Monday, June 30, 2014

Sorting Out the Synclavier 4

This is a good news/bad news post. The good news is great, so let's start with that.

The good news is, John has reverse engineered some of the Synclavier cards, and he sent me a couple to test! Now that I have all of my voices sorted out, no problem. In fact, this sort of thing is big time fun for me.

Here are the test cards of the new SS1, which is the first card in a voice card set for 8 FM/Additive voices. On the right is a complete card, which is what I tested for this post, and on the left is a card with interchangeable chips, which I will test next month (I'm still not perfectly clear on a methodology for that yet).

Here are the old and new SS1's next to each other. The new SS1 is on top - vintage 2013 - and the old SS1 is on the bottom - vintage circa 1983 - so there is about thirty years between these cards in age! Long story short, the new card functioned perfectly, so that is great news. So John could listen for himself, I recorded two test tracks; a Reference with the original SS1, and a Test with the new one.

I just recorded the stereo outputs of the Synclavier directly into a Lexicon FW810s, and used a single stereo track in GarageBand to record it: The very simplest possible solution. There is no EQ, compression, reverb or anything; the tracks are totally dry. I rendered the tracks as uncompressed AIF files, so they are large, but they are also CD quality resolution (Which is better than the Synclavier!). Here they are:



My 56 year-old ears can't tell any difference. Oh, and sorry for no fadeout: These are test tracks.

So, this first test being successful sure is a good feeling! I've only ever been a virtuoso on one instrument, and that's the Synclavier, not the guitar, so it's awesome for me to see new life being breathed into it. This sequence is vintage 1986, by the way, from when I was a guitarist in a techno-rock band called B-Rock. I prepared it for a studio album that never happened.

Okay, now for the bad news part.

After many months of wrestling with a Mac interface and getting the voice cards sorted out, I decided to return to my original setup with a Pericom dumb terminal - basically a VT-100/640 with a bigger screen - and when it finally arrived - the last piece in the puzzle before I could start recording Fuga Electronica - it was damaged in shipping (Note the broken base). What a letdown. So, there are only two other Pericoms known (!) and I'll try with another. But I'll also go ahead and get a VT-100/640 as they are more common.

Saturday, May 31, 2014

Sorting Out the Synclavier 3

John finally got time enough to send me a replacement SS2 card, so I got the Synclavier back up to 32 stereo voices, and for the first time since I got the machine last August, it's working 100% correctly.

I used hockey pucks to raise the keyboard up so I can program it standing.

Here's the replacement SS2. It's vintage 80's.

Finally, 32 happy stereo FM/Additive voices.

For the first time in ten years, I can listen to all my old sequences.

Next month I'll get the Pericom terminal, and then I should be in beeswax for Fuga Electronica.

Wednesday, April 30, 2014

Book Review: Bach, Music in the Castle of Heaven, by John Eliot Gardiner

John Eliot Gardiner is a giant in the early music/period instruments world, and he's basically been active all of my life. He's one of those musical animals I don't understand, though; a conductor. So, I want to get this out of the way right up front: This is a monumental work, but it is not aimed at composers like me. In fact, I'm not sure who the target audience for this book is. Musicologists - another breed beyond my comprehension - perhaps. There wasn't anything for the theorist in me either, and I can't help but think that even an educated connoisseur would be completely baffled by this book.

He's English and I'm an American - US and Canadian dual-citizen, so I mean American in a broad sense - so there are the usual u-phillic/z-phobic English spelling fetishes to deal with, but barriers to this book go well beyond that. Since it's a book about music, he employs the curiously English crotchets and quavers terminology - something I refused to even learn because it's so patently ridiculous - and he has this vast classically-educated vocabulary, which I find tiresome, frankly. His approach seems to be, why use, "assistant" when I can whip out the, "amanuensis?" Since my writing and composing philosophies can both be reduced to three words - economy of expression - I don't like this at all. I found myself having to leave the well-ordered comfort of my wingback reading chair to shuffle off to the computer to look up some obscure word quite often, and I have a Master of Music degree!

Then there is the sheer length of the book - over five-hundred pages - so it took me over a month to slog through it. Seriously, I could only stand to read some several pages at a time. A large part of the problem is that the book is filled with subjective opining on Bach's music, and even though I'm a Lutheran, and so open to theological rationale, I didn't agree with much of it. Another huge, gaping maw is the utter absence of musical examples - I think there was only one - to back up any of his opinions with technical analysis.

