Mix differences between SQ lp and Quad 8-track versions of the same recording.

[Q-Eight]

I've often wondered if the Q8 of Blood, Sweat & Tears "Greatest Hits" mistakenly received the SQ master instead. Hear me out on this one.
We do know that many SQ titles received modified mixes because even Columbia/Sony engineers knew that SQ was less than exemplary. So, two tapes containing differing mixes must have been made at various stages of production.

An example would be B,S&T's Greatest Hits. Now, I've never heard the SQ LP, but frankly, I don't think I need to. If we compare the mix on the Q8 versus the unreleased mix on one of the Mike Robin Reels.... the mixing and editing is almost identical. However, it seems that on the Q8, the separation has been cranked down so that if a Horn is supposed to play out of back left, you're still getting horn in front left at about 30% volume. It's like they were blending things back together after the main mix had been done. Yet on the Robin Reel, that horn is plainly in back left and nowhere else.

I often wonder if this is the case with one of the later Santana titles.... is it Festival? or Borboletta? (I can't remember).

 

[furui_suterioo]

unreleased mix on one of the Mike Robin Reels

I had been wondering about the source of a download I have of BS&T "unreleased mix", the mix is completely discrete and it sounds too good to be from a Q8, it's one of the best quad mixes I've ever heard.

 

[Doug G.]

I believe it's been established, in the past, that some 8-track quads were made from SQ masters. I think there are two different versions of DSOTM, one truly discrete and one SQish. There are others, too.

Doug

 

[furui_suterioo]

I believe it's been established, in the past, that some 8-track quads were made from SQ masters. I think there are two different versions of DSOTM, one truly discrete and one SQish. There are others, too.

Doug

I've heard that the Atom Heart Mother UK Q8 is also like DSOTM, discrete mix on Q8 in UK and SQ master on Q8 in the US, is this true?

 

[fredblue]

I've often wondered if the Q8 of Blood, Sweat & Tears "Greatest Hits" mistakenly received the SQ master instead. Hear me out on this one.
We do know that many SQ titles received modified mixes because even Columbia/Sony engineers knew that SQ was less than exemplary. So, two tapes containing differing mixes must have been made at various stages of production.

An example would be B,S&T's Greatest Hits. Now, I've never heard the SQ LP, but frankly, I don't think I need to. If we compare the mix on the Q8 versus the unreleased mix on one of the Mike Robin Reels.... the mixing and editing is almost identical. However, it seems that on the Q8, the separation has been cranked down so that if a Horn is supposed to play out of back left, you're still getting horn in front left at about 30% volume. It's like they were blending things back together after the main mix had been done. Yet on the Robin Reel, that horn is plainly in back left and nowhere else.

I often wonder if this is the case with one of the later Santana titles.... is it Festival? or Borboletta? (I can't remember).

i think the one you're thinking of is "Festival", which sounds equally terrible on Q8 and SQ, a clangy reverby mess with no separation. "Borboletta" is a great mix on Q8 and the SQ decodes well enough.

 

[Q-Eight]

I believe it's been established, in the past, that some 8-track quads were made from SQ masters. I think there are two different versions of DSOTM, one truly discrete and one SQish. There are others, too.

Doug

That's pretty much exactly my point ; though with the DSOTM Q8's, it was pretty definitive. Rarely do we come across such an example from Columbia, but I think B,S&T might be one. The Q8 mix sounds like the discrete mix, just collapsed somehow. Hard to explain.

i think the one you're thinking of is "Festival", which sounds equally terrible on Q8 and SQ, a clangy reverby mess with no separation. "Borboletta" is a great mix on Q8 and the SQ decodes well enough.

Thanks for clearing that one up. I'm a casual Santana fan at best and couldn't recall which witch was which witch.

