Monday, September 29, 2008

Purple people eater

My latest acquisition -- an Ensoniq Fizmo:

This really wasn't in my plans, but I couldn't pass up the opportunity to pick it up from a local seller.  And I don't have any other synth that does wavetable scanning. so it'll be something new.  I'm not sure if I'll wind up keeping it or not.  Once the Solaris I have on order comes in, I won't have any place to put the Fizmo; I've already got three tiers on my A-stand and I don't want to add a third tier to my Z-stand, since the V-synth has to go on top and that will make it too high.

This Fizmo has the time-bomb LM2940 voltage regulator.  Because of that, I haven't powered it up yet, and I'm not going to until I get a chance to replace the regulator.  The seller gave me the replacement part, which he had received from Ensoniq but never installed.  The synth came out of my car and went straight to the workbench.  It may be next week before I get to it; my wife and I are appearing in a ballroom dance competition this weekend, and that is taking up most of my spare time this week.

Saturday, September 27, 2008

Words of Wisdom from Alan Parsons

As quoted by Music Thing:

Never trust anyone in the music business with a vowel in their name.

Friday, September 26, 2008

Deciphering Roland serial numbers

I found this writeup on decoding the serial number of any Roland or Boss product manufactured since 1989.  All Roland/Boss products manufactured since that time have a serial number consisting of two letters followed by five digits.  Here is the serial number of my JD-800:


As it turns out, this contains two pieces of information: the actual serial of a unit of a given model, and the month of manufacture.  The format is rather strange.  The second letter and the first digit give the month of manufacture.  These count like this: A0, A1, A2... A9, B0, B1, B2, etc.

A0 = March 1989.  As you can see, there is no clean correspondence between the letters and years; you have to count it up.  Or, the link above contains a table where you can look it up.  C7 is the 27th month since March 1989, which amounts to August 1991.  Since I bought the JD-800 in the summer of 1992, evidently the store had had it in stock for nearly a year before I purchased it.

The first letter and the last four digits give the actual serial, which is the count of units of that model manufactured.  To figure this, you count the letter as A = 1, B = 2... J = 10, K = 11, and so forth, and you append the four digits to that number.  So, a theoretical serial number of BD75191 would indicate the 25,191st unit of whatever model it was.  However, looking at the JD-800 model number above, you can see that it begins with Z.  What is Z?  Well, Z = 0.  So, my JD-800 is the 9,377th one manufactured.  A thought about this: The JD-800 did not sell that well when it was first introduced, which explains why the store had it in stock for so long, judging from the manufacture date indicated in the serial number.  And, mine must have been built near the end of the production run (actually, I wasn't aware that they had made as many as 9000 of them).  

Looking at the rest of my Roland gear, none of them have a serial number beginning with any letter other than Z.  I don't have a Fantom, a JV, or any of the more popular Boss pedals, and likely of the Roland gear I do have (other than the Juno-106, which was manufactured prior to Roland's adopting this system), none of them were made in quantities as large as 10,000 units.  The lowest serial number of any Roland gear I have is ZR02527, which belongs to my Boss DD-20 delay (manufactured May 2003).  

The system can go up to a unit number of 259,999 (which would be Y__9999).  The page I linked to says that Roland has sold more Boss CH-1 chorus pedals than that, and that when it overflowed, they simply started over again with Z__0000.  An interesting thing to note is that the date codes will overflow in November 2010, and judging from the above example, they will probably just start over again with that month becoming A0.  The CH-1 chorus was in manufacture in 1989, so sometime in the future, it might be possible to find a pair of a very old and very new CH-1 with the exact same serial number.  

JoMox to discontinue the SunSyn?

It appears so.  Enough parts are left for a handful more units, and after that... no more.  The culprit is our old bugaboo, discontinued ICs.  JoMox isn't saying which ones specifically, but I'm guessing it's an op-amp or OTA.

Monday, September 22, 2008

Release Velocity

If you've spent time studying the MIDI specification, or ever looked at any MIDI data dumps, you probably know that a Note On message is sent by a MIDI keyboard whenver you press a key on the keyboard, and that the Note On message is three bytes long:
  1. The first byte indicates that the message is a Note On message, and contains the MIDI channel number.
  2. The second byte indicates which note was played.
  3. The third byte is a data byte that indicates the velocity, which is a measure of how quickly or slowly the key was pressed.

Each Note On message must, eventually, be balanced by a Note Off message that tells that note to stop sounding. (Otherwise you wind up with stuck notes.) The Note Off message is sent when you release the key -- let it up or take your finger off.

A Note Off message, like the Note On message, is three bytes long. The first two bytes are the same: the first byte contains the Note Off type code and the MIDI channel number, and the second byte indicates which note was released. So what does the third byte in a Note Off message do? It contains what is called the release velocity. The release velocity is a measure of how quickly the pressed key was allowed to come back up to its rest position.

