Tom Fine: "Tape hiss "reduction" is even worse because it usually leaves
artifacts worse than the tape hiss."
Agreed. Usually my policy towards that is I will use noise reduction and
listen first to see particularly what the opening sounds like. If it's full
of those warbles, wobbles and chemical
sounds I will back out of it and go with the track as is. I'd rather hear
hiss than the alternative.
There is a guy on the web, whose stated specialty is removing "grinding
noises," that uses noise reduction in such an aggressive way that
practically nothing he transfers is listenable.
best,
Dave Lewis
Lebanon, OH
On Wed, Jun 4, 2014 at 7:49 AM, Tom Fine <[log in to unmask]>
wrote:
> When did the obsession with click and tick removal begin? It may go back
> to broadcasting and the original Scott Impulse Noise Reducer. I think it's
> one of the dumbest obsessions in professional audio, right up there with
> obsessive tape-hiss "reduction" schemes in digital remastering (aka "Suck
> The Life Out Of Everything Code"). It's been proven time and time again
> that ordinary people enjoying recorded music hear right through disk
> surface noise and tape hiss. If one concentrates on using the best possible
> source material and doing superb analog playback, the known imperfections
> of analog media shouldn't be a problem.
>
> Even dumber than the obsession with click and tick removal is the nip and
> tuck method, which creates permanent micro-errors in the timing and pacing.
> If one must mess with the dubbed tape, the John RR Davies method of
> "scrubbing away" the oxide where the tick or click lived was a better plan,
> because it preserved the entire time spectrum. In today's world, the only
> suitable method I've heard to both preserve the right audio sound quality
> and remove the (hopefully few) bad ticks and pops is to do it by hand in a
> waveform editor, to literally write-out the tick with the pencil tool. I
> will admit that some modern software solutions are quite good, and it's
> hard to hear the artifacts over speakers. But, if one listens carefully on
> headphones, the artifacts lurk in the background -- or the sonic balance
> has been reduced or altered vs an unprocessed playback. Tape hiss
> "reduction" is even worse because it usually leaves artifacts worse than
> the tape hiss.
>
> -- Tom Fine
>
> ----- Original Message ----- From: "Ted Kendall" <
> [log in to unmask]>
> To: <[log in to unmask]>
> Sent: Wednesday, June 04, 2014 7:12 AM
>
> Subject: Re: [ARSCLIST] Odd warble
>
>
> Yes, come to think of it, declicking in 1984 was a choice of the Garrard
>> MRM (fast muting), the SAE 5000 (switching between direct and delayed
>> signal paths) or the Packburn (heavy top cut). Still, the worst of it, to
>> my ears, is the cut and shut tape editing. we all had to do it then, but
>> some of us took more care than others.
>>
>> As regards full track tapes, undoubtedly the best arrangement, all things
>> being as they should, is a full track head. Playback of a partial width
>> involves degradation of s/n ratio and may result in incorrect frequency
>> response - spread gaps were not uncommon in those days, so the response
>> from the full width was an average of those obtained by the differing
>> effective gap lengths across the tape. This also affects the crossfade
>> effect on splices, for which the full tape width is desirable. That said,
>> phasing effects are both unpleasant and irremovable.
>>
>> My own approach is to start with a decent transport - Richard and I both
>> like the A80, and I use a Telefunken M10A for the really awkward squad -
>> the trailing capstan holds the tape against the head so well it usually
>> doesn't even think about misbehaving. Then, a combination of a narrow guard
>> band butterfly head and azimuth correction software (Cedar) usually ensures
>> that the signal comes off straight and square.
>>
>>
>>
>>
>>
>>
>>
>>
>> On 04/06/2014 11:33, Randy A. Riddle wrote:
>>
>>> The "warbles" show up in the same places on both the cd and lp versions
>>> of
>>> these Radiola lps.
