I think you're dancing around a central issue -- archival transfer/preservation vs.
pleasure-listening and commercial transfer/remastering.
There's a very good argument to be made for raw transfers, high resolution ones at that. By raw, I
mean sans analog EQ and sans analog NR. See my posts a while back about my visions for a tape head
to DSD stream system, cutting tape electronics completely out of the process. The same argument
applies to a properly loaded phono cartridge. In other words -- electromagnetic inductance through
as straight a line as possible to bits and bytes. I outlined the possibilities from there in my
previous posts, so I won't repeat them here.
But as for the pleasure-listening and commercial transfer/remastering process, I think aesthetics
matter more than perfect adherence to a curve designed for a perfect world. In the real world, the
"encode" curve was not always followed well and thus tweaking is needed. Plus, in the commercial
world, personal preferences and aesthetics are what keep mastering engineers working -- and some of
the worst-sounding LPs and later CDs were mastered by people who either operated like robots and
made no aesthetic decisions or made bad aesthetic decisions due to lack of knowledge, inexperience
or simply bad taste.
Now, what I was envisioning in my previous posts was a system for TAPE where both needs could be
satisfied with one pass of the tape. If the head were aligned properly, and the transfer was very
high resolution through a sonically-neutral or invisible chain (which may not exist to some ears),
then the DSD stream should be able to be played back through any impedence-matched tape preamp
and/or NR unit to get the same sound as analog playback from that same head. The same could happen
with disk transfers, assuming the cartridge was properly loaded and thus behaved properly. So in the
archives would be this unprocessed stream that is akin to the reel of tape or the record on the
shelf. What is done with it is the option of the ultimate end-use of the audio. DSP will undoubtedly
end up the way to go when automated processes, cost-effectiveness and quantity are the key
operators. When a pleasure-listening is involved or a commercial product sold on the aesthetic of
the producer or remastering person, then playback through an analog chain will likely be preferable.
Just to be clear, I am not advocating any one-size solution to any of this. We all have unique ways
and in the real world many things that don't work on paper work just great for the ears and the
soul. Conversely, in the real world, many things that work on paper and should work on the
soundwaves are highly discordant.
Finally, as I said in my original posts, I don't have a dog in this fight. I know what works for me
and it may or may not work for everyone.
-- Tom Fine
----- Original Message -----
From: "Richard L. Hess" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Monday, August 25, 2008 12:14 PM
Subject: Re: [ARSCLIST] RIAA EQ software
> Hi, Folks,
> I think perhaps this discussion could become more valuable if we think about some of the different
> classes where we might need to apply this.
> Since I don't DO discs, I would like to start with an analogy from the tape world.
> No, Tom, I don't connect my A-D directly to the head and do NAB/CCIR/IEC equalization in the
> software. For one, you need an integrator as well as the equalizer as the voltage output rises 6
> dB per octave off the tape head.
> BUT, I do record the machine outputs and the Dolby/dbx/Telcom outputs as TWO stereo pairs when I'm
> doing music work. I may not deliver the non-decoded file, but I may. In this way, if there is ever
> any question about the processing, we can do the analog re-processing without having to replay the
> analog tape.
> I also use the DSP to correct for using instrumentation recorders for playing back audio, so in
> that sense I do add the NAB curve (or some approximation that better matches the actual sound on
> the tape since one might suspect that the recorder wasn't very accurate to begin with). As you
> know, instrumentation recorders are essentially constant current record which means that there is
> some playback equalization for the thickness loss and other losses in the tape path. The only
> pre-equalization is typically to adjust for high-frequency losses in the 300 kHz - 2 MHz range,
> depending on tape machine and that's only adjusted at the highest speed (60-120 in/s).
> Conversely, I attempt to dial out all audio EQ when I'm reproducing instrumentation tapes on audio
> recorders--but that probably won't happen again now that my instrumentation recorder population
> has increased beyond all reasonableness <sigh/smile>.
> There is another use I make of DSP and that's correcting odd off-speed replay when I need to, but,
> for my common off-speed replay (like 3.75 -50% varispeed to obtain 1.88 in/s), I have a separate
> preset that has been equalized to the 1.88 in/s MRL tape on the machines.
> Now back to discs.
> I would think that there is a time and a place to use both systems. I would guess that there might
> be a good reason to use hardware RIAA EQ (and by the way, the National Association of
> Broadcaster's publication of that curve is now available as are the tape standards at
> www.richardhess.com/tape/history/ ) for playback of "late model" discs and use software for the
> wide variety of other EQs.
> There is a movement afoot (partially pushed by TracerTek -- the people who sell DC7 and flat
> preamps) that flat transfers with digital EQ is superior. I see Chas. Lawson's similar post, but
> I'm surprised analog components drift as much and as quickly as he said, but I don't doubt it
> I think that dividing the discs into different groups and then discussing each group might be more
> informative than lumping all discs together.
> The original question was for RIAA but there are few RIAA discs that we should be transferring,
> even in Canada, as the true RIAA standard came into effect what, 55 years ago? I don't think it
> was adopted immediately, was it? I recall seeing London/Decca FFRR recordings (but maybe they were
> done with the RIAA curve and they just used the acronym for marketing) into the 1960s and perhaps
> the 1970s.
> The freeware from Australia that I pointed Scott to last night does "delta" equalization, so it
> appears that you can tell it that you used an RIAA preamp to make the transfer and then it changes
> RIAA to FFRR or whatever.
> Finally, when I have attempted to do RIAA playback, I have noted a WIDE variety of apparent
> responses from built-in preamps in receivers. I use an RTS-405 preamp which has adjustable R and
> C cartridge loading and I first adjusted it by ear based on several LPs and a CDs that I believed
> to be mastered from the same source. After doing that I found I was within about 2 dB of the CBS
> test record at 12 kHz. I tweaked a little more to bring the CBS test record into alignment. This
> setup is much brighter and has more detail (without sounding bad) than most of the preamps in
> receivers. Of course, that exacerbates surface noise, but, when properly adjusted there was
> surprisingly little difference between the CDs and the LPs other than the noise of the LP.
> This has led me to speculate that at least one source of the critique of CDs as "sounding harsh"
> comes from a combination of overly bright speakers and rolled off RIAA preamps so that some of the
> disc noise is reduced.
> So, I think this whole area needs further discussion and narrower categories.
> At 09:09 AM 2008-08-25, Tom Fine wrote:
>>OK Charles, I'll take the bait ;) ...
>>What DSP RIAA "decoder" do you recommend? Do you also do this for tapes (ie take a flat feed at
>>either head levels or no-EQ amplified level and decode in the computer)?
>>-- Tom Fine
> Richard L. Hess email: [log in to unmask]
> Aurora, Ontario, Canada (905) 713 6733 1-877-TAPE-FIX
> Detailed contact information: http://www.richardhess.com/tape/contact.htm
> Quality tape transfers -- even from hard-to-play tapes.