| Author |
|
crapfromthepast
MusicFan
Joined: 14 September 2006
Location: United States
Online Status: Offline
Posts: 2192
|
| Posted: 10 September 2022 at 7:17pm | IP Logged
|
|
|
"Respect" would have been a standout in any year, but that it stood out in the extraordinary field of pop in 1967 is even more remarkable. In my DJ sets from the '80s and '90s, it was a guaranteed floor-filler (for me, wedged between Stevie Wonder's "Signed Sealed Delivered" and Johnnie Taylor's "Who's Making Love"). The mono 45 and mono LP version have a fade starting about 2:18 and running to about 2:22. The stereo LP version runs at the same speed as the mono 45, with the fade starting about 2:18 and running to about 2:26.
Stereo
I'm not sure where the stereo version first appeared on CD. According to Discogs, it may be Atlantic's Aretha Franklin Best Of (1984). (This was one of a series of 1984 "Best Of" releases on Atlantic. The full series included collections from Joe Tex, Sam & Dave, Aretha Franklin, Booker T & The MG's, Otis Redding, and Wilson Pickett. I'm not sure how many of them made it onto CD.) Discogs shows the release date of this CD as 1984, but I'm skeptical that it came out at the same time as the LP/cassette.
The same analog transfer is used on just about all other CDs that include the stereo version of "Respect", including:- Atlantic's 2-CD Aretha Franklin 30 Greatest Hits (1985) - digitally identical; it's possible (likely?) that this came out on CD before Best Of
- Warner Special Products' Atlantic Soul Classics (1985)
- Silver Eagle Records' 2-CD Rockin' Down The Block (1987)
- Time-Life's Classic Rock Vol. 5 1967 (1988) - fade is about 3 beats shorter
- Atlantic's Golden Age Of Black Music 1960-1970 (1988) - digitally exactly 3.4 dB louder, but fade starts too early at 2:12 and is about 6 beats shorter
- JCI's Soul Sixties (1988)
- Atlantic's Atlantic Rhythm And Blues 1947-1974 (1991 rerelease) - fade is about 2 beats shorter and has added noise reduction
- Time-Life's Rhythm And Blues Vol. 10 1967 (1991) - fade is about 4 beats shorter
- Time-Life's Solid Gold Soul Vol. 2 1967 (1991) - digitally identical to Rhythm And Blues Vol. 10 1967, fade is about 4 beats shorter
- JCI's Only Rock And Roll 1965-1969 (1994)
- MCA's 2-CD Soul Train 25th Anniversary Hall Of Fame (1995) - mastered a little loud
Some outliers that don't use the same analog transfer as the above:- Heartland/Warner Special Products' 2-CD Real Rock (1987)
- Sony Music Special Products' 2-CD 35 Years Of Rock And Roll (1992)
- MFSL's full-length Aretha Franklin album I Never Loved A Man The Way I Love You (1993, UDCD 574) - EQ'd so that horns don't shine through as well as on 1984 mastering
If you must hear the song in stereo, 30 Greatest Hits is a great Aretha overview, which sounds just fine and is very inexpensive nowadays.
Mono
1985: The mono version first appeared on CD on Atlantic's multi-disc Atlantic Rhythm And Blues 1947-1974 (1985 original release). While my modern-day ears agree with Atlantic's choice to use the hit mono mix in their first multi-disc box, early adopters of the CD format wanted stereo. This set was not received well, to the point where Atlantic redid the entire set and rereleased it in 1991, replacing many mono tracks with stereo versions where available. Sonically, the 1985 set is OK overall. The EQ is a little flat-sounding, but there's plenty of dynamic range, the fades are left intact, and there's no added noise reduction. (The 1991 does indeed have added noise reduction on some tracks.) On this track, the left and right channels have some phase issues (showing up when out-of-phase-summed), but if you use just the left channel or just the right channel, you'll be fine. Other CD releases ignored the mono version until...