You know what this book reminds me of? Ulysses by James Joyce. I know it's great, but I get very little out of it (Some of the details of Bach's day-to-day life were nice, but most of the book is describing the passions and the B Minor Mass). So, I guess I'd recommend this book to hearty souls, but your basic, b-flat musician would probably enjoy James R. Gaines' Evening in the Palace of Reason or Christoph Wolff's Bach: The Learned Musician much more, as I did. I couldn't put either of those down.

Bottom line is, for me, this book would have been twice as good if it was half as long and chock full of musical examples

Monday, March 31, 2014

Sorting Out the Synclavier 2

In the previous post, I identified voice card set four - voices 25-32 - as being bad (There was distortion and artifacts in the waveform). Well, I decided to take a break and enjoy my Synclavier as a 24 voice machine for a while, as I just wanted to play with it. I soon got frustrated with running out of voices though, so today I identified which specific card in that voice card set was causing the distortion and artifacts. This was just a matter of replacing the cards in voice card set 3 - voices 17-24 - with the ones from voice card set four, one at a time: When the distortion and artifacts returned, I'd have the culprit singled out.

As you can see by all the floppies everywhere, I have been hard at work recovering ancient sequences and timbres. Some of the sequences go back to my techno-rock band days in NYC from '84 to '88. It's almost spooky hearing those after all these years (Though I did have to mute the FX tracks to keep it within 24 voices). Some of these are very, very good, and I guess I'll release them at some point.

Here's the view from my couch: The Mac Mini is running a 65" Plasma TV, and that's the Lexicon FW810S mixer. I have a bluetooth keyboard and Magic Trackpad on the coffee table as remotes. Besides being my recording studio, his doubles as my home theater too, as I have a DVD Mac Mini and a BluRay player. The sound system is a Bryston 2B-LP Pro running Turbosound monitors. It's quite amazing sounding.

Below is the bad voice card set I identified last month. I just systematically swapped them into the voice card set for voices 17-24 until the problems returned. Left to right, the cards are called, SS1, SS2, SS3, SS4, and the stereo output double-card.

Here's a look at the upper bin again, and the card set for voices 17-24 is on the right. All I had to do was introduce the suspect cards one at a time until the troubles resurfaced. Needless to say, I've done this many times over the years... decades.

It turned out that it was the SS2 card that was bad. As soon as I introduced it, the distortion and artifacts returned in the exact same manifestation as when I had all four voice card sets installed.

I went ahead and tested the SS3 and SS4 while I was in there, but they were fine, so I didn't bother with the stereo output dual-card. Output problems and wavetable errors manifest in completely different ways.

Now, If I can get John to stop globetrotting long enough, I ought to be able to get a replacement SS2 from him, and then get the monitor issues sorted out (Yes, there is even more to do).

Monday, February 17, 2014

Sorting Out the Synclavier

Keeping in mind that I'm dealing with a very complex vintage computer music system, it is no surprise when things need to be sorted out. The vexing problem has been background distortion and ghost notes. Since I've been a Synclavier guy since the 80's, this is by no means new to me. I worked on my own system a lot over the years, and even built one for John McLaughlin to use at the 1987 AES show in NYC when his was suffering some problems. So, I dove in and sorted the voices out today (An all-day job).

One thing you need is good lighting. Oh, and dig those 5.25" floppy drives!

Then, I needed to remove all of the voice cards except those for voices 1-8, which you can see at the left of the bin.

Here are the removed cards. Each row of five cards is for 8 stereo voices. Today, all of this could be on a single chip! However, without the incredible digital to analog converters, you could not duplicate the sound, even though these voices are 8 bit!

After booting up as an 8 voice stereo system, I used a sequence to torture test them. Voices 1-8 turned out to be fine.

So, I went on to the next step, which was to install voices 9-16. Before doing that, however, I used a pencil eraser to clean the backplane contacts.

Here is the upper bin with 16 voices, which also tested fine.

On to 24 voices then, which also tested fine. I was certain by this point that the problem was with voices 25-32.

Sure enough, when I reinstalled voices 25-32, the distortion and ghost notes were back.

Here is the culprit voice card set.

