On 3/29/2015 1:26 AM, Corey Bailey wrote:
> I would like to respectfully disagree (Agree to disagree?).

Well, that's how things should be! Let a hundred flowers bloom.

> First of all, I have, and use, the software you mentioned. I've tried
> the comparisons, not only with Audition 3 and DC-8, but Sound Forge 9,
> Pro Tools and Sonic Solutions (haven't tried after-the-fact phono EQ
> with Cedar or Pyramix) and I agree with those who are on the side of
> hardware EQ in the analog domain. Audiophiles have been arguing for
> years about the virtues of one phono preamp over another. The
> differences that they are really hearing can be defined as the "time
> constants" of a given design. The differences in time constants are
> simply the result of the type of parts used in a particular design and
> how they are arranged, regardless of weather we are discussing solid
> state or vacuum tube circuitry. And, as you know, the debate between
> "toobs" and solid state circuit designs rages on.

Not relevant; whether or not the differences in hardware RIAA preamps 
are due to different time constants, there's no excuse for making an 
RIAA preamp with the wrong time constants these days. The formulas are 
known. (Not simple, but known.)

  EQ in the digital
> domain does not make allowances for part tolerances or varying circuit
> designs.

Huh? It doesn't need to. A flat transfer is (or should be) a flat 
transfer, and parts tolerances will only affect channel balance (easily 

I find that when digital EQ is applied to a flat record
> transfer, the result is somewhat lifeless sounding although much more
> precise, I'm sure. I have to agree with Gary Galo that making a "flat"
> transfer does not allow for the headroom needed for the turnover
> frequencies unless you are willing to make your reference level around
> -25DBFS.

Actually I've done and published measurements and calculations showing 
that standard opamps, used with +/-5V supplies, have enough headroom to 
handle the output of standard-sensitivity cartridges; the big barrier is 
large scratches.

  Not adding the roll-off can make sense if a considerable amount
> of digital processing is needed to reduce noise or remove scratches,
> etc. This is where accessing an external analog EQ is useful for post
> processing (I use a GML8200 for this) although it has to be done in real
> time. If digital processing is going to be required, using a higher
> sample rate and bit depth is also beneficial.

I always use 24-bit recording. The descratching programs seem to work 
better at 24 bits. I do record at lowish levels, sometimes around 
-25dBFS. sometimes lower, sometimes higher. Since I use a fixed-level 
card (CardDeluxe) and no level control in the preamp, the level in the 
computer depends on how hot the disc was cut. But with 24-bit recording, 
there's no real penalty for recording at max=-25dBFS.

> As one who works on both the outside and inside of analog mixing
> consoles (have for years) and, although it wasn't your main point, I can
> tell you that the electronic design world has moved on from the NE5532.
> The disclaimer here is that I have never been a fan of multiple op-amps
> in one package and the NE5532 is the one I would use to make my argument.

Actually, for serious designs, I too prefer single-opamp packages; the 
problem isn't so much leakage from one amplifier to the other inside the 
package, but simultaneous loads on a single pair of power supply pins 
and closely-spaced PC traces, leading to crosstalk. But for a simple 
"bivouac" flat-EQ preamp, a 5532 will do a more-than-adequate job. Even 
powered by batteries, definitely non-ideal but, again, workable.


This email has been checked for viruses by Avast antivirus software.