jaybird said:
I don't think it's the depth of the groove that matters; rather, the amount of out-of-phase information in the mix.
A) The latter is generally the cause of the former BUT
B) Mastering engineers can choose to enhance leave the same or suppress the vertical information by their cut.
So that explains as discussed above why shallow-cut versions of the exact same stereo mixes - and some CDs decode very flat and two-dimensional compared to their deeper-grooved counterparts? I perform the exact same playback AND RECORDING demonstrations discussed above and more every year for the junior college sound engineering kids I teach.
Every year they accuse me of playback manipulation behind the scenes til I make `em play the same records in the front of the room themselves without touching anything other than the disc and the tonearm. The same holds true for cutting on the lacquer. As everybody knows, not only can you set the groove depth of a cut on any stereo lathe/head, you can also adjust the ratio between lateral and vertical - especially on the Fairchild lateral-and-vertical lathe we have which goes through the same L+R/L-R matrix to convert to 45/45 stereo.
I've cut pieces on there that do MARVELOUS upon matrix playback, and the same exact master tape in the same exact condition cut on the e.g. stereo 45/45 Westrex head on a Scully lathe with no other differences whatsoever are back to sounding flat and boxy.
The reverse has proven to be true as well. Some cuts on the Scully/Westrex with the groove depth set to normal have decoded alright - nothing spectacular - but then you take the depth up a little bit and leave everything else alone and WOW.
The aforementioned Patsy Cline's Greatest Hits (1967 - recorded between 1959-63) at the top of this page is the perfect example for the reasons stated - and the results are visible in the disc itself. Perform the playback test as I indicated it above. Start off with a normal 140-gram 1967 Decca pressing, then go to the 90-gram MaxiFlex series vinyl which sound very flat and boxy when decoded - and then play the aforementioned 45 RPM Super-Sonic 2-disc edition.
If you look at the pictures of the latter, you can see how much more vertical information is there compared to all the other pressings - and when you play the disc, it shows up in how well the decoder works.
Keeping with the Patsy Cline theme - the conventional MCA masterings/pressings of the `Sweet Dreams' soundtrack - albeit including the fact that some numbers have a new backing track added in to the classic vocal performance - the difference is striking from the normal MCA commercial release cut on the Neumann at Sterling Sound which has very shallow grooves and barely decodes at all and the RCA Record Club edition cut on the Scully at Columbia Nashville and pressed at RCA in Indianapolis which was cut deep as it would have been back in the 60s.
Exact same mix, one cut with a deeper groove and one cut with a shallower groove and no other difference. You perform the same test the same as I do every year in class and you see what you think.
jaybird said:
Matrix-encoding, for quad, utilizes phase shifting in precise amounts, applied by the encoder, according to the quad mix of the recording. Any quad effect you get from stereo records, CD's, etc. are purely at random. They weren't planned for.
No argument there. But like I said - there can be VAST differences in the type, quality and depth of these ``accidental surrounds'' depending on the cut, as I prove to the music engineering kids every year in class.
jaybird said:
Decoders for the QS/RM work best for extracting a surround effect from stereo recordings, I use my Surround Master, in the Involve 4.1 setting, for this, and the results are quite interesting. Some records deliver more of an effect than others, but more of the detail in the music is revealed, too.
Again - between the extraction just discussed, all Lt/Rt combination matrices all work off the same principal. If you work in a movie theatre in the classic days of Dolby Surround cinema, you know about the four trim-pot adjustments on the back for Left, Right, L+R and L-R.
Even though Dolby Surround in the 70s never had split surrounds/5.1 and was built to enhance center channel dialogue, you can take any Dolby Stereo test film (or any similarly mixed stereo music with or without the Dolby Surround matrix encoding) and turn the L-R trimpot all the way down.
You will still get perfectly fine 3-channel Left, Center and Right stereo, but the surround channel(s) will effectively be turned off or rendered very low by comparison. The reverse is also true. An operator can increase the trimpot value for L-R and leave everything else alone and all of a sudden theatre patrons are drowning in surround.
Some prints would even come in from other theatres in or out of the same chain with little white cards in suggesting a percentage setting for the surrounds trimpot (/, \, |, etc). Doing so would often increase attendance and therefore revenue at our theatre the same as they would in the bigger movie palaces.
We used to get in all kinds of trouble from theatre brass for running the normal stereo intermission music with normal amounts of L-R signal, and
A) piping it through the Dolby Surround decoders
B) cranking the L-R trimpot up to max in doing so
til we forgot to turn the trimpot back down again for the feature
After the third or fourth time getting caught doing that, we'd do the same thing as the guys on eBay are now trying to hawk their own upmixes by simply playing normal stereo material through a Dolby Surround or other matrix decoder the same as we did and turning up the L-R in the process.
The resulting 4-track inline 4-channel tapes most people can't tell the difference, since the ones we'd choose would have the most active surrounds compared to other music - regardless of maxing out the L-R decoder trimpot or not.
Discrete 4-track playback was easy enough in the theatre as every house had provisions for discrete 4-track magnetic surround that hadn't been used in 20 years - but the discrete 4-track amps were still there. So we'd wire it up to take the hots from the left and right for the center amp and blend the two rear channels we made back into one and pipe that through the surround amp.
Once 5.1 came around and we had split surrounds after that, we just left the center channel off and piped the 4 discrete tracks of our 4-channel inline tapes we made through the 4 discrete amps and let the public enjoy their intermission music in synthesized quad.
What the eBay guys are doing on the other thread is nothing different - except they are trying to SELL theirs as ``authentic record-label engineered quadraphonic-on-purpose'' with artist involvement and the whole nine yards when all they did was the same as we did and the same as the radio music library production houses did -
i) run the normal stereo feeds through one decoder or the next (sometimes simultaneously and choosing the best surround channels of all of them for the final 4-channel mix),
ii) max out the surround settings on each one
iii) record the result on 4-channel inline tape
and send them out to the radio stations which would then re-encode the matrix on their end prior to broadcasting.
40 years later I'm still amazed at the number of ``REAL'' i.e Drake Chenault etc produced in the fashion discussed above ``quadraphonic'' programming reels floating around that have nothing to do with
a) the real period-specific quadraphonic mix by the label if there was one
b) later versions that would be remixed from the multitracks by e.g. Dutton Vocalion UK etc.
So groove depth DOES make a difference and you'd be wise to try the above comparisons yourself before you judge. yes they are still all ``accidental'' surrounds, but that doesn't change the fact that different mastering and cutting techniques do affect the results coming out of the decoder just the same as different matrices.
So try it and see what you find out.