So, we are all operating 1974 vintage or thereabouts equipment. It performs quite well.
A recapping and a tuneup will bring it up to par.
Up to par for CONSUMER tolerances yes, especially after having modern transistors and resistors and etc installed. Not for commercial or manufacturing tolerances. Remember the tolerances for consumer equipment and for playback are a LOT looser than those for commercial equipment or manufacturing.
Remember that they could spend much more on the modulator technology. there was no need to keep the unit cost down as with the demodulators.
It's like the Wonka-Vision Wonka Bar. So OK commercial and/or manufacturing gear don't have to be 100 times the size, just 100 times the quality so an ordinary person on the other end can experience it with ordinary (not even superior) quality (for that it would have to be 1000 times better).
I don't imagine there is a whole lot of difference between 2/3 and 3/4 cutting speeds.
Why not just run at the old speeds?
As everyone here knows, the very very first CD-4 titles produced were mastered at 11.11 RPM for playback at 33 because the cutterhead capabilities of the time were insufficient to be able to sustain a long constant reference tone without melting the bobbins in each channel of the cutterhead from overheating.
And if you find some of those early Princeton Labs test LP's, the sound quality was so terrible upon playback vs a 4-track tape modulated live to CD-4 and then demodulated live by the demodulaters that they almost didn't get the funding to proceed with re-tooling the cutterheads and cutter systems to work with 16 RPM.
The simple explanation for non-engineers of what they did for the ``live'' modulating and demodulating was basically tweak the 38 KHz center frequency subcarrier of two low-power FM Stereo transmitters down to 30 KHz, sent the front channels over the basebands of each one and then send the rear channels over the two subcarriers and reverse the process when connected to the demodulator.
Of course the
actual explanation is, they used
three subcarriers, one at 38 KHz for the stereo mixdown, one at 67 KHz, one at 76 KHz and one at 92 KHz, as they tried to introduce for FM Discrete Quadracasting. As we all know 19 KHz is only for the pilot tone that lights up the FM Stereo light on a receiver, 67 KHz is for Radio Reading Services for the Blind, 76 KHz is for subscription commercial music services and 92KHz is for hospitals, cable stations and other such installations where the transmission channel might be on 67 KHz for the transmitting station and the reverse direction might be 92 KHz on a second station transmitting from the receiving end for extended point-to-point or so-called narrowcasting applications before the advent of inexpensive data services.
So anyway back to the problems with Half Speed Mastering on Lacquer. 16 RPM cutting for playback at 33 has an ENORMOUS amount of bass-contour effect, not to mention, even tho the top end has the benefit of all that headroom, you still end up with a somewhat pinched and lifeless sound compared to the real-time mastering when you don't slam the cutterhead to its' limits and beyond like rock and roll is fond of doing.
The same effect can be heard when mastering, say a 78 RPM record at 45 RPM. Even with the correct size 2.7 - 3.0 mil stylus and a good cartridge, if a 78 is mastered at 45 for recording into the computer, all the natural life and ambiance of the original recording will be gone in the transfer.
Try it sometime. Of course, if you can find the same title mastered in both Stereo as well as CD-4 in addition to being mastered in a 45RPM SuperSonic audiophile edition, then that would be best.
But if you don't have that, and you don't have the 78 stylus and want to hear a slightly less pronounced effect of the same thing, play one of those audiophile SuperSonic classical music 45RPM discs from the 70's back at 33 and record into Adobe Audition using my tutorial here:
http://bsnpubs.websitetoolbox.com/post?id=3024365 and then play the same record back at its' original 45RPM speed and record again and then compare the two.
Now do the same with some famous performance of a classical work that was recorded in both Stereo as well as CD-4. Record the stereo version at 33 into AA.3.0 or whatever using the same CD-4 stylus as a Quadradisc and then do the same to the CD-4 WITHOUT demodulating (even switching on the FM multiplex filter so you don't get carrier tone interference to make the test more equal) and you can PLAINLY hear what we're talking about.
Or just watch this video for a considerably more crude example:
http://www.youtube.com/watch?v=oT8gmiCbuTI. By the number of raves on the post about the version transferred at 33 over the version transferred at 78, the number of deaf or partially deaf non-engineering listeners in the general public can be ascertained (looks over top of glasses).
