JediJoker
2K Club - QQ Super Nova
A matrix encode can't exist without a discrete source from which to encode it. No one mixes directly into an encoder.
Chicago 2 Quadio Mix missing vocals
- Thread starter ElBlancoChocolatte
- Start date
Help Support QuadraphonicQuad:
That's exactly what CBS did, and Sigma Sound on the PIR quad mixes in Philadelphia. They ran two separate signal chains from the mixing board during the session - one 4-channel output directly into a 4-channel 1" reel-to-reel for the discrete master, and another 4-channel output into an SQ encoder that fed a 1/2 inch deck for the SQ-encoded 2 channel master.
Arthur Stoppe (who was the main quad guy at Sigma) told me that CBS's instruction was to do it this way and not to make SQ encodes from a 4-channel tape as they felt the extra tape generation could induce phase-related issues that could impair the quality of SQ decoding. For the same reason, CBS told Sigma to have all their LPs cut flat with no EQ or other mastering at Frankford-Wayne, so I presume this was probably the case with most (or all) CBS releases.
There was also an article in an early '70s issue of Billboard about the way that Ed Michel, who produced all the QS quad mixes for all the Impulse! jazz (and a lot of ABC Records) releases with Baker Bigsby, mixed for quad at the Village Recorder in West LA. According to that article, they used two studios simultaneously connected by tie-lines, one set up for creating the discrete 4-channel master and the other set up for creating the QS-encoded 2 channel master. The article suggested that they'd walk between the two studios and listen to how the mix sounded through the two different configurations - I believe the QS room probably had a 4 channel -> QS encoder -> QS decoder -> speakers routing option so they could make sure the way they were mixing discretely would also "pan out" (pun intended) when it played back in QS.
I also think in a lot of cases, the discrete master was actually the afterthought - in some cases they knew there was going to be a Q8 or QR release, but I think a lot of the time the only reason a discrete master exists is that the attention to detail of the recording engineer on the session led to one being created "just in case." They certainly had no idea that people would be seeking them out 40 years later, that's for sure. The LP was the best-selling medium of the time, and so I think most (or a lot, anyway) of matrix releases were mixed to get the most out of the crappy decoders of the time rather than to make good discrete mixes. That was definitely the case at CBS, and (in my opinion) why the majority of the early Columbia quads can sound weird and disjointed when you hear them discretely. In fact, I think Sigma's PIR mixes may be an outlier amongst matrix quad mixes in that they only mixed them to sound good discretely and never checked the SQ decodes to hear how they sounded decoded because, according to Mr. Stoppe, 'they sounded weird no matter what we did.'
Arthur Stoppe (who was the main quad guy at Sigma) told me that CBS's instruction was to do it this way and not to make SQ encodes from a 4-channel tape as they felt the extra tape generation could induce phase-related issues that could impair the quality of SQ decoding. For the same reason, CBS told Sigma to have all their LPs cut flat with no EQ or other mastering at Frankford-Wayne, so I presume this was probably the case with most (or all) CBS releases.
There was also an article in an early '70s issue of Billboard about the way that Ed Michel, who produced all the QS quad mixes for all the Impulse! jazz (and a lot of ABC Records) releases with Baker Bigsby, mixed for quad at the Village Recorder in West LA. According to that article, they used two studios simultaneously connected by tie-lines, one set up for creating the discrete 4-channel master and the other set up for creating the QS-encoded 2 channel master. The article suggested that they'd walk between the two studios and listen to how the mix sounded through the two different configurations - I believe the QS room probably had a 4 channel -> QS encoder -> QS decoder -> speakers routing option so they could make sure the way they were mixing discretely would also "pan out" (pun intended) when it played back in QS.