Release velocity has been in the MIDI specification since the beginning. As it turns out, the most common mechanism used in scanned keyboards to measure velocity can also provide release velocity "for free"; the software which controls the keyboard scan just has to notice it. So it's a bit surprising that many velocity-sensing keyboards don't do anything with it. Of the synths that I own, only the Oberheims (Matrix-1000 and Matrix-6R) provide it as a routeable parameter. The JD-800 uses it for one purpose, which is providing a surrogate velocity value for a note that is retriggered in solo mode when a higher-priority note is released. I was surprised to find that the V-Synth, which I'm now using as my master keyboard, doesn't provide release velocity as a routeable parameter, or do anything at all with it as far as I can tell. However, it does output release velocity in the MIDI data it sends.

So why don't more synths use it? I'm not sure. One thought I had is that release velocity, as far as I know, does not have any significant effect on any non-electronic keyboard instrument that I can think of. It might make a slight difference on a pipe organ that has tracker (purely mechanical) action. On a piano, it would effect how quickly the dampers recontact the strings, but I doubt that there is any practical effect, as I've never heard any piano player mention it. Given this, it is probably true that few players have worked at developing any skill in controlling release velocity. I had never really done anything with it up until a couple of weeks ago, when I decided to use it to control envelope release time on a Matrix-1000 pad patch for something I was putting together. I found it rather difficult to control at first; it's odd, seemingly counter-intuitive, to think about release velocity.

I have put together a video demonstration of release velocity on the Matrix-1000 (using the V-Synth as a controller). When I put this together, I found something interesting and rather bizarre about the implementation on the 1000. Since release velocity is obviously not available until key release, I had assumed that the synth would use a default value of 0 for that parameter through all of the envelope phases prior to release. No so. What it actually does is this: at the time a note is released, the release velocity is obviously applied to the note through the release phase. However, that release velocity value is retained and applied to the next note played by that voice! The value of the release velocity parameter will be the value of the release velocity for the last note played on the same voice, until the next note is released. This can lead to some wild and hard-to-predict results, depending on which voice allocation method is selected. You can hear this particularly in the video's last segment, where some notes in a chord have wild pulse width modulation and some don't, depending on previous notes and how the voices were assigned to the chord. What you do with the previous note can effect the next note. It's a bit like having a demented version of polyphonic aftertouch.

(P.S.: I had to re-shoot part of the video due to problems with my camera. All complaints about continuity errors will be cheerfully ignored.)

Monday, September 15, 2008

Patch Bay 101

If your studio has more than one or two synths and effects, you need a patch bay. If you have more sources than your mixer has inputs, you need a patch bay. (Easy to have happen these days, with nearly all synths now on the market having multiple outputs.) If you're walking on a pile of cables criss-crossing the floor, and you're afraid to disconnect any of them because you don't know what they do, you need a patch bay.

So what does a patch bay do? It gives you a place to bring all of your signal sources and destinations together, so that you can connect things in any arbitrary order using short patch cords. In that respect, it's no different from a modular synthesizer. However, a well laid out patch bay can also make your life easier by having pre-made "default" connections that correspond to the way you normally connect things in the studio, but can be easily overriden when desired. And, by making it more convenient to change connections, it can enable the use of things that might otherwise be too much trouble to cable up -- like auxiliary outputs from synths.

Your standard 19" rack-mount patch bay consists of two rows of 24 1/4" jacks each on the front, with a like amount on the rear. (Some pro patch bays use a different connector called a "TT" connector, which looks like a 1/4" connector but is smaller; it's about 1/6" in diameter. You'll probably want to avoid those for a home studio, because you obviously have to buy different cables for them.) Each jack in the rear corresponds to a jack in the front; there's a direct connection. The rear is where you make your (relatively) permanent connections: you bring all of your signal inputs and outputs to jacks in the rear. Each connection in the rear causes the connected input or output to appear at the corresponding jack in the front.

(Some patch bays designed for large-studio use have other connection means in the rear, such as large multi-pin connectors, punch-down terminals, or lugs for soldered connections. These are used by pro studios because they are more reliable means of connection. They do, however, make it much more difficult to change things around. Home studio users should probably stick to the patch bays that have 1/4" jacks in the rear.)

The patch bay is a completely passive device, requiring no power, and so any jack can be used as either an input or an output. However, by convention, outputs are connected to the top row, and inputs to the bottom row. And the patch bay is designed with this in mind: the process of setting up those "default" connections relies on it. In recording engineering terminology, the process of setting up default connections in a patch bay is called normalling. When a "behind the panel" connection exists between a top row jack and the jack below it, the two are said to be normalled together. Any such pair of jacks can be, with no patch cords connected, in one of three states: full normalled, half normalled, or non-normalled.