This was an easy sort out, all things considered, since it was the final set that was causing the problems. So right now, I have 24 good voices, but that's enough to start working on Fuga Electronica!

Monday, January 20, 2014

Fuga Electronica 2: v1.0 MIDI Files Complete

Another couple of weeks, and another milestone. This week I got the v1.0 MIDI files complete in Encore for all nine of the fugues. All I have done for these first versions is to get all of the staves in order so that the MIDI files will end up in the top tracks of the Synclavier's sequencer, and I put articulations in so that the durations of the notes is what I want. I will wait to do dynamics until I have the first Synclavier versions recorded and saved. Here are the first pages for each score/MIDI file.

Here is the opening Allegro. It was originally a Sonata-Fugue for two guitars, so to make the transcription process easier and less prone to errors - I spend a LOT of time fixing transcription errors - I used 8vb treble clefs. There are 5 because in the development area five voices are used briefly, and the new sound is used for the following canon as well.

The Lament was originally for string trio, so I just had SAB clefs before. I added a contrabass track to double the last statement in the bass an octave lower, and I used Alternate Bass and Contrabass tracks to fatten up the final cadence.

Track 3 is the Valse Macabre, which was originally a three-voice fugue, the subject of which is a twelve-tone row, for wind trio. So again, SAB is what I had before, and I just added a Contrabass track to fatten up the last bass statements by doubling them an octave lower.

The Andante was originally for string quartet, so I added the CB track to make it like a string choir, and I use one Alternate Alto track to bring out the only statement in diminution that is in the piece. I'm trying to make the Synclavier "orchestrations" as basic as possible to save voices for the sound effects I want to develop and employ. I'll do traditional orchestrations for this and a few others in Logic Pro X to add to the Synclavier's synthesized timbres, but that will be later this year, after I have the Synclavier versions perfected.

Track 5 was originally a very elaborate Ricercare for wind choir functioning as a wind quartet with very wide part ranges, using auxiliary instruments to accomplish the range shifts, and leaving out the contrabassoon. All I did was add the CB clef, and the setup for the Jubilate was fine.

My Scherzo Comico was originally for chamber orchestra, so this one needs all 8 tracks; SATB plus Contrabass, plus alternate alto, tenor and bass tracks for the hilarious final cadence.

This little Allegretto was originally a solo guitar piece, so it has the 8vb treble clefs, and I only needed an alternate bass part, and nothing in the contrabass octave at all.

The penultimate Adagio is the epic five-voice Ricercare for symphony orchestra I completed last year - which finished all of the music I needed for this album - and I was able to just delete the wind and brass parts that duplicated the string parts, and cut-and-paste those into the string section that weren't dups, to end up with the "virtual string choir" version here.

For the Finale, which was originally a Ricercare for one, two, or four guitars, I had to use seven of the eight available Synclavier tracks. This is mostly because the final thematic statement/cadence is a series of five and six-voice guitar chords. The seventh track, then, is just for the contrabass doubling of the late bass statements.


Okay, now I will spend some time in the Synclavier documentation to figure out the best workflow. The process is actually pretty straight forward: I set the Synclavier up to record the correct number of tracks on the right MIDI channels, activate MIDI Sync, press record, and simply play the Encore files into the Synclavier's sequencer from my Mac (Ive done this many times before, but it has been many years!). Once the raw MIDI tracks are in the Synclavier, then I will develop and assign sounds. Since I have bunches of disks - 5.5" floppies! - of timbres I've programmed since 1984, it shouldn't take too awfully long to get initial versions, actually.<.p>

Sunday, January 12, 2014

Fuga Electronica 1: Synclavier Transcriptions Complete

I have been highly motivated to get these done, so it took less than two weeks! Using the templates mentioned in the previous post, everything went as planned. With just a generic string section sound, I was able to work out the tempos and the number of staves for the final versions. Here's how it looks now:

I was figuring about 40 minutes total, and I'm 30 seconds short of of that, which is fine. The pieces range from 0:55 to 9:55, so there is nothing longer than ten minutes, which is also perfect. The next step will be to set up individual templates with the correct number of staves, so that all the tracks will begin with Track 1 in the Synclavier's sequencer. Then, I can sync the Synclavier directly with Encore and record the MIDI tracks into it in real time. After I assign timbres, I'll have the initial versions!

I figure another two weeks and I'll have the individual versions ready to record into the Synclavier.