>>>
>>> Goldin seemed to use different pressing facilities over time for his
>>> discs
>>> - perhaps a full track tape was played back with 2-track heads for
>>> mastering both the lp and cd versions of the release. In 1984, though, I
>>> was thinking that your choices for cd mastering and pressing were much
>>> more
>>> limited and probably wouldn't be done at the same plant.
>>>
>>> It's a mystery we'll probably never figure out unless I get some more
>>> concrete info from Goldin himself on how he mastered these releases.
>>>
>>> Randy
>>>
>>>
>>> On Wed, Jun 4, 2014 at 6:13 AM, Tom Fine <[log in to unmask]>
>>> wrote:
>>>
>>> The digital artifact could well be an uncorrectable error with the
>>>> 1600/1630 system used to master the CD.As I recall, very brief
>>>> uncorrectable errors don't completely mute, or the mute is so short it
>>>> sounds like a digi-click-warble, like playing over a bad scratch on a
>>>> CD,
>>>> where there's a click/spike instead of program audio but not a mute.
>>>>
>>>> That said, after hearing from Richard Hess and others, the warble could
>>>> be
>>>> from the analog domain, a bad nip and tuck splice played back
>>>> incorrectly
>>>> (splice made in FT, tape played for CD master on 2T machine).
>>>>
>>>> By the way, 2T playback of FT tapes is fairly rampant in the music
>>>> remastering world. I hear it on a lot of mono jazz CDs. Even if you get
>>>> the
>>>> azimuth perfect, there will still be channel differences on 50+ year old
>>>> tapes because of dropouts, warps, splices, etc. I can't understand why
>>>> every professional player-back of reel tapes doesn't have a FT head.
>>>> There
>>>> are lots of mono tapes out there. The problem with summing 2T channels
>>>> (another common practice) is that 50+ year old tapes rarely travel the
>>>> tape
>>>> path perfectly, so there ends up being "country lane-ing" and thus
>>>> azimuth
>>>> shifts and flange/phase artifacts.
>>>>
>>>> The latest flagrant example of all of these problems is the Record Store
>>>> Day special issue "Gems From Sun Records Vol 1." All of the content is
>>>> MONO, but try summing a stereo playback or playing the record with a
>>>> mono
>>>> cartridge. You will not be happy with the results. Listen especially to
>>>> the
>>>> middle cuts on each side, which seems to have been made from the worst
>>>> warped/damaged tapes. Also the Charlie Rich cut. For most cuts, summing
>>>> to
>>>> mono produces pumping phase/flange problems, indicating the master tapes
>>>> are badly warped, probably from vinegar syndrome, and aren't passing
>>>> over
>>>> the 2T play head smoothly. When played back in stereo, it's not annoying
>>>> because the anomolies are out on the sides and the high-spl content is
>>>> in
>>>> the middle.
>>>>
>>>> I think Richard Hess and maybe others use special thinner FT heads,
>>>> which
>>>> just read the center half of the track. I very much agree with this
>>>> method
>>>> and would definitely purchase such a head if I were about to undertake a
>>>> large job of 50+ year old FT tapes. I'm wondering about the ideal
>>>> head-track width, would it be akin to 1 track of a 2-track head but in
>>>> the
>>>> middle of the head, or more akin to 1 track of a 1/2" 3-track head
>>>> (about
>>>> half again as large), in the middle of the head? The goal would be to
>>>> read
>>>> the "meat" of the track but not the edges, where warpage and shrinkage
>>>> leave less signal and more artifacts. I think you'd still have
>>>> country-laneing problems, but could the head itself be "cupped" to allow
>>>> warped top and bottom to just hang in space while the unwarped center
>>>> passed over the gap? Could guidance be made to allow that through the
>>>> tape
>>>> path?
>>>>
>>>> -- Tom Fine
>>>>
>>>> ----- Original Message ----- From: "Ellis Burman" <
>>>> [log in to unmask]>
>>>> To: <[log in to unmask]>
>>>> Sent: Tuesday, June 03, 2014 3:04 PM
>>>>
>>>> Subject: Re: [ARSCLIST] Odd warble
>>>>
>>>>
>>>> I'm unaware of any digital restoration tools in 1984. Those came a
>>>> little
>>>>
>>>>> later, around 1987 (Sonic Solutions and Cedar).