As you might expect, non-normalled means that no default connection exists. The difference between full normalled and half-normalled is this: In a full normalled setup, whenever a patch cord is inserted into either the output (top jack) or the input (bottom jack), the normalled connection is broken. Inserting a cord into the output jack routes the output to the cord and leaves the input jack connected to nothing, unless a patch cord is inserted there also. Similarly, inserting a patch cord into the bottom jack connects the signal from the cord to the input, and the ouput is left connected to nothing until and unless a patch cord is inserted there also. In a half-normalled setup, the bottom jack behaves the same, but the top jack behaves differently: inserting a patch cord there does not break the normalled connection. Assuming that the bottom jack has no patch cord inserted, then inserting a patch cord in the top jack effectively creates a Y-connection: the output signal is now routed to both the normalled input and the inserted patch cord.

Here's a tip: Less expensive patch panels usually don't support full normalled connections; the choices are half normalled and non-normalled. That's okay, because as it turns out, full normalled connections are not all that useful. When you patch into an output jack, creating a Y-connection is often what you want. And if you don't, it's no big deal to insert an unconnected cord or a shorting plug into the corresponding input to interrupt the normalled connection. Full normalling just doens't buy you much over half normalling, at least not for electronic music and home-studio purposes.

Nearly all patch panels let you choose the normalling on a per-jack-pair basis. The mechanism for doing this varies. Below you see a photo of my two patch panels:

And the rear:

Both of these are inexpensive models, but the top one is somewhat less expensive. As you can see from the rear photo, it's an "open frame" design, with the case not fully enclosed. There's a reason for this: Each set of four jacks (two front, two rear) is on its own circuit card. The card is held in by a retaining nut that clamps it against the retaining strip in the rear. To change a card from half-normalled to non-normalled or vice versa, you undo the nut, pull the card out, flip it around, and put it back in. This makes it very easy to change. The jacks and retaining nuts are all plastic. It's not the most rugged patch bay in the world, but for home studio use it's fine. It came with one spare card, so I have a replacement if one goes bad, although there isn't much that can go wrong with it.

The bottom one is a little bit higher-priced. It's more rugged, with all metal construction. Like the top bay, jack pair normalling is changed by flipping the card. Problem: To get at the cards, you have to remove the front panel. Since the front panel is what holds it in the rack, what this means in effect is a massive disassembly job whenever you want to change the normalling. To save a little time (and avoid having to undo and redo all of the rear connections), I've developed the trick of leaving it mounted and detaching the body of the patch bay from the front panel by undoing the hex screws that you see in the front. Then, I can pull the body a few inches away and get at the cards.  See below:

However, getting it reattached, lining up the screw holes and all 48 front-panel jacks, is quite a balancing act. So if you need to change the normalling on a pair, it's a big time-consuming production. Because of that, I tend to use the bottom bay for more permanent things, like mixer inputs and effects sends/returns, and I use the top panel for things that get changed more often.

So how should one configure a patch bay? Other than obvious things like grouping similar sources and keeping stereo pairs together, I have one tip to give: use normalling as much as you can. Think some about where you usually route things and normal those connections. For example, I usually want my Juno-106 run through a compressor, in order to control the unpredictable level excursions that its chorus creates, so I have its outputs (stereo) normalled to a dbx 266XL's inputs, and the 266's outputs are normalled to a pair of mixer inputs. Once you've done that much, normal any other pairs of inputs and outputs you can round up (unless doing so would create feedback). If you can't think of a normalling that makes sense, choose randomly. Why? Because normalling makes your life simpler. Once you start working with the setup, you'll realize that either the normalling you set up works, or else you need to change it; either way, you'll know what to do. But anytime you have an output that you can match up with an input in a normalled or half-normalled configuration, do so.

Concerning labeling: One of my panels has a scribble strip, but I've found that the markings get rubbed off as I insert and remove patch cords. So I prepared some labels in a drawing program in my computer, put them in a line, printed it, and cut out the lines of labels as thin strips of paper. I then taped it under the jacks with cellophane tape.  Example:

That's pretty much all there is to it. A few miscellaneous tips: Keeping in mind the number of cables you will need, you might want to think about locating the patch bay to minimize your average cable run length, in order to reduce your cabling expense. If you have one place (say, your mixer) where a large number of cables are going, put the patch bay near it. Since the patch bay will have the effect of combining all of the signal grounds for everything connected to it, try to have all of your gear driven by the same electrical circuit if possible, to avoid ground loops. And do use decent quality cables.  You will probably want some means of being able to identify the cables.  You can buy different color cables, or wrap pieces of colored electrical tape around the ends to identify them, or print your own cable labels on your inkjet printer.  You can see I've taken all of those approaches in the rear-panel shot above.  It doesn't really matter, as long as you can make them distinct.

I have prepared a tutorial video that shows (1) the use of half-normalled connections in a patch panel, and (2) how to change the normalling in one jack pair.