1992: Rhino did a new analog transfer for their Aretha Franklin box set Queen Of Soul (1992), which was mastered by Ted Jensen. This box set is stellar in every way, from the track selection to the sound quality to the packaging. The box uses some mono mixes, which is great. (Not all mono, though; some early tracks are in stereo.) There's a slight difference in volume between channels (about 0.2 dB), but no desychronization between left and right channels. The same analog transfer is used on:- Rhino's Billboard Top Rock 'N' Roll Hits 1967 (RE-1 reissue, 1993) - digitally about 0.2 dB quieter
- Rhino Special Edition's budget Aretha Franklin collection Chain Of Fools (1993) - digitally about 0.2 dB quieter
- Rhino Special Products' 2-CD Entertainment Weekly Presents Pure Party (1993)
- Rhino's budget Soul Hits Vol. 1 (1993)
1994: Rhino did another new analog transfer for their Aretha Franklin collection Very Best Of The '60s (1994), which was mastered by Bill Inglot and Dan Hersch. This song is a little louder than others on the disc, so that the peaks are reduced a bit in the last half of the song (when compared to the earlier analog transfers). You can see it on the screen, but I doubt that you'll be able to hear any problems with it. The EQ boosts the high end, typical of the mid-'90s Rhino masterings. The same analog transfer is used on:- Rhino's More Stadium Rock (1996)
- Rhino's Atlantic Ultimate '60s Soul Smashes (1998)
- Rhino's Instant Party Disc Regular Strength (1999) - differently EQ'd digital clone
- Rhino's New Millennium Soul Party (2000) - differently EQ'd digital clone
- Time-Life's Legends Of Soul Vol. 4 Aretha Franklin (2001) - digitally exactly 1 dB quieter
2018: Atlantic/Rhino did one more analog transfer for the 2-CD Aretha Franklin The Atlantic Singles Collection 1967-1970 (2018), which was mastered by Scott Levitin. It's pure mono (out-of-phase-sums down to dithering noise). The dynamic range is great, the EQ is great (no high-end boost), there's no evidence of noise reduction on the fade, and best of all, the fade extends about a beat longer than everything else listed above! "Respect" is superb here. Everything else I've heard from this CD set is also breathtaking - it even fixes the dropout of the intro of "Since You've Been Gone"!
My recommendation
Not to downplay the Queen Of Soul set, which is terrific, but I wholeheartedly recommend the 2-CD The Atlantic Singles Collection 1967-1970 (Atlantic/Rhino 571349, released 2018). The tracks are all mono, like the 45s, and the sound quality is exactly what I'd want from these tracks. I haven't seen better presentation anywhere of these essential 45 sides from the Queen Of Soul.
__________________
There's a lot of crap on the radio, but there's only one Crap From The Past.
|
| Back to Top |
|
| |
LunarLaugh
MusicFan
Joined: 13 February 2020
Location: United States
Online Status: Offline
Posts: 327
|
| Posted: 12 September 2022 at 6:59pm | IP Logged
|
|
|
The best thing about the more recent Atlantic Singles Collection is that the new transfers were made using the Plangent Process.
This is real audio nerd type stuff but basically the process involves recovering the bias tone of the source tape to solve any
pitch fluctuations that may occur in six decade old tapes as well as during the recording itself. This means you hear what was
originally printed to the tape virtually how it sounded when it was recorded. An online acquaintance of mine pioneered this
technique and everything I've heard it used on sounds absolutely brilliant.
__________________
Listen to The Lunar Laugh!
|
| Back to Top |
|
| |
crapfromthepast
MusicFan
Joined: 14 September 2006
Location: United States
Online Status: Offline
Posts: 2192
|
| Posted: 12 September 2022 at 8:59pm | IP Logged
|
|
|
(From https://www.mixonline.com/recording/analog-tape-101-part-3-b ias-magic-373029: "Bias is a Radio Frequency (RF) signal - anywhere from 60 kHz (for cassette decks) to 432 kHz (for the Ampex ATR-100 Series) – that is mixed with the audio signal.")
