| Author |
|
crapfromthepast
MusicFan
Joined: 14 September 2006
Location: United States
Online Status: Offline
Posts: 2243
|
| Posted: 12 January 2023 at 9:20pm | IP Logged
|
|
|
The promo copies of Donald Fagen's Kamakiriad CD (1993) were pressed on gold CDs, with a little sticker that read:
SPECIAL LIMITED EDITION
SUPER BIT-MAPPED
GOLD DISC
<Reprise logo>
2-45230-DJ
The matrix number of the promo CD (according to Discogs) included the notation 452302DJ.
Some people swear by this particular pressing.
From Roger Nichols: "I mastered a special 20 bit version of Donald Fagen's "Kamakiriad" album using the prototype of Apogee's UV-22 process for squeezing 20 bits onto 16 bit CDs."
From Steve Hoffman: "...it's a real "full range" digital recording. Try and find the 24 Karat Gold CD version. To me, that's the keeper; I play mine quite often." "From my own listening comparing the retail CD and this SBM promo, the SBM disc exhibits (slightly) tighter bass and better treble extension."
The commercial copies were pressed on aluminum, of course. I found three different pressings for the US commercial CD releases. All have a notation in the matrix number of "RE-1" (a first reissue), "RE-2" (a second reissue), or "RE-3" (a third reissue). I didn't see (on Discogs) any commercial releases without an "RE" desgination; the only one I found without "RE" is the gold promo disc.
I got to run a null test to compare the gold promo CD against an RE-1 commercial release. I found that the gold promo disc sounds exactly like the RE-1 commercial release, except that the gold promo CD has its left and right channel desynchronized by exactly one sample. Specifically, I can read both files into Audacity, line up the left channels to be perfectly in synch, invert one of the files, get perfect cancellation down to dithering noise (around -74 dB) in the left channel, and high-frequency bleed-through in the right channel. I can then shift one of the files by exactly one sample to get the right channels to be perfectly in synch, and I get perfect cancellation down to dithering noise (around -74 dB) in the right channel, and high-frequency bleed-through in the left channel.
So either the gold promo CD has a mastering defect, or the RE-1 commercial CD has a mastering defect. Which one? I can tell by performing an out-of-phase sum (OOPS) test. I break the left and right channels into two mono files, invert one of them, add them, and listen to what bleeds through. For the gold promo CD, I hear lots of high-frequency stuff. For the RE-1 commercial CD, there's quite a bit less high-frequency stuff. Specifically, the hi-hat is significantly more muted (because it's close to mono in the mix), and same for the lead vocals.
So... the gold promo CD has a mastering defect in which the left and right channels are out of synch by exactly one sample. Same EQ and dynamic range as the RE-1 commercial CD, same... just about everything, but the soundstage will be a tiny bit off for high frequencies for the gold promo CD.
I don't expect you can hear this with your naked ear. It eluded some mastering engineers. But you can be assured that not only does your RE-1 commercial CD sound every bit as good as the hyped gold promo CD, it actually sounds better because the left and right channels are perfectly synched.
I don't know what Reprise did differently for the RE-2 and RE-3 versions of the commercial CD. If I had to guess blindly, I'd guess that there's a small change in length somewhere, where a track or two is lengthened or shortened by a few samples of silence. I'll be happy to run some more null tests, if anyone wants to send me files.
__________________
There's a lot of crap on the radio, but there's only one Crap From The Past.
|
| Back to Top |
|
| |
AutumnAarilyn
MusicFan
Joined: 22 August 2019
Online Status: Offline
Posts: 181
|
| Posted: 14 January 2023 at 11:34am | IP Logged
|
|
|
I never bought into SACD or Gold disc phenomenon. Thanks
for clarifying in scientific audio terms what I already
assumed.
The whole "fear of missing out" concept is what drives a
lot of this. If you have an appetite for buying quantity
in today's vinyl market or even SACD/Gold disc market,
you probably only want to buy one copy. Since one doesn't
know how it will sound, probably the earliest in country
of origin on the best pressing might be best unless you
want to buy many multiples without logic.