I also think in a lot of cases, the discrete master was actually the afterthought - in some cases they knew there was going to be a Q8 or QR release, but I think a lot of the time the only reason a discrete master exists is that the attention to detail of the recording engineer on the session led to one being created "just in case." They certainly had no idea that people would be seeking them out 40 years later, that's for sure. The LP was the best-selling medium of the time, and so I think most (or a lot, anyway) of matrix releases were mixed to get the most out of the crappy decoders of the time rather than to make good discrete mixes. That was definitely the case at CBS, and (in my opinion) why the majority of the early Columbia quads can sound weird and disjointed when you hear them discretely. In fact, I think Sigma's PIR mixes may be an outlier amongst matrix quad mixes in that they only mixed them to sound good discretely and never checked the SQ decodes to hear how they sounded decoded because, according to Mr. Stoppe, 'they sounded weird no matter what we did.'
Yes CBS did using a position encoder, with a discrete module to tap off a discrete version. The two versions were done at the same time. Early encodes would of been done from the discrete master.
Not correct, in the case of Columbia/Epic. They were done at the same time (in parallel), the same mix one encoded and one discrete.
What does that have to do with what I just said? I'm not familiar with that Q8 but I never said that discrete mixes were made from SQ. They were mixed, while being monitored through an SQ full logic decoder, resulting in an SQ encoded version and a discrete version of the exact same mix.
We are in agreement, the point being that they never made entirely separate quad mixes for the LP and tape variants of the same album.
@MidiMagic said the following, you quoted the point in bold, and I gave an example of it: the Live It Up Q8 uses an SQ decode of the quad mix instead of the discrete version.
OK, I was confused about what you were talking about. Didn't that happen with one version of "Dark Side Of The Moon" as well? That was not a normal occurrence, nor common practice!
Yeah, the US Capitol Q8s of Dark Side and Atom Heart Mother used decoded SQ as well. The discrete versions were exclusive to the UK EMI Q8s until the quad mixes were reissued on Blu-Ray a few years ago.
There are many instances where quad and stereo mixes differ. On the Q8 you can hear a piano all the way through “Mississippi Queen”, not just a few notes at the end. Same with “Operator” where the piano is in the right rear and you hear it all through the song. There are songs like “Crimson and Clover” where one mix has chorus cross corner and another where chorus goes around the room.
MidiMagic
2K Club - QQ Super Nova
- Joined
- Jul 5, 2010
- Messages
- 2,083
Remember that the matrix version can be optimized to take advantage of the best features of the particular matrix system.
They can mix both the discrete and the matrix at the same time with two mixers. That way, they can make them SOUND the same through the monitors even though the mixes are electronically different.
I always mix directly into the encoder. The encoder plugs into the bus inserts of my mixer. There is never a discrete source other than the original parts on their channel strips. The monitors are fed by the decoder.
And in some cases they used two mixer boards so the SQ could be tweaked separately. Like this:
Or maybe a mixer with QS encoding -> QS decoder.
That makes sense - especially in the early days when they did not see that there was a loss of information.
This also makes more sense.
This is similar (but not the same as) decoding the SQ to make the discrete.
Last edited:
And for some things, like Santana Lotus I think, there never was a discrete tape created - just the encoded tape.
The Sony Japan SACD of Lotus used the discrete tape, which turned out to have just reverb and the audience mics in the rear speakers.
I don't believe that to be correct. Lotus was released single inventory SQ encoded but the Sony Japan SACD contains virtually only applause in the rear channels. Too discrete to have been extracted from a decode of an encoded mix. I prefer the SQ decode to the discrete as there is a bit more rear channel activity.
David Bowie "Serious Moonlight" on the other hand was done as an encoded tape, no discrete version that I know of.
But weren’t a number of new bonus tracks newly mixed for the Japan SACD?
Could they have just gone back to the multis and simply added the discrete applause rears to the existing SQ tape when creating the DSD quad master for continuity with the new bonus tracks?
That would seem consistent with the previous knowledge (as I recall) that no discrete quad tape for Lotus was ever made - and not particularly hard or troublesome to do since the multis were already pulled for mixing the bonus tracks.
I’m not an expert on the album (and don’t even really like it that much to be honest)…I’m just thinking out loud.
It's the guitar playing the melody line in the intro. Sounds very empty without it. But yeah, if you've never heard it.
It's the guitar playing the melody line in the intro. Sounds very empty without it. But yeah, if you've never heard it.