 

[jaybird100]

I have a Harman Kardon 75+ quad receiver. I've got a turntable and a Panasonic RS-845 quad 8-track tape player hooked up to it. The amplifier has a mode selection switch which includes the options mono, stereo, SQ matrix 1, SQ matrix 2, enhanced stereo 4-ch discrete. I'm basing this discussion on the Columbia SQ lp and quad 8-track tape version of the Broadway revival cast recording of "No, No, Nanette" (1971). When I listen to the 8-track tape using the 4-ch discrete mode, the mix is fantastic. On numbers with the chorus, the voices come from the back channels and don't exist on the front channels. The tap dancing goes from front left to front right to rear right to rear left. It's incredible to listen to. When I listen to the SQ lp using either one of the SQ matrix modes, the back channels seem to exist only to provide a kind of theatre sound effect. There isn't any remarkable front/back separation. I'm new to quad. I've always been interested in it but never had the time or money. I decided to start out with quad 8-track tapes - but the selection is very limited and, many times, lp tracks were left off for space reasons. In the last few weeks, I decided to try out a few lps - starting with SQ lps because it's less complicated to get going with. I know this is old technology, but does anyone out there have any words of wisdom regarding whether or not SQ lps did not produce the same effects as their quad 8-track tape counterparts? Could it be that something is wrong with the SQ matrix functionality of my receiver?

Many albums were, in fact, mixed differently for the SQ and discrete versions. The SQ system was flawed from the beginning, from the parameters chosen for the matrix. SQ decoders of the early era were not especially able to offer decent separation from front to back. It was only about 3dB, center front to center back. SQ emphasized lateral separation (left to right), so sophisticated logic circuitry was needed to enhance front to back separation. Unfortunately, these logic systems had audible artifacts; a "pumping" effect. I had one of the earliest SQ decoders, Sony's SQD-1000. That pumping was very obvious. As for the SQ functionality of your receiver, I'd guess it's working as designed. Later decoders offered better separation with less pumping. The best SQ decoding I've personally heard is with the Surround Master v.2 from Involve Audio. You can read a lot about it in the forums here on QQ. It will decode SQ with excellent separation and no pumping, and will do the same with QS and EV. It also does a remarkable job of creating quad from stereo.

 

[Eggplant]

There are also some liberties taken by the recording companies in that a really discrete mix could not be properly encoded for SQ LPs, so they used a modified mix.

I’ve never heard of this. Are you saying Columbia watered down a quad mix before creating the SQ master? It is true that remix engineers avoided certain no-no’s — center-back imagining or placing a source in all four channels being two well-known ones. Several early discs including Art Garfunkel’s Angel Claire violated the first restriction yet I don’t know of any that ignored the second. Perhaps we can thank that second restriction for insuring that most Columbia quads had lead vocals up front and center where they belong instead of the awful and wrong vocal X 4 (or 5) still common today.

But those restrictions were to avoid anomalies when playing in stereo or mono (center back vocals disappeared in mono, as they did with QS). If we’re talking quad decoding, there were plenty of things done in SQ that just flat didn’t work — meaning they did not decode well into the original soundstage. Most of these involved stereo fields, as in a grand piano across the fronts and a chorale spanning the rears. This very setup is found on Silent Eyes, the final track on Still Crazy After All These Years. There is no decoder that can handle this — the result is always a big gooey mishmash, and an unstable one. SQ simply didn’t work for multiple stereo sources and never would, which is what inventor Benjamin Bauer tried to tell Columbia but it fell on deaf, overly enthusiastic ears.

SQ works for ping-pong mixes with mono sources and rear ambience, but that’s about it.

 

[MidiMagic]

I’ve never heard of this. Are you saying Columbia watered down a quad mix before creating the SQ master? It is true that remix engineers avoided certain no-no’s — center-back imagining or placing a source in all four channels being two well-known ones. Several early discs including Art Garfunkel’s Angel Claire violated the first restriction yet I don’t of any that ignored the second.
...
SQ works for ping-pong mixes with mono sources and rear ambience, but that’s about it.

Placing a source equally in all four channels is totally impossible to do in SQ (unless no logic and no blend is used - then center front drives all channels equally).

Center-back imaging was avoided to prevent parts from disappearing when the matrix recording was played on a mono radio.

Center back imaging disappears in mono playback in all matrix systems except BMX, UD4, H, and UHJ.

SQ is not nearly as good for back ambience as QS and EV are. The separation is much lower. But ambience is good to place at center back.