The bias frequency is way higher than what we can hear (around 20 kHz), so it's usually ignored by the audio processing chain.
It's a pretty brilliant idea to use the tape bias itself as a synchronization device.
I have a bunch of FM radio airchecks that I originally recorded onto Hi-Fi VHS tape. If you look at a spectrum of the audio that comes out of the FM receiver, the audio ends cold at 15 kHz, there's nothing between 15 kHz and 19 kHz, and there's a very pronounced pilot tone at 19 kHz, which is used for FM stereo (in the US). I always thought that the pilot tone would be helpful for fixing the speed of a recorded FM broadcast (you just adjust the speed until the 19 kHz tone is truly at 19 kHz). It turned out that Hi-Fi VHS tape was extremely accurate and didn't need any speed correction. It also turned out that some receivers didn't output anything over 15 kHz, and a lot of cassette decks (which would benefit from the speed correction) didn't record up to 19 kHz. Oh well.
The Plangent Process people have gone well beyond using that pilot tone for speed correction. They're actually using the bias tone to do jitter correction, by reading the bias tone off the tape, and doing local speed corrections to set the bias tone off the tape to what it should be (like a 432 kHz sinusoid for the Ampex ATR-100 Series).
It's really a genius idea.
__________________
There's a lot of crap on the radio, but there's only one Crap From The Past.
|
| Back to Top |
|
| |
mjb50
MusicFan
Joined: 28 April 2021
Location: United States
Online Status: Offline
Posts: 265
|
| Posted: 13 September 2022 at 9:24am | IP Logged
|
|
|
For what it's worth, the new mastering is objectively quite excellent, but I personally hear no appreciable difference between the original mono Atlantic Rhythm and Blues 1947–1974 mastering and the new The Atlantic Singles Collection 1967–1970 one, since there are no noticeable speed issues on the old transfer at all.
What's interesting to me about the Plangent Processes system is that the first and last steps are not all that difficult, technologically. The actual speed correction method is simple; it's just varying the playback sample rate, which is what vibrato effect plug-ins do. Capturing a ~100 kHz tone from a tape is within reach, too, although doing a really good job of it requires a custom deck which can play very high frequencies without adding noise, and the ability to capture at 384 kHz sample rates. (On a budget, I'd sample at 96 kHz while playing the tape at 25% speed, although I'm skeptical I could do it on an ordinary deck without adding audible noise and I'm not sure I'd really get the best "image" of the bias tone.)
The real trick, the middle third, is analyzing that captured bias tone, so that your software knows exactly what adjustment to make to the playback rate from moment to moment. It's complicated enough that Plangent Processes has the market cornered. They've probably also refined their system over the years to make it more complicated than just this.
When I'm doing speed correction, it's a similar process, but I'm eyeballing frequency plots and just adjusting the playback rate of the entire track, not changing it from moment to moment. I'm also not looking at the bias tone, but at some other tone that I know the ideal frequency of, e.g. musical notes or electrical noise (mains hum or tube TV interference). In the end, I'm only making the average frequency meet the ideal, to get the overall pitch right, not be precise to the millisecond or fix wow & flutter or slowly changing tape drag.
I've long been hoping for a similar system for correcting the wow caused by an off-center hole on a record. I was once in contact with a vibrato effect developer who was going to do it, but he fell out of touch. To work for what I wanted, it just needed to be a little more configurable by the user. I'm surprised no one has done it yet.
Edited by mjb50 on 14 September 2022 at 12:22pm
|
| Back to Top |
|
| |
|
|
Active Topics 
Memberlist 
Search 
Help
Register 
Login
Top 40 Music on Compact Disc : Chat Board
Topic: Aretha Franklin - Respect