There's actually a company called Better Records that
listens to different versions and rates each side. Some
of the differences are attributed mostly due to the last
owner's needle drag and in what sequence it came of of
the pressing line. Again earlier is assumed better with
the "old sneaker wear logic" which says that they'll be
more imperfections and distortions as the stamper wears
down.
|
| Back to Top |
|
| |
Ringmaster_D
MusicFan
Joined: 08 July 2010
Location: United States
Online Status: Offline
Posts: 212
|
| Posted: 14 January 2023 at 3:23pm | IP Logged
|
|
|
I can shed some light on the RE-2 release. The story is
that shortly after release Donald Fagen wanted exactly one
drum fill changed on "Countermoon" at 1 min 25 sec. I only
have the RE-1 release, so I've never heard the difference
myself. Over on the Hoffman forum they describe it as
this:
On the first version there's just a cymbal swell going into
the 2nd verse, on the latter version there's a little
syncopated drum fill.
|
| Back to Top |
|
| |
AdvprosD
MusicFan
Joined: 12 June 2020
Location: United States
Online Status: Offline
Posts: 354
|
| Posted: 14 January 2023 at 9:05pm | IP Logged
|
|
|
Ringmaster_D wrote:
I can shed some light on the RE-2 release. The story is
that shortly after release Donald Fagen wanted exactly one
drum fill changed on "Countermoon" at 1 min 25 sec. I only
have the RE-1 release, so I've never heard the difference
myself. Over on the Hoffman forum they describe it as
this:
On the first version there's just a cymbal swell going into
the 2nd verse, on the latter version there's a little
syncopated drum fill.
|
|
|
Well, now I have a reason to dig out my old CD and see which version I actually have. To me, a drum fill is almost meaningless. But to Donald, I assume it made the whole song better in his opinion.
__________________
<Dave> Someone please tell I-Heart Radio that St. Louis is not known as The Loo!
|
| Back to Top |
|
| |
crapfromthepast
MusicFan
Joined: 14 September 2006
Location: United States
Online Status: Offline
Posts: 2243
|
| Posted: 16 January 2023 at 12:44pm | IP Logged
|
|
|
I got rips of the files from the RE-3 release, and compared ran null tests to compred them to RE-1.
Tracks 1 and 3-7 of RE-3 are digitally identical to RE-1 (after accounting for differing amounts of silence at the beginning and end of the tracks).
Track 8 has an insignificant difference. It looks like the track was resampled so that the samples on RE-3 are just a little out-of sync with the samples on RE-1. There are edit points at 2:47, 4:45, and 5:10 that remove a handful of samples at each edit point. There aren't noises or clicks that are removed; it's just audio in seemingly random places. Bottom line: don't worry about it. Track 8 on RE-3 sounds exactly like track 8 on RE-1.
Track 2 ("Countermoon") is fascinating. The RE-3 and RE-1 cancel down to dithering noise except for added drum fills. The fills and accompanying reverb are just laid down on top of the RE-1 track, with no additional mixing. (Sorta like the 45 version of "In The Air Tonight".) You can tell that it's a live drummer playing the fills, not just a snare sample.
There are a total of 14 added drum fills at 1:03-1:04, 1:13-1:15, 1:24-1:27, 2:01-2:03, 2:12-2:13, 2:22-2:24, 2:41-2:44, 3:08-3:10, 3:18-3:20, 3:28-3:30, 3:39-3:40, 3:49-3:51, 3:58-4:01, and 4:09-4:11.
Personally, I think the added drum fills do fill in the space nicely and do improve the track a bit.
I wonder how much it cost to bring in the drummer for an extra session, record everything, create a new mix for "Countermoon" (even if it's just plunking the drum fills right over the old track), create a new master that includes the new "Countermoon" mix, and getting it into production. Clearly, it was worth it to Donald and Walter (who produced the album).
__________________
There's a lot of crap on the radio, but there's only one Crap From The Past.
|
| Back to Top |
|
| |
crapfromthepast
MusicFan
Joined: 14 September 2006
Location: United States
Online Status: Offline
Posts: 2243
|
| Posted: 16 January 2023 at 3:08pm | IP Logged
|
|
|
This was posted by Mal on the Steve Hoffman board in 2021:
We Forgot The Drums!
by Roger Nichols
If you have already purchased Donald Fagen's Kamakiriad album, you may have a collector's item. It turns out that in the mix of Counter Moon, the snare drum fills were turned down so low that you need a microscope to hear them. (I'll let you ponder that last sentence.)
Donald called me up and said that when they press the next round of CDs, he wanted the drum fills up where you could hear them. (The nerve of some people.) The catch was that he didn't want to go in the studio and re-mix the song because, other than the missing fills, it was perfect.
The drum fills were recorded on the Sony 48 track digital machine in stereo at 48kHz. They needed some reverb added to them and then mixed into the existing mix which was at 44.1kHz.
If you kept notes from one of my earlier scribblings, you will recall that when we mixed I printed the mix to a Fostex D-20B time code DAT machine while simultaneously transferring SMPTE from the Sony 48 track "just in case" we needed to lock the mix up with the original multi-track. I guess this would be that "case".