There were several different methods used to make both matrixed records and Q8 tapes from the same master tape:

1. Make the discrete tape, then encode it with the 4-corners encoder.

2. Make the matrix recording, then decode it to the Q8 tape.

3. Make two different mixes, one for Q8 and one for the matrixed record.

4. If the stereo mix made a good image in matrix, they used that.

5. Some bootleggers just stick quadraphonic labels on stereo recordings.

I would bet that the SQ 1 is a 4-corners decoder and SQ 2 is a 10/40 blend decoder.

 

[sjcorne]

An example would be B,S&T's Greatest Hits. Now, I've never heard the SQ LP, but frankly, I don't think I need to. If we compare the mix on the Q8 versus the unreleased mix on one of the Mike Robin Reels.... the mixing and editing is almost identical. However, it seems that on the Q8, the separation has been cranked down so that if a Horn is supposed to play out of back left, you're still getting horn in front left at about 30% volume. It's like they were blending things back together after the main mix had been done. Yet on the Robin Reel, that horn is plainly in back left and nowhere else.

The SQ's of BS&T's self-titled second album and the greatest hits are among the worst I've heard. There's barely any channel separation. To me, this suggests they used the 'watered down' quad masters rather than the fully discrete versions on the Mike Robin Reels. Audio Fidelity's SACD of the second album also used the 'blended' version of the quad mix with the single version of "Spinning Wheel".

The mix on the Robin Reel would be perfect if it wasn't missing one of the brass parts during the chorus of You've Made Me So Very Happy".

SQ simply didn’t work for multiple stereo sources and never would, which is what inventor Benjamin Bauer tried to tell Columbia but it fell on deaf, overly enthusiastic ears.

I don't think it even does a single stereo source well. Whenever the drums are supposed to be panned in stereo across the front channels only, it sounds like they emanate from all around the listener in SQ.

 

[fredblue]

i'm fascinated by the suggestion because, setting to one side the "BS&T Greatest Hits Q8 that plays like decoded SQ" anomaly already cited, i've yet to encounter a CBS SQ Quad mix of Rock/Pop/Soul material that when decoded through the Surround Master didn't appear to pan out to the same cardinal positions as the discrete SACD and/or Q8 presentation.

are there CBS (and associated label) mixes where elements bleed into more than one channel in decoded SQ fashion that are more hard-panned in one channel in discrete form? yes but as mixes go the SQ and discrete presentations seem to resemble one another.

even taking that Garfunkel "Angel Clare" mix with instances of 'Centre Back SQ no-no' positioning as an example, the Surround Master acquits itself well with the SQ compared to the Q8 and SACD.

intriguing...

edit: oh yeah and thanks sjcorne
for the reminder about that BS&T S/T.. kinda bored of that album in Quad nowadays tbh and would rather spin the Quads of "Mirror Image" and "New City" any day of the week

 

[Marcsten]

You have to remember that the SQ decoders of that era were notoriously poor. They did not have the technology to properly decode the encoded albums until later models of decoders arrived about the time quad was disappearing. Today we have decoders, like the "Surround Master 2", and late era vintage decoders like the "Tate II" that will do a much better job on the SQ than the primitive decoder in your receiver.

The Q8 holds the real quad master, but at the price of additional wow, flutter and hiss. There are also some liberties taken by the recording companies in that a really discrete mix could not be properly encoded for SQ LPs, so they used a modified mix. You can hear this if you have a Q8 and an SQ LP with a modern decoder, then play a title like Paul Simon's "Still Crazy After All These Years", or others. There are just some things that Matrix encoding and decoding just can't do.

Be glad you have both!

All correct, as always. But also remember that BenBauer and the CBS folks who developed the SQ system required their (CBS group) releases to be exactly the same as the stereo versions when played back in stereo. This means that if the stereo mix has something in only the left channel, the SQ mix was required to locate the sound in the front left channel and so on. Where they had latitude was where the stereo mix had center cannel info. SQ would generally speaking send the rear channel info to the middle somewhere in stereo, so that material could be isolated in a back channel. All the considerations meant that the CBS SQs sometimes had less exciting rear channel info. Of course, if it was not a CBS SQ (English releases like Pink Floyd, etc.) they did not stick to this dogged compatability so far as I understand it. But regardless, the decoder in the HK isn't going to really give you any separation to write home about (probably 3db) so you won't be able to hear the difference.