Same as the transferred-at-78 version has more life, you'll notice conversely on your own tests that the real-time-mastered LP version has PLENTY of air and space where the CD-4 version does not, same as the 45 version has more air and space in real time than the 33 version, but NOT if you play the 45 back at 33 and resample digitally to 45 in the computer.
That's like some of the old early RCA Dynagroove cutting tests in the `60s where they were trying to save studio time by cutting stereo at 45 for playback at 33, taking the tape machines from 15 IPS up to 20-1/4 IPS, or worse yet, trying to cut a disc at 66-2/3 RPM. Actually it wasn't really 66-2/3, it was 56.35 since all they did was use what the ``78'' became when the ``33'' was set for 24, just like the Mattel doll people used 105, because that's what the 78 becomes when the 33 is set to 45.
Anyway, if you don't have one of these and want to hear what it sounded like, just take one of the old 16 RPM music discs cut in the 60's (either the 7-inch or the 12-inch), or see if that Lanny Kennedy is still a member here and see if he will sell his granddad's 16 RPM Will Kennedy Dancetime Orchestra records that has a 33 RPM counterpart, or take one of the 16 RPM Prestige Jazz series of double-albums on one disc, play it at it's normal 16 RPM playback speed and then take the same 33 LP album from the same period and play it back at it's native 33 RPM and see how you like it.
Or do what everybody else does with a 16 RPM title that has no 33 counterpart: Play it at 33, record it with Dolby NR onto a reel to reel deck at 15 IPS (or 7-1/2 if you don't have 15, or onto a Chrome or Metal cassette on a deck with a High Speed Dubbing setting if you don't have that) and then play the resulting tape back at it's normal speed with the Dolby out and the EQ setting to Normal instead of Chrome.
But watch out. Make sure you have the correct size playback stylus. The Prestige Jazz and most other mono 16-RPM 12-inch LPs from the late 50's and early 60's are cut with a one-mil mono stylus the same as most of the 7-inch 16's with the 45-size hole. The Will Kennedy and the other Stereo 16 RPM's were all cut with a normal 0.7 mil Stereo stylus, so no worries there.
Here's where it gets dicey: The 12-inch department-store 16 RPM's like Allied Artists and such are all recorded with a 0.5 HALF-MIL stylus, smaller than the Stereo stylus and commonly found on talking book players for the blind. Same holds true for the the seven-inch 16-RPM flexidiscs made by Auto-Com as well as the nine-inch Seeburg BMS 1000 discs with the 2-inch hole
http://www.youtube.com/watch?v=bgo6u-qEECg. The guy in this video is the only one on YouTube who is not destroying his Seeburg disc on the first pass from playing it with a stylus that is too big for the groove.
The ones you REALLY have to watch out for though are the Highway Hi Fi records that went into `56 Chryslers. They have a very small 2 inch label on a 7-inch disc with a small hole, and were made from a 90 gram biscuit, same as for a normal non-Dynaflex ten-inch vinyl, but the grooves are a QUARTER-MIL, 2/3rds smaller than a stereo LP stylus and 75% smaller than a mono stylus. Playing a Highway Hi-Fi record with anything other than it's own stylus or a remanufactured one for a modern cartridge will be tantamount to playing an LP with a 78 stylus, destroying the disc on the first pass.
But all those have LP counterparts except for the demonstration discs with the announcers on them, and all those have already been transferred using compatible equipment, so there should be no need for transferring a Highway Hi-Fi record. If you find one, frame it and hang it up as art. Or Earl or Bill or...
But anyway, playing back a disc at slower than its' original speed will still sound pinched and lifeless the same as playing back a disc faster than the recorded speed because slow speed will dump the bass into the mud and the 2x speed of 66 over 33 or the 1.35 speed of 45 over 33 makes all the high frequencies move out of the range of the cutterhead and tape playback equipment.
The only reason it works the other way around, is that 11-1/4 IPS is 75% of 15 IPS (normal mastering tape speed) and 33-1/3 RPM is 74.0740740...% of 45, bringing the bass up out of the mud and leaving the treble in at least the same range of treble, just not as high, ergo 3/4 speed mastering is the magic threshhold. That, plus the difference between 74.074074074.... and 75, like the difference between the 45.00 RPM in the U.S. and the 45.45 RPM at which European discs are cut is negligible.
