Another Day in Paradise

…not. Sometimes you just want to shoot your computer.  Today I did some remastering work on three or four songs for the new CD I’m planning to release, and then attempted to burn a test CD so I could listen to it on various playback systems.

My nice new HP computer, which burned an audio CD yesterday without complaint, has now decided that it has other ideas. PreSonus Studio One, which has a swell utility page for mastering CDs, couldn’t manage to initialize the DVD/CD drive at all. iTunes burned a CD, but it had crackling noises in it — on some players, though not on others.

I spent two solid hours on the phone with HP tech support. At least they answer the phone, and the young lady in the Philippines was very patient. But I’m afraid we’re wandering into one of those horrible dead zones where HP says their drive is fine, it’s the fault of the PreSonus software (or maybe the fault of the CD player in my stereo, though it’s fine with every other CD in my collection), while PreSonus (whom I have only just contacted — no response yet) says no, it’s the drive, our software is fine.

Me, I think it’s the drive. I think the burn laser is misaligned. The good news is, the computer is still under warranty. The bad news is, how can I get them to send me a replacement drive if they insist that it’s the software?

If I ever get this squared away, I swear to God, I’m going to lock down this system. No more installs. No more updates. No more swapping out drivers. If anyone even dares suggest it’s the other vendor’s fault, I’m going to drive over to their house with a golf club and break their windows.

Here’s the deal: Computer technology has gotten too complicated. We don’t need all this shit. Hell, I didn’t need Windows 7. I was doing fine with XP. But you know, it’s the new OS, and I’m buying a new computer, and it’s supposed to be better. Besides, up until this month I was busy writing magazine articles on music software. I knew editors would want to read reviews that were written using Windows 7 as a test system.

I think I’ve just retired. Fuck it. I still have an assignment from Keyboard for a product review, but I am finding it more and more difficult to get psyched up to write it. If I had been able to burn a nice CD today, I would have been in a good mood, and this weekend I would probably have started on that review. But now … no way.

Footnote: The next day, it occurs to me that I still have my old PC. Transferring the files, installing the software, and burning a CD on it only took about two hours, and next time it will be quicker. Now I can get back to worrying about what I need to improve in the mixes themselves.

There is more than one way to skin a cat. Personally, I like to dangle them by the tail over a vat of boiling oil and then dip them ever so gently. Falconer’s gloves are a helpful aid in this endeavor, as the cat tends not to be happy about the prospect.

Testing What, Exactly?

Artificial intelligence is an area of research that fascinates a lot of people. I’m not entirely sure why. It seems to me that promoting human intelligence would be a far more useful activity. But I digress.

In dissecting the article on David Cope (see previous blog entry), I made a passing reference to the Turing test. That got me curious. The Turing Test is supposed, by some people at least, to address the question, “Can machines think?” The way the Turing Test works is this: You’re having a conversation with an entity, and you don’t know whether the entity is human, or whether it’s a computer. The “conversation” is held by typing words into a computer, and reading the entity’s responses on the machine.

The thesis of the test is that if you can’t tell whether the entity with whom you’re conversing is human or a computer, the entity is exhibiting intelligence. It has passed the test. Several serious criticisms have been made of the Turing Test. (If you’re interested, read the Wikipedia article. That’s what I did.) Basically, it has been blown out of the water. It isn’t even worth discussing, except that people keep bringing it up because it’s simple and obvious and seems, at first glance, to be a meaningful benchmark.

I claim that I could beat the Turing Test with any software program ever devised. In less than 30 seconds, I could tell whether I was dealing with a human or with a machine. I would type the following message: “Hey, I just won 2.6 million dollars in the lottery!!! I’d like to share the wealth. I’ll write you a check for $10,000 today, but only if you’ll meet me for lunch. Where would you like to go for lunch?”

My suspicion is that if I typed that text, the experimenter would stop the test. They would say, “Hey, that’s not fair. You can’t test the entity by inviting it to leave the testing environment. You’re supposed to restrict your conversation to X, Y, and Z, and only typing is allowed.” In other words, you’re not allowed to interact with the other entity as if it were a real human being.

That’s another reason, one that has apparently been missed by other critics, why the Turing Test is a fraud. An artificial set of conditions is set up, and the test can produce results (supposedly, reliable ones) only within the artificial parameters of the test.

This is a bit like testing whether humans have depth perception by forcing them to view test objects out of one eye at a time.

It’s also a dandy example of reductionist thinking. Scientific research has thrived on reductionism for hundreds of years, but we now seem to be reaching a point where reductionism no longer works. The simple questions, the ones that can be answered using simple artificial tests (Galileo rolling balls down an inclined ramp, for instance) have all been answered. What we’re now grappling with are questions so complex that they can only be answered by looking at the test subject in its native environment. To move forward, we need to look at all of the factors that may be coming into play.

The design of the scientific test itself is always one of the factors. It can’t be left out of the analysis.

This is why things like testing new drugs before releasing them have become so difficult. To test a drug, you have to give it to human subjects and then observe what happens. But once a chemical enters the human body, it can affect any of the thousands of metabolic processes that are taking place in the body. And the effects may not show up for months, or years. One possible outcome: Thalidomide babies. Not only that, but the drug can interact with any other drug that the person is taking, and also with any chemical that the person comes into contact with in their environment.

Scientists have gotten pretty good at predicting some of these effects and interactions, but of course they make many mistakes. It’s not their fault. The problem is not inept scientists; the problem is that reductionist scientific testing methods just don’t answer complex real-world questions very reliably.