 

[Marcsten]

The SQ's of BS&T's self-titled second album and the greatest hits are among the worst I've heard. There's barely any channel separation. To me, this suggests they used the 'watered down' quad masters rather than the fully discrete versions on the Mike Robin Reels. Audio Fidelity's SACD of the second album also used the 'blended' version of the quad mix with the single version of "Spinning Wheel".

The mix on the Robin Reel would be perfect if it wasn't missing one of the brass parts during the chorus of You've Made Me So Very Happy".



I don't think it even does a single stereo source well. Whenever the drums are supposed to be panned in stereo across the front channels only, it sounds like they emanate from all around the listener in SQ.

I have the BS and T greatest hits on SQ and also an Audionics Tate. I find the album inconsistent. Some songs have decent mixes while others suck. I always chalked it up to the fact that as a sampler album it had different engineers on different mixes. The self titled album, as I recall has only one or two SQ tracks on it. The rest are actually just stereo, which is why I never bothered to get it and just played the greatest hits.

 

[sjcorne]

The self titled album, as I recall has only one or two SQ tracks on it. The rest are actually just stereo, which is why I never bothered to get it and just played the greatest hits.

The entire album was definitely remixed for quad.
https://www.discogs.com/Blood-Sweat-And-Tears-Blood-Sweat-And-Tears/release/4140098

 

[fredblue]

hmm.. feels like we're getting fixated on a couple of stray BS&T outliers.... the overwhelming evidence doesn't support the idea that CBS did two Quad mixes as a matter of course, the discrete master mix and another mix doctored to cope with the SQ system's foibles/shortcomings...

 

[Marcsten]

The entire album was definitely remixed for quad.
https://www.discogs.com/Blood-Sweat-And-Tears-Blood-Sweat-And-Tears/release/4140098

As I say, I don't own it, so I am referring to the Larry Cliffton Quad Incorporated manual which I just dug out of my junk, I mean, 'archives" and it reflects the the only song actually encoded into SQ on the self title album is Spinning Wheel.

hmm.. feels like we're getting fixated on a couple of stray BS&T outliers.... the overwhelming evidence doesn't support the idea that CBS did two Quad mixes as a matter of course, the discrete master mix and another mix doctored to cope with the SQ system's foibles/shortcomings...

I always assumed they were the same. The ones where I owned both sounded the same, but that was a pretty small sample. I think the only thing they did to help out the logic circuitry was to boost the midrange compared to the stereo LP. And although they may have adjusted the mix to work better with the decoders, I always thought that the Q8 version was the same.

 

[fredblue]

afaik PIR's engineers stopped caring pretty sharpish about how their mixes were going to pan out when put through the SQ encode>->decode process and instead mixed them as to how they were going to sound in discrete form first and foremost.. although from what i can gather amongst other things provision was made for SQ considerations such as a "Centre" position (i.e., "Room Centre" by doing diagonal pans since SQ couldn't cope with the same info in all 4 channels at once) and they also addressed the side walls/phantom side thing that's been discussed several times here lately.

 

[steelydave]

i'm fascinated by the suggestion because, setting to one side the "BS&T Greatest Hits Q8 that plays like decoded SQ" anomaly already cited, i've yet to encounter a CBS SQ Quad mix of Rock/Pop/Soul material that when decoded through the Surround Master didn't appear to pan out to the same cardinal positions as the discrete SACD and/or Q8 presentation.

are there CBS (and associated label) mixes where elements bleed into more than one channel in decoded SQ fashion that are more hard-panned in one channel in discrete form? yes but as mixes go the SQ and discrete presentations seem to resemble one another.

even taking that Garfunkel "Angel Clare" mix with instances of 'Centre Back SQ no-no' positioning as an example, the Surround Master acquits itself well with the SQ compared to the Q8 and SACD.

intriguing...

edit: oh yeah and thanks sjcorne
for the reminder about that BS&T S/T.. kinda bored of that album in Quad nowadays tbh and would rather spin the Quads of "Mirror Image" and "New City" any day of the week

This is absolutely correct - I don't think there's a single CBS title where they mixed an album twice, once for SQ and once for discrete. There was no mixing desk automation back then, at least not at CBS's NY quad studio, so doing two mixes would've required re-doing everything by hand (and by ear) twice. Doing one quad mix was difficult, expensive, and time consuming enough for a format that wasn't selling much, there was no way they were doing it twice!