The half-speed-mastered version sucks all the air and space out, along with most of the life and vibrancy carried thereby. It's the same with the standard stereo Half-Speed-Mastered LPs by Columbia or whoever. And that you can't do anything about, it's endemic to the process. On rock-and-roll, there's not much air and space there to begin with, so, say CD-4 Black Sabbath for example, you'd probably never be able to tell the difference.
Now, the inbetween version of that was the 2/3rds speed cutting, more or less 22 1/2 RPM for playback at 33. The first of these was of course Japanese since they had half-speed-mastering lathes for all their CD-4 45 RPM calibration discs with various quadraphonic Muzak titles on the back. So, they improved their cutterhead and modulation electronics once again to be able to cut at roughly 2/3rds speed.
Guess what? Most of the air and space was back, most of the bass contour effect was gone and most of the pinched effect in the treble was gone. Why did it not catch on? Because they didn't do it until the end of 1976, by Spring of 1977 with a few exceptions, most of the last CD-4 titles had already been cut, places like JVC Hollywood were going back to stereo cutting and by the end of 1977 a year later, quad in ALL formats was pretty much a done thing.
Now the 3/4 speed cutting i.e cutting at 25 RPM for playback at 33 would reduce those artifacts even further, but cutting at 33 for playback at 45 would eliminate those artifacts altogether, and then 33 RPM lathes are a lot easier to come by than 25 RPM, and then the RIAA recording curve could more easily have its' bands lowered to work with 3/4 speed cutting a lot easier than trying to adapt a lathe for 25 RPM besides lowering the RIAA curve, not to mention the better sonic response recording at 33.
Cutting onto DMM would be the only way to go as well, because now, at 45 RPM playback speed, your original 23 minute LP lacquer side at 33-RPM now has to be 23 minutes at 45 or 30 minutes at 33. The only way you'd be able to master that with any kind of decent groove depth and decent volume is on a DMM, since not only does it cut longer sides than a lacquer at the same volume, but it also has a greater range in the treble. What ends up as a brittle-sounding top end in real-time DMM cutting maybe just what the CD-4 process needs at 3/4 speed cutting,at 33 for playback at 45 we'll have to see.
You never know. They could have had exactly what they needed if they'd have waited a few years for the SX-84 DMM heads to come out for the VMS-80 or VMS-82 Neumann lathes for BOTH matrix as WELL as discrete-disc cutting. By then they could have had 5.1 or even 7.1 utilizing both matrix and discrete technologies at the same time like the 1969 test disc uses SQ laid over QS laid over EV laid over DY to create some crazy kind of 12-channel or 14-channel matrix presentation.
I have the test LP and the film to which it syncs around here someplace. It's been years since I played it for anybody, but I imagine even if somebody HAD a matrix decoder that could retrieve 12 different kinds of phase differences all at the same time, even with 21st century digital steering logic, I doubt very seriously an all-matrix 12- or 14-channel program would have got off the ground.
But, I imagine, using a simple 90-plus and 90-minus matrix, i.e. splitting the 180-degree out-of-phase between the two channels and then introducing the signal from a retooled London Box originally designed by Ben Bauer to give a discrete rear-center and then using the CD-4 (or in this case UD-4) to isolate certain tracks or instruments, I think once we got all the kinks worked out it would work fine.
After all, if it's the difference between not recording CD-4 and doing so. If the project was successful enough to warrant it, you could then design a new modulator.
Only one problem behind that, though, back to the Willy Wonka bar on the Wonka-Vision example.
The only reason we'd even GET funding in the FIRST place for even TRYING is if it would:
A) eliminate the bass contour effect,
B) restore the air and space, and
C) take away the pinched effect on the treble that plagued the original recordings.
If a huge company like RCA using people at the influential Princeton Labs came This Close to not getting CD-4 approved in the first place because they could only cut at eleven RPM, (like they came This Close to not getting CED Videodiscs approved because they could only cut at 45 RPM [1/10th speed]) and came This Close to got getting the green light after they perfected CD-4 cutting for 16 RPM, a year later, then we as a little bitty research group with no ties to like the Government's Lawrence Berkeley Labs or any huge corporation like RCA with its' own Princeton research lab and much of its' own funding, well, we'll have to do a LOT better than RCA or the LBL research teams JUST to get testing approved, nevermind going ahead with manufacturing.
The Dorren demodulator is being built to work (the best) on the (original) CD-4 standard. It's got to play what was recorded with that very modulator. That standard has not changed since 1974. If Lou decides to build a new modulator, it must conform to the same standards.
