On 3/28/2015 6:41 AM, Tom Fine wrote:
> As for flat transfers, I recommend reading Gary Galo's research on this.
> I have yet to hear RIAA software that sounds half as good as a
> high-quality phono preamp with high-fidelity material. I haven't messed
> with lower-fidelity 78s. Gary's research indicates digital EQ do not
> take into account certain facts and factors of analog components, and
> thus do not mirror the cutting curve as well as a properly-designed and
> well-built analog preamplifier. Horses for course, but I'll always
> choose analog for phono EQ and preamplification. Paul's point about
> trying multiple curves is correct, but one can figure out a
> good-sounding curve without over-playing a disk.
I read Gary's comments on software implementation of the curves, and I'm
sorry to say that, for at least some current software, he's wrong. His
basic contention is that software-implemented curves are "linear phase",
and so have different phase characteristics from hardware-implemented
curves (which are "minimum phase". I did some experiments, and verified
that at least the software-implemented curves in Adobe Audition and DC
EIGHT are in fact minimum phase, and so have the same phase
characteristics as standard hardware-implemented curves. I published
these findings in a letter to audioXpress several years ago. Gary's
criticisms of software-implemented curves may have been correct some
years ago, but (at least for the software I tried) are no longer.
There are still potential advantages to hardware compensation, so that's
what I use when I'm confident a disc is RIAA. If I'm not confident of
that, I transfer flat and do the compensation in software.
Incidentally, you cam make a preamp for flat phono transfers from a
couple of NE5532 opamp chips, some resistors, some batteries and RCA
jacks, and a box to put it all in.
This email has been checked for viruses by Avast antivirus software.