I meant to mention this when I did my post, but I have a theory on this missing guitar.
It's probably on YouTube somewhere, but when the 5.1 mix of Pink Floyd's Wish You Were Here came out on the Immersion box, there were a number of featurettes put out and one of them was about the remix. One of the things James Guthrie mentioned was that when he was doing the mix, he got to a part in one of the songs that had a synth lead in the stereo mix that was completely missing from the multitrack tape. He of course investigated, wondering if maybe there was another slave multitrack that contained the synth parts, but nothing could be found - but then he noticed while playing back the multitrack tape, that the section with the synth part on it had been physically spliced into the master tape, ie the tape looked like this:
[ ORIGINAL TAPE WITH FIRST PART OF SONG ] [SECTION WITH SYNTH SOLO] [ ORIGINAL TAPE WITH REMAINDER OF SONG ]
(I know, quite the artist's rendering
)This led him to the theory that the band had mixed the whole song only to later decide that they wanted to add a synth lead to that middle part, but didn't want to remix the whole track to do so. So what they did was played back that stereo mix in the studio, had Rick Wright play the missing synth part live, and fed the stereo mix and the synth part through the mixing board on to another stereo master tape. So what you end up with there is a synth performance that (unlike the rest of the album) exists nowhere except for on that stereo bounce-down/mix-down (whatever you want to call it) tape. Then, being the quality-control freaks they are, rather than using that tape for the whole song because it would add another generation of tape hiss, they cut the section of the tape with the solo on it at the beginning and end, and then did the same to the original master tape that was missing the solo, removing it and swapping in that second-generation tape that contained the solo in its place.
Now the reason I bring up this long and complicated anecdote is because my best guess is a similar thing happened to Chicago - they mixed Wishing You Were Here in stereo, thought it was done, and then later had the same realization as you, that the intro sounded kind of empty. I think it's hard to understate how getting a great mix was in the pre-mixing console automation days was akin to capturing lightning in a bottle, because every knob-twiddle, fader move, outboard effect had to be done in real-time as the tapes were running, which is why you often saw 2 or 3 guys manning a mixing desk in some of those old photos. So rather than remix the whole song, Chicago did what Pink Floyd did and spliced in an intro where they'd overdubbed the melody guitar on the original stereo mix, and that explains (to my mind) why it's missing from the quad mix, because it doesn't exist anywhere except for the final stereo master.
The reason I've gone to such convoluted ends to come up with this explanation is that the quad mix was done by Wayne Tarnowski, who worked on every album the band did starting with Chicago II and was the main recording and stereo mix engineer on every album they recorded from Chicago VI though Chicago XI. He wasn't some hired gun who came in to do this quad mix unfamiliar with the band, and unlike the other Chicago quad mixes, Chicago VII was mixed in quad just months after the stereo version, so I think Tarnowski would've been intimately familiar with the material. The missing acoustic guitar in Wishing You Were Here is a pretty glaring omission - it's not some obscure production detail like the snare drum in the second verse missing some reverb or something, it's an integral part of the original stereo mix, and I don't think it would've been left out simply due to carelessness.
I was only ever familiar with "Wishing You Were Here" from AM radio. When I bought the SQ, LP I didn't notice anything missing, and I've never listened to the stereo LP! The idea that the "missing" section was spliced into the stereo mix makes perfect sense.
Remember as well that singles (vs album tracks) often differed from album tracks, often with parts missing or added. So variances from stereo to quad are to be expected as well. I more often find new 5.1 mixes disturbing when dominant sound elements (of the stereo version) get buried in the new mix (like in Roxy Music).
On the other hand I can't think of a single case of a quad mix that I didn't like because it differed from the stereo.
Remember as well that singles (vs album tracks) often differed from album tracks, often with parts missing or added. So variances from stereo to quad are to be expected as well. I more often find new 5.1 mixes disturbing when dominant sound elements (of the stereo version) get buried in the new mix (like in Roxy Music).
On the other hand I can't think of a single case of a quad mix that I didn't like because it differed from the stereo.
Similar threads