Same standards, yes, same parameters, yes. Same design, no. There's a lot better ways today of both creating as well as retrieving CD-4 signals, just as in the modern day there are a lot better ways of retrieving magnetically recorded signals from the legacy media formats which contain them.
The main difference between Lou's design and the old school demods is the FM limiter stages and the sharper (digital?) filtering.
Which in the opinion of this as well as several other engineers is a major, major improvement. Utilizing the design of the original CD-4 process notwithstanding, this is up there with the conversion from 11 RPM cutting and 16 RPM cutting, or even better, from 16 RPM cutting to
22 1/2 RPM cutting. So, it's NOT the same animal.
But there is nothing wrong with 1974 analog electronics
technology, on the contrary, it was quite good.
Well, we also want to have as much of it be valve-design oriented as possible. If there could be a such thing as a valve-design CD-4 modulator---or even demodulator that was only digital in the places it ABSOLUTELY needed to be, then that would even be better.
The reason we're trying for that is we're going off of the valve-design DMM cutting
test discs that never got off the ground.
When RCA, Mercury and Columbia were mastering their Living Stereo Living Presence and Legacy Masterworks for their Anniversary Reissue audiophile CD's, done in the late 80's and early 90's, they were going back to the original 3-track half-inch work parts (unedited session masters) for the first time since the original stereo LP mixdowns were made in the `50's and `60's.
They'd align each one up individually on the playback deck in order to transfer each to digital so that it could be tuned up in a new post-production session and prepared for the final CD release, but inbetween time,they cut DMM ref copies of each tape they aligned through what, according the slip inside the test pressings we have, were identified as Gotham Audio valve amplifiers. Even compared to the CD's that came out, the DMM test pressings cut with a valve amplifier sound
F A B U L O U S.
So we can imagine how terrific a new CD-4 record would sound if
A) it was mastered at 25 RPM for playback at 33 or better yet 33 for 45
B) sent through a newly redesigned Quadulator (hybrid valve/digital)
C) mastered through a valve amplifier
D) onto DMM and
E) pressed onto 180G grainless red or blue dye
And if you wanted to know why they invented DMM to begin with it wasn't for better fidelity.
RCA invented DMM for cutting CED videodiscs with way back in 1979 and then gave it to Neumann/Teldec in trade for some circuitry they had invented and couldn't figure out what to do with that RCA thought would work well with some CED improvements before LaserDisc took over the videodisc market.
Then Telcec and Neumann perfected DMM and then tried to SELL it back to RCA for something they GAVE AWAY, so if you run into some legacy RCA mastering engineers who have a decidedly negative opinion about DMM and wondered why they never got their own DMM lathe and farmed everything out, now you know.
But really, DMM was invented because skinny engineers were becoming the norm and they didn't have the body insulation we did (in the form of 3X shirts and 54 waist pants or whatever) for cutting 10 and 12 hours a day in a 55-degree room to preserve the lacquers from groove bounce-back before we could get them into refrigerators. Guys wanted heat in the mastering room and the only way to do that was change the cold and hard from the room temp to the cold copper and the hard diamonds with which to cut.
Go to the library and look up photos of old mastering engineers from the bygone eras.
There's enough spare tires in the room to outfit a fleet of 18-wheelers for a lifetime.
But on the other hand, I don't doubt that a new design
could probably tighten things up a bit.
Absolutely. Which is exactly the point.
If germanium transistors were used, all bets are off.
But those were generally out of service by then.
So, once again, professional and manufacturing gear has to be a number of degrees better than its' consumer counterpart, therefore whoever redesigns the new modulator would have to use germanium transistors then for the very reason you mentioned.
So maybe when all is said and done, we might be able to cut CD-4 DMM at 33 for playback at 45 with a brand new modulator with a brand new design and blow the lid off the audiophile vinyl market.
Your homework for next time is Chapters 8, 11 and 13, pages 135-151 in your workbook, and show up at 5 to setup for the sessions Friday and Saturday night with a downbeat promptly at 8. Standard four-hour Union Scale sessions, musicians get to go home at 12, depending on how fast you strike, you MIGHT get out of here by the time the sun comes up.
Failure to show up will GUARANTEE you at LEAST a one-grade penalty on the next exam and maybe more depending on how the session went and what kind of mood I'm in afterward when I grade everybody's performance.
Class dismissed. See you Friday night at 5 (looks over top of glasses).