Also, if they were going to do a second mix that was only for discrete sources (Q8, QR, whatever) why do all the CBS Q8s exhibit the telltale signs of SQ mixing, like 4-corner discrete separation, nothing mixed in all four speakers, no front instrument reverbs in the rears, etc? Surely if they were unencumbered by those rules they could've been more creative with their mixes, like the RCA engineers were at the time. I know for a fact that all the PIR quads done at Sigma Sound were mixed to 4 track discrete 1/2" 15ips tape, and from that 4 track tape, a 2 channel SQ-encoded master was created - I believe CBS New York even eliminated a step in that process, running simultaneous 4 track discrete tapes and a 4-channel mixing bus output into an SQ encoder and then on to a 2 channel SQ encoded tape, the thinking being that less tape generations means few phase issues. In any event the quad mix is the quad mix is the quad mix - any differences are just anomalies introduced by the terrible SQ encode/decode process - look no further than some of the D-V releases of previously SQ-only releases like the two O'Jays albums, where the decoded SQ version at times barely resembles what's on the discrete tape.

As for the BS&T self-titled album, there are actually two Robin reels of the two different mixes, and we know from that Robin Reel listing that a lot of early CBS quads had multiple mixes - Laura Nyro Eli had 3 (if you count the one which AF released on SACD, which is different than the other two) and Johnny Winter's self-titled album also had two. It's been a while since I listened to the two BS&T self-titled quad mixs, but my recollection is that the released mix has more differences in it than just the couple that @sjcorne mentioned - I don't feel like it was simply a "blended" version of the unreleased mix. It's also worth noting that the BS&T self-titled album wasn't released in quad until the very end of 1973, despite having been mixed much earlier (all the Robin reels were from albums that were released in early 1972) so maybe they knew it didn't decode well in SQ, or it was mixed before CBS had settled on SQ, and they felt that the circa '73/'74 decoders were more capable of handling the mix, who knows...or maybe the SQ encoding was done with a crappy SQ encoded circa 1970 or 1971 - garbage in, garbage out, as they say.

 

[Eggplant]

the telltale signs of SQ mixing, like ... no front instrument reverbs in the rears

Huh?
Why is that a problem?
There are definitely CBS SQ albums that have stereo rear reverb from mono front vocals — Redbone’s Message From a Drum being one.

If you’re right that would explain the curious lack of any trace of the fronts in the rears on so many CBS Q8’s. As a youngster I found this ultra-discreteness cool — it was fun to pick out isolated instruments. Nowadays I just find it weird.

WEA mixes, of course, had tons of rear ambience AND the dreaded vocal X 4 (as found on America’s two quads). Over at Arista, all they did was vocal X 4 AFAIK. Amazing to think this was all due to SQ vs CD-4 and not artistic choice. At least one WEA quad, Seals & Croft’s I’ll Play For You has thunderous reverb restricted — for no good reason — to the fronts.

Now I really officially hate SQ.

 

[Marcsten]

Huh?
Why is that a problem?
There are definitely CBS SQ albums that have stereo rear reverb from mono front vocals — Redbone’s Message From a Drum being one.

If you’re right that would explain the curious lack of any trace of the fronts in the rears on so many CBS Q8’s. As a youngster I found this ultra-discreteness cool — it was fun to pick out isolated instruments. Nowadays I just find it weird.

WEA mixes, of course, had tons of rear ambience AND the dreaded vocal X 4 (as found on America’s two quads). Over at Arista, all they did was vocal X 4 AFAIK. Amazing to think this was all due to SQ vs CD-4 and not artistic choice. At least one WEA quad, Seals & Croft’s I’ll Play For You has thunderous reverb restricted — for no good reason — to the fronts.

Now I really officially hate SQ.

Play the Burton Cummings SQ. The rears on both the SQ a nether Q8 are full of heavy duty reverb on many songs. 'Scare' the opening track is really heavy,