>>>>>
>>>>> That said, these sound like interpolation artifacts. Maybe there was
>>>>> some
>>>>> crude form of interpolation back then, or even just using editing to
>>>>> fix a
>>>>> loud thump. It is made MUCH worse IMHO because, even though the audio
>>>>> is
>>>>> mono, they used a stereo cartridge and interpolated each channel
>>>>> separately, instead of combining the channels and working from a mono
>>>>> source file. The ear is VERY sensitive to minor phase differences
>>>>> between
>>>>> channels. If the work was done on a mono file instead, the
>>>>> interpolations
>>>>> would be much more palatable, as they would not be constantly
>>>>> distracting
>>>>> your ear away from the center mono image.
>>>>>
>>>>> It doesn't sound like any (broadband) noise reduction was applied, but
>>>>> I
>>>>> guess "impulse noise reduction processing" would be an appropriate
>>>>> term.
>>>>>
>>>>> (the other) Ellis (from Los Angles)
>>>>>
>>>>>
>>>>> On Tue, Jun 3, 2014 at 5:04 AM, Randy A. Riddle <
>>>>> [log in to unmask]>
>>>>> wrote:
>>>>>
>>>>> Tom --
>>>>>
>>>>>> Is the "warble" directly because of "nip and tuck" or do you think
>>>>>> it's
>>>>>> exacerbated by the use of digital tools?
>>>>>>
>>>>>> Were these digital tools commonly available around '84?
>>>>>>
>>>>>> I'm thinking the best description in the discography I could use
>>>>>> might be
>>>>>> "artifacts from noise reduction processing". Some releases are worse
>>>>>> than
>>>>>> others - this is one of the more annoying examples.
>>>>>>
>>>>>> Randy
>>>>>>
>>>>>>
>>>>>> On Tue, Jun 3, 2014 at 7:33 AM, Tom Fine <[log in to unmask]
>>>>>> >
>>>>>> wrote:
>>>>>>
>>>>>> It sounds like a combination of two bad practices -- overuse of
>>>>>>> digital
>>>>>>> "tools" which cause artifacts, and also bad nip and tuck splicing to
>>>>>>>
>>>>>> remove
>>>>>>
>>>>>>> bad ticks. I doubt that's a direct transfer to master of the disk. I
>>>>>>>
>>>>>> bet
>>>>>>
>>>>>>> the disk was dubbed to tape at some pre-digital time, the old nip and
>>>>>>>
>>>>>> tuck
>>>>>>
>>>>>>> method was employed, and then when it was time to remaster for CD,
>>>>>>>
>>>>>> someone
>>>>>>
>>>>>>> decided to over-use the newest "tools." The company that bought
>>>>>>>
>>>>>> Goldin's
>>>>>>
>>>>>>> retail operation, Radio Spirits, is notorious for terrible digital
>>>>>>> processing. The only people I know who consistently make good audio
>>>>>>>
>>>>>> from
>>>>>>
>>>>>>> OTR transcriptions are Art Shifrin in Queens and a guy last name
>>>>>>> Ellis
>>>>>>>
>>>>>> who
>>>>>>
>>>>>>> sells MP3 and WAV downloads out of Virginia.
>>>>>>>
>>>>>>> Speaking of Vic Damone, here's a dub of scatchy Mercury 78 promoting
>>>>>>> Damone's first 78 album:
>>>>>>> https://dl.dropboxusercontent.com/u/55748706/Mercury_Promo%
>>>>>>> 20-%20Virgil_Trucks.mp3
>>>>>>> some context: Jackie Smith was Mercury's pop promotions person out of
>>>>>>> Chicago. She ended up being a pioneering businesswoman in the
>>>>>>> midwest.
>>>>>>> Virgil "Fireball" Trucks pitched for the Detroit Tigers and later for
>>>>>>>
>>>>>> the
>>>>>>
>>>>>>> Washington Senators. He and Smith must have worked together during
>>>>>>>
>>>>>> WWII.
