rich
customer who is prepared to cover the cost.

Also, in Cubase, and I guess any DAW, you can set timing to audio and program tempo changes. Oh man, I should earn money for free advice.

[Adrian Tuddenham[_2_]]

Luxey wrote:

On Wednesday, 11 April 2012 09:25:51 UTC+2, Adrian Tuddenham wrote:


Hells Bells! That costs more than my entire kit.

It certainly looks like exactly what is needed - I must find a rich
customer who is prepared to cover the cost.

Tried Vari Pitch, within Cubase?

I have tried it in Pro-Tools and Audacity, but neither of them allows a
graph to be drawn.

This is the sort of problem I need it for:

http://www.poppyrecords.co.uk/other/.../Motoring3.mp3

This is not a regular 'wow', but one cause by an intermittent fault on
the weight motor of the recording turntable.

Badly 'wowed' recordings with a regular 'wow' are easy to recover by
analogue means; when the recording is off-centre (as many of them are)
it is relatively simple to re-centre the disc on playback. Usually the
centre hole of the pressing is at fault; but I have come across
pressings where the plates slipped a long way off centre - and direct
recordings where the disc was centred on the drive pin instead of the
centre pin. (At a guess, caused by an inebriated operator.)

Even wax cylinders which have gone oval and 'wow' twice per rev can be
corrected by analogue playback:

http://christerhamp.se/phono/tuddenham.html


--
~ Adrian Tuddenham ~
(Remove the ".invalid"s and add ".co.uk" to reply)
www.poppyrecords.co.uk

[Scott Dorsey]

Adrian Tuddenham wrote:

That's interesting. We don't get many Red Seals this side of the pond,
but it would be helpful if you could let me know the turnover
frequencies (or time constants). There have been several attempts to
draw up definitive tables of turnovers for different makes at different
times, but they always seem to be incomplete, in dispute or inaccurate.

It's a two pole filter. F1 is specified at 500 Hz, F2 is 1590 Hz and
set for 13dB at 10 Khz. Maybe.

Even wackier is the Columbia stuff! Columbia used a three-pole filter
like the modern RIAA curve on some of their later 78 and early LP pressings.
F1 at 100 Hz, F2 and 500 Hz, F3 at 1590 Hz, and something like 16dB boost
at 10 Khz.

None of these numbers are guaranteed, and the might be totally wrong. But
they aren't any worse than the ones in the McIntosh preamp book.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[William Sommerwerck]

Even wackier is the Columbia stuff! Columbia used a three-pole filter
like the modern RIAA curve on some of their later 78 and early LP
pressings. F1 at 100 Hz, F2 and 500 Hz, F3 at 1590 Hz, and
something like 16dB boost at 10 Khz.

Picky point... For recording, F1 is a zero. F2 is a pole. F3 is a zero.

Unless you're trying to produce a constant-amplitude recording across the
entire audible spectrum (which is generally undesirable, for a number of
reasons), you'll always need a zero/pole/zero recording EQ, with frequencies
similar to those given. In this regard, all pre-emphasis is pretty much the
same.