To the non Gear Slut it would be simply a matter of synchronizing both machines, patching the analog outs into a mixing console, adding some reverb to the drum fills, record the results onto another DAT machine, and edit it into the album master. But that would have only taken an hour to do and therefore goes against the grain of my very existance. And besides, running the original mix through anything would have made it a generation down from what we want our public to expect from us.
Ok, follow along closely. Since both the 48 track at 48kHz and the DAT machine at 44.1kHz were referenced to video sync during the mix, both sample rates had a common denominator. Therefore, 3 minutes 24.003 seconds is in exactly the same place on both tapes. If I lock them up the same way for this drum fill fix, they will achieve sample accurate lock even though they are at different sample rates.
One other little twist to our story. The original tape was 48 track digital, but I was at Walter's studio in Maui and he only had a pair of the new Sony 3324-S 24 track digital machines. I checked the 48 track tape and found that the drum fills we needed were on the first 24 tracks. Now we were about to find out how compatible the 24 track and 48 track Sony machines really are. We put the tape on the 24 track machine and it played back perfectly, time code track and all.
Let the syncing begin. The machines locked and released to video sync for the common clock. The drum fills were in the right place, but to make absolutely sure, I substituted the regular snare for the fill track to see if it matched up with the snare already on the mix. I panned the whole mix to the left and the snare only track from the 48 track to the right. I recorded the results into Digidesign Sound Tools and zoomed in to the sample level to make sure that everything lined up. It was perfect through the whole song. I then copied part of the regular snare into the fill tracks at the beginning and end of the tune for later reference. Now we could get down to business.
Remember, the quest here is to stay digital and be able to prove afterwards that nothing was changed in the original mix except where the fills were added. I ran the AES digital output from the Sony multi-track into the "B" input of the Roland SRC sample rate converter. I fed the digital output of the DAT machine into the Digital Domain FCN-1 box so that I could have multiple outputs from the DAT machine. One output was fed to the "A" input of the Roland SRC while the other was fed to the digital input of the four channel version of Digidesign Pro Tools. The time code from the Sony multi-track was also fed to Pro Tools so everything would have the same reference. Ok, catch your breath for a second and we will continue.
Reverb comes next. The digital output of the Roland SRC is now 44.1kHz referenced to the sample clock of the DAT machine containing the master mix. This converted signal was fed into the digital input of a Lexicon 300 reverb. A slight amount of ambiance was added to the drum fills so they match the sound of the snare on the master mix. The digital output of the Lexicon 300 was then ready to be fed to Pro Tools.
I had to make two passes to get everything into Pro Tools because the four channel version will only let two of the channels come in digitally. (Remember, "it's always something!") This is why I put the snare reference on the front and back of the fill tracks.
Remember, the master DAT mix has gone through only the FCN-1 box on its way to Pro Tools which adds a .06ms delay. The stereo drum fills have gone through the Roland SRC which adds a 2.48ms delay and the Lexicon 300 which added another 3.40ms. Our calculator tells us that the fills are going to be 3.40 + 2.48 - .06 = 5.82ms behind where they should be. By zooming in on Pro Tools I could see that the reference snare was behind the mix snare by exactly 5.82ms. I slipped the fill tracks earlier so everything lined up perfectly. It is great how sometimes things work out like they are supposed to.
Now to mix everything together. In Pro Tools, you can assign everything to the same digital output and mix them together. I played the four channel file and set the proper level for the fills. It sounded great. For you detail mongers out there, no level change was made to the original mix and the fill tracks were digital black between fills. The resulting output was a clone of the master mix except where the fills were digitally added in.
To check the clone status of our results, I performed two tests. First, I locked up the two versions of the mix and fed them to the SIM machine. It said they were perfect except where the fills came in. Second, I made two files in Sound Tools. One file was the original master mix, and the other was the new mix with fills. I digitally inverted the phase of the entire original file. I then digitally summed it with the new composite file. When I played back the resultant file, all that was left was the drum fills. Everything else completely canceled out. Satisfied?
The new mix plus fills was copied digitally to a DAT tape and sent to the mastering facility to edit into the album master. From now on all of the Donald Fagen CDs will have the version with the added snare drum fills.
So try this next time you are faced with a situation where you need to re-mix because you forgot something in the mix. When the artist asks you how long it will take, tell him five, maybe ten minutes, tops.
__________________
There's a lot of crap on the radio, but there's only one Crap From The Past.
|
| Back to Top |
|
| |
|
|
Active Topics 
Memberlist 
Search 
Help
Register 
Login
Top 40 Music on Compact Disc : Chat Board
Topic: Donald Fagen - Kamakiriad CDs (1993)
