Re: [ARSCLIST] Pristine Audio (?!)

[Tom Fine <tflists@xxxxxxxxxxxxxxxxxxx>]

Right, but I would argue that the Robert Johnson material -- at least based on the online sample --  
is a prime case of over-processing, the biggest problem of too many digi-tools and not enough 
critical ears. With a good bit of this old source material, there is a point where the digi-tools 
create more unpleasant artifacts than they improve audibility. The ear is pretty good at ignoring 
white-ish background noise like is rampant on carefully cleaned, decent-condition 78's. The ear, at 
least an ear that doesn't have shot treble sensitivity, is not good at ignoring digi-hash and 
"sparkle-grit", compounded by digi-flangeing errors, around instruments, especially when the digital 
artifacts pump loud and soft with the musical content as in the Johnson sample.

I'm a big believer in the Art Shifrin way of doing disk transfers. Hunt down the best source 
material you can find. Clean the record, and clean it well. Select the proper stylus, usually 
through experimentation. Make sure to use the correct EQ curve for the source material. In some 
cases, frequency-range-limiting is perfectly acceptable and desireable, as often in the case of 
spoken-word records, but beware the effects of cutting a range of fequencies has on the audible 
harmonics (oftentimes, undesireable noise is out of phase to certain frequencies and cutting it 
makes those frequencies jump out; this is the case often with rumble-hi-passing for instance). And 
finally, make a high-resolution transfer so that any digi-processing you have to do has the most 
information to work with. My experience is, following the above guidelines produces transfers that 
are about as audible as they're going to get. You can conservatively tweak this or that with 
digi-tools but anything aggressive usually does more harm than good.

By the way, digi-tools often get abused in regular remastering for CD. A great example is the 
Universal-licensed reissued of Ray Barretto "Acid" that came out in the past couple of years. It was 
remastered in that timeframe, says from the master tapes. The original tapes are hissy and the 
remastering engineer got too aggressive with digi-NR. The result is clearly heard on the title 
track -- horrid digi-swish floating around where much less annoying tape hiss was on the original 
LP.

There are also plenty of terrible gray-market CD's out there made from LP transfers where someone 
then must have applied the default settings of some cheap digi-tool, the result being digi-hash and 
"sandy grit" where impulse-removal removed peaks of music or took out too much content with the 
scratches in a patch of bad surface. Another bit of wisdom from Art Shifrin -- remove the ticks and 
pops by hand. My preferred method in Sony Soundforge is to literally write them out of the waveform 
with the pencil tool. Anything of quick time duration can be done this way. Bigger pops from things 
like a vinyl "zit" can be mitigated by either fudging a write-out or cutting the pop's level to 
bring it in line with the surrounding waveform. Another thing Art taught me -- don't even try to get 
grooved disks totally error-free. The ear gets used to crackle and a few short, not loud pops. Take 
out the stuff that stands way out from the musical or spoken content and the brain ignores the rest 
if the content is compelling enough to justify a transfer.

Finally, my biggest pet peave with grooved-disk source material is something that just cannot be 
fixed yet -- groove distortion. This is a gritty, square-wavey distortion caused by groove damage or 
over-cutting or bad pressing from the get-go. In almost every case when I encounter records like 
this, I just throw up my hands. This is a big reason why I avoid the mythical "original pressings" 
of many classical and jazz titles -- they weren't all that good back in the day and they are almost 
guaranteed to be ruined after they've sat in some guy's collection for 30+ years. Big exception to 
this would be some 1960's European classical and jazz records, which were pressed extremely well and 
hold up if carefully stored and not over-played. For someone like me, Mosaic Records is a god-send 
because all those Verve and Mercury and Impulse albums they reissue in those box sets were usually 
not mastered or pressed very well back in the day. Having a careful transfer of the original master 
done up by someone like Malcolm Addey puts you a lot closer to the control room on the session day 
than the original-release LP, in my opinion.

-- Tom Fine

----- Original Message ----- 
From: "Steven Smolian" <smolians@xxxxxxxxx>
To: <ARSCLIST@xxxxxxxxxxxxxxxx>
Sent: Tuesday, December 04, 2007 7:17 AM
Subject: Re: [ARSCLIST] Pristine Audio (?!)


> I suspect there may be a source problem.  It's likely that Pristine had the 78s of the classical 
> material and was working from a copy of the blues items.  That might account for the differences of 
> opinion.
>
> Steve Smolian
>
> ----- Original Message ----- 
> From: "Scott Phillips" <scottp@xxxxxxxxxxxxxxxxxxx>
> To: <ARSCLIST@xxxxxxxxxxxx>
> Sent: Tuesday, December 04, 2007 2:00 AM
> Subject: Re: [ARSCLIST] Pristine Audio (?!)
>
>
> I vote for Tom Fine's opinion.... although I have to qualify it as I
> don't know how bad the raw file sounded like before processing. I do
> have a reasonably state of the art computer sound system. Again, one
> man's opinion.
>
> -----Original Message-----
> From: Association for Recorded Sound Discussion List
> [mailto:ARSCLIST@xxxxxxx] On Behalf Of Steven C. Barr(x)
> Sent: Monday, December 03, 2007 9:18 PM
> To: ARSCLIST@xxxxxxxxxxxxxxxx
> Subject: Re: [ARSCLIST] Pristine Audio (?!)
>
> ----- Original Message -----
> From: "Tom Fine" <tflists@xxxxxxxxxxxxxxxxxxx>
> > That sample track of Robert Johnson is full of digital hash in the
> > treble,
> especially around the
> > slide guitar but also the upper notes of the voice. It sounds very
> > bad, to my
> ears. I'll take the
> > record surface noise in the early 90's Sony reissue set over the
> > digital hash
> any day. One man's
> > opinion ...
> > ----- Original Message -----
> > From: "Graeme Jaye" <graemejaye@xxxxxxxxx>
> > > Hi Steven
> > > On 03/12/2007 you wrote;
> > > SCBx> Is anyone out there familiar with Andrew Rose...Pristine
> Audio...
> > > SCBx> or their restoration process? Is it really as good as he/they
> say(s)?
> > > Yes - I know Andrew (we are in the same line of business) and he has
>
> > > done some remarkable restoration work, particularly on early
> material.
> > > He gets some amazing results.  He was also the man who uncovered the
>
> > > Hatto scandal.
> > > If you haven't already done so, have a listen to the sample track he
>
> > > has posted on his site, I know you will be impressed.
> > >
> Well...that gives me two conflicting opinions...?! I couldn't tell much
> from the available samples...partly because my PC's sound system is
> several miles north of "state of the art"...and partly because 65 years
> of abuse and a couple of major head injuries have left me with only
> fragments of the hearing I once had (which may be why I find that
> listening to 78's as such to be quite adequate...?!).
>
> Steven C. Barr
>
>
> --
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.503 / Virus Database: 269.16.13/1169 - Release Date: 12/3/2007 10:56 PM