>>>>>>
>>>>>>> "Merc" was a smiley-faced cartoon "trademark" for the label, briefly
>>>>>>> replacing the stern-faced Mercury romanesque messenger head. It's
>>>>>>> clear
>>>>>>> that a record company wanting to take itself seriously would do away
>>>>>>>
>>>>>> with
>>>>>>
>>>>>>> "Merc" in its promotions, as Irving Green & Co quickly did. The other
>>>>>>>
>>>>>> side
>>>>>>
>>>>>>> of the 78 is a Vic Damone side from that album. I'm not sure if the
>>>>>>>
>>>>>> record
>>>>>>
>>>>>>> was meant to be played in stores or on-air. I didn't use anything to
>>>>>>>
>>>>>> clean
>>>>>>
>>>>>>> up the sound, so it's noisy. If someone wants to play with digi-tools
>>>>>>>
>>>>>> on
>>>>>>
>>>>>>> it, ping me offlist and I'll send you WAV.
>>>>>>>
>>>>>>> -- Tom Fine
>>>>>>>
>>>>>>> ----- Original Message ----- From: "Randy A. Riddle" <
>>>>>>> [log in to unmask]>
>>>>>>> To: <[log in to unmask]>
>>>>>>> Sent: Tuesday, June 03, 2014 6:26 AM
>>>>>>> Subject: Re: [ARSCLIST] Odd warble
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Here's a sample that includes some "warbles". It's a 14 mb .wav
>>>>>>> file
>>>>>>>
>>>>>> that
>>>>>>
>>>>>>> was ripped directly from the cd "Pop Singers on the Air!".
>>>>>>>>
>>>>>>>> The cd was issued in 1984, so this would have been some kind of
>>>>>>>>
>>>>>>> technology
>>>>>>
>>>>>>> available during that time. My guess is that it was some kind of
>>>>>>>>
>>>>>>> hardware
>>>>>>
>>>>>>> box that did click removal that was adjustable - in the sample, you
>>>>>>>>
>>>>>>> hear
>>>>>>
>>>>>>> the "warble" in addition to some broader surface noise on the
>>>>>>>> original
>>>>>>>> transcription. Again, you can hear the same "warbles" on both the
>>>>>>>> lp
>>>>>>>>
>>>>>>> and
>>>>>>
>>>>>>> cd versions of some of the Radiola releases, so it's in the master
>>>>>>>>
>>>>>>> tape
>>>>>>
>>>>>>> and
>>>>>>>> not an artifact of the media.
>>>>>>>>
>>>>>>>> https://dl.dropboxusercontent.com/u/54210054/warble-sample-
>>>>>>>>
>>>>>>> radiola.wav
>>>>>>
>>>>>>> Randy
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> On Tue, Jun 3, 2014 at 12:52 AM, Paul Stamler <[log in to unmask]>
>>>>>>>>
>>>>>>> wrote:
>>>>>>
>>>>>>> On 6/2/2014 8:03 PM, Ellis Burman wrote:
>>>>>>>>
>>>>>>>>> Hi Randy. Can you make short mp3 of just that section? Maybe 10
>>>>>>>>>
>>>>>>>>>> seconds
>>>>>>>>>> or so? Then you could use a higher bit rate, and hopefully
>>>>>>>>>> preserve
>>>>>>>>>>
>>>>>>>>> it
>>>>>>
>>>>>>> better. If you make the section short enough, you can even send it
>>>>>>>>>>
>>>>>>>>> as
>>>>>>
>>>>>>> an
>>>>>>>>>> uncompressed WAV.
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Or a .flac file.
>>>>>>>>>
>>>>>>>>> Peace,
>>>>>>>>> Paul
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>
>>>>> --
>>>>> Ellis
>>>>> [log in to unmask]
>>>>> 818-846-5525
>>>>>
>>>>>
>>>>>
>>>>>
>>
>>
|
