Matrixing

When I went off on my little Doctor Who ramble earlier I didn't get around to finishing up my earlier topic.  One of the interesting things people do - again, I mostly encounter this in the Zep and Dead boot communities - is matrixing.  This, I think, says interesting things particularly about the Dead tape community.  Back in the day, before widespread digital availability of everything, before the Internet Archive or Dick's Picks, the currency of the realm in Deadhead circles was "crispy" or "crunchy" soundboards.  It's not a nomenclature I understand.  I find it interesting the attributes attributed to sound, that what was desirable in those days was a sound that was hard and brittle, as opposed presumably to soft and yielding.  It tracked with the tendency, in that early digital era, for noise reduction, the sterile narrative of "progress".  Hiss was seen as inherently bad, and accordingly recordings were reprocessed to be dry, airless; space was interpreted as noise.  Now it is different.

The centennial release of the recordings of Robert Johnson restores the hiss, and this is seen by many, including myself, to be an improvement.  A similar change has occurred over time in the Dead community.  An understanding of the differences between line sound and sound as heard by the audience, the ambience of the venue, has spread with greater availability of recordings.  Certain sound configurations of the Dead, particularly their 1974 "Wall of Sound" getup, sound very different in audience captures than on soundboard recordings, on which the band's sound is not markedly different than it was in prior years.

Thesis and antithesis gives way to synthesis, in the best scholastic style.  One can superimpose an audience recording, for the space, onto a line recording, for the clarity of sound, and get some significant measure of both, just like you can take the fuzzy colour signal of an off-air Doctor Who recording and superimpose it onto the crisp lines of a B&W film print to get a crisp colour picture.

This also comes about in commercial recordings, in various forms.  I first encountered the concept through the discovery of some old Duke Ellington recordings from the early '30s.  The tracks were recorded into two mono recorders at the same time, but these recorders were set up in different places in the room, with different acoustics.  Recordings from both masters were commercially issued.  This happens a lot too - when distributing recordings there is sometimes a mix-up about which recording is the master, and so the wrong recording gets used.  You could write a very boring book about all the different versions of Bob Dylan's "Blonde on Blonde" that have been released.

This phenomenon actually accidentally redeemed one of Peter Bogdanovich's most notorious flops.  In 1975, Bogdanovich, who was a very successful director at the time, resolved to revive the Hollywood musical with a slate of Cole Porter songs.  Today, of course, we know that every time this is attempted it pretty much winds up a dismal and embarrassing failure, but at the time Bogdanovich may be excused for believing he had a shot, as the heyday of the musical wasn't that far behind.

There were a couple problems.  First was that his star was Burt Reynolds.  Since this was 1975, test audiences demanded more Burt Reynolds, which was not generally something that led to the improvement of the film.  Second was that a lot of people in Hollywood really personally hated Bogdanovich.  He has a reputation in many quarters of being a weasel and a snake.  I don't know the man personally, and I am making no personal judgement on him, merely observing that a lot of people did, in fact, very much want to see him fail.

So he did.  His response to this failure was to actually take out a full-page groveling apology in the trades, which doesn't necessarily do a lot to argue against his being sort of a sniveler that I can see, and then the film was buried.

The interesting thing is that a fairly anonymous film editor by the name of Jim Blakely decided that there was some value in the film Bogdanovich was trying to make, and so put together his own cut of the film based on Bogdanovich's script and first cut.  At some point he quietly swapped out this version for the master, and subsequently died.

In 2011, Netflix was trying to make a name for themselves in the industry, and they were doing it by acquiring as much content as cheaply as they could, which meant acquiring the streaming rights to content that nobody in their right mind would possibly want to see.  After this, the unthinkable happened: Somebody actually watched "At Long Last Love" and said "Hey, this is actually pretty good!"

People still hate Peter Bogdanovich, but I think they are OK with Jim Blakely.

Back to Duke Ellington.  What was discovered was that combining these two commercial mono recordings resulted in a stereo recreation of the room ambience - that's basic matrixing for you!  It's still a pretty heavy use of matrixing that I can tell - taking mono recordings and turning them into stereo.  Of course, it is possible to have more than two audio channels on a recording, but anything over two channels has been and possibly always will be the province of rich, overwhelmingly male audiophile nerds, the acceptable uses of technology determined by cultural considerations.  That's a rich topic for another time, probably.

The other example of matrixing recordings in order to create stereo is the case of Good Vibrations.  If one had to pick one song to be most identified with Brian Wilson and the Beach Boys, well, I'd pick Good Vibrations.  A towering accomplishment as well as a bit of a pyrrhic victory - Brian logged 75 hours in the studio recording the single, which isn't particularly practical for a song that's going to spend a week at number one and then be replaced by some song that probably took ten minutes to record and five minutes to write.

All of the Beach Boys' early recordings were in mono even after most other bands had adopted stereo.  A lot of this was a function of Brian's huge input into all aspects of the recording process.  Since his dad had beaten him deaf in one ear, stereo wasn't a huge priority for him, even as the record company was pressuring him to adopt stereo recording (early cover slicks for "Smile" declared it to be a stereo recording, which it almost certainly wouldn't have been).

Over time the reputation of the Beach Boys' '60s recordings grew, and in 1996 a stereo version of their landmark "Pet Sounds" album was released, newly mixed from the original master tapes.  This wasn't possible with "Good Vibrations", though, because at some point the complete vocal sessions for the single were lost.  Beach Boys nerds bemoan this loss particularly.  Three CDs worth of sessions for the backing tracks exist, and a stereo mix can be put together from this, but the vocals exist only on the released mono single.

Which means, in turn, that any stereo version, well, it's going to be "mixed" in a sense, but big chunks of the recording being taken from pre-mixed commercial recordings does make it an exercise in matrixing, even if these days it's extremely sophisticated and complex.  Here's the recording I go to when I want to hear a stereo version of Good Vibrations:

https://www.youtube.com/watch?v=rfX-5Ok5pzo

I honestly wouldn't really know the vocals no longer exist in stereo from listening to this.  It's quite convincing.