Holistic thinking got a sort of bad rap in the Seventies, because a lot of hippies were trying to apply it in simplistic ways. But I think we need to go back there and have another look.

One of the huge problems today in what passes for political discourse in the United States is that simple solutions are proposed (mainly by right-wing knuckle-draggers) for complex problems. This is reductionist thinking in full flower. But that’s a topic for another time.

I wonder if kids today are being taught in high school how to think about complex problems. I hope they are, because they’re going to have a bunch of them.

Coping Mechanisms

Can a computer program ever compose music that’s as good as what a first-rate human composer can devise? I claim that the answer is “no.” And I’m going to prove it.

I started thinking about this question after reading a new piece about David Cope, the retired UC Santa Cruz professor who has devoted his life to developing software that composes music. Cope has gotten a fair amount of press over the years — quite a lot, really, considering the marginal quality of the music his programs have produced. He has also garnered a fair amount of criticism. Since I’m one of the critics, I felt not only curiosity about what he’s up to now, but a sense of obligation. I don’t want to be unfair to the guy, after all. Maybe my view 20 years ago was too harsh.

The new article (you can read it online at http://www.miller-mccune.com/culture-society/triumph-of-the-cyborg-composer-8507/) was written by Ryan Blitstein. Blitstein’s bio identifies him as a journalist, but does not mention any musical training or musical credentials, a fact that might alert us to potential weaknesses in his presentation. In what follows, I’ll weave back and forth between Cope’s own work as I understand it and the claims made for it by Blitstein. Cope’s own web pages are at http://artsites.ucsc.edu/faculty/cope/index.html.

In a nutshell, Cope has written several programs over the years that analyze musical materials fed into them (melody lines, chord progressions, rhythms, polyphonic textures, voice leading, and so on) and then produce new music that conforms to or arises out of the analysis. In his early work, for instance, he input hundreds of four-part chorales by Bach. After processing this data, the software was able to synthesize entirely new four-part chorales that were more or less in the style of Bach.

Not having studied exactly how the program goes about its analysis, I’m not qualified to judge how good it is at doing that. Nonetheless, as an academic exercise that may prove useful to musicologists, it’s a remarkable achievement of software design. This may be Cope’s real contribution to the universe of discourse.

The claim is repeatedly made, perhaps by Cope or perhaps by those around him (I don’t have any direct quotes from him on this subject) that listeners who are not aware that the music was composed by a computer can’t tell. They believe, or assume, that it was composed by a human being. Or so we’re told.

It may be true. But we may want to remind ourselves that until recently most people believed the Earth was flat. Human judgment is notoriously prone to error. In any case, I’m not aware of any double-blind studies in which listeners

Do the Twist

Today I’m in pain, and it’s because yesterday I was having so much fun playing music. No, playing the piano doesn’t hurt. Neither does playing the cello (though I’ve been having a little problem with one finger, thanks for asking). What’s painful is using my computer music workstation.

Doing this type of work involves four separate components — the QWERTY keyboard and mouse (which we’ll count as one component, since they sit side by side), the computer screen, a pair of large stereo speakers (again, one component), and a five-octave music keyboard. The difficulty is, it doesn’t seem to be possible to get all four of those components into an ergonomically healthy physical arrangement.

The computer screen and QWERTY/mouse are in a good arrangement, considered by themselves. The table is the right height, as is the screen. But the music keyboard is off to the right, at a right angle to the computer table. In order to work with a music program, I find myself sitting in a twisted way, with my left hand near the mouse (yes, I’m left-handed) and my right on the music keyboard. This twists my right shoulder back at a fairly sharp angle. And while I’m editing on-screen, which I do a lot, I’m hearing the left speaker channel, essentially in mono, with my right ear.

I can roll backward and turn so that I’m facing the music keyboard and have the speakers directly in front of me in a good listening position, but then I can’t see the computer screen without twisting my head to the left, and I can’t reach the mouse at all.

I’ve seen charming (and expensive) pieces of studio furniture that are intended to address this type of problem. I used to have one in my office at Keyboard, in fact. This design puts the QWERTY keyboard and mouse on a little pull-out tray under the music keyboard, and the computer screen behind the music keyboard, between the speakers.

Swell idea, but in my experience it never quite worked. The pull-out tray is so low your knees bump into it, while the music keyboard is perched so high that it’s not at a good playing height. Plus, if the tray is pulled out (which it needs to be in order for you to use the QWERTY keyboard or mouse), the music keyboard is too far away to reach comfortably. And if your eyes aren’t good (mine aren’t), the computer screen will be so far away that you’ll constantly be leaning forward to see it.

Playing music should be comfortable. You want to be concentrating on the music, not constantly rubbing your shoulder. This is another one of those darn conundrums. I have no answers, I’m just grumbling.

Slotted

So I ordered a new PCI card for my new computer in order to get it to work properly with the Yamaha mLAN driver. But — whoops, silly me! I looked at the back of the computer, saw metal tabs over card slots, and figured, hey, I know how to put in a PCI card.

If I had taken the trouble to look at the specs of the computer before ordering, I might (maybe) have noticed that it uses PCI Express. Confusing name, because PCI Express ain’t PCI. The two card types are fundamentally incompatible, although in theory, your software won’t know the difference. (That is, I want to emphasize, the theory.)

So I spent $35 on a PCI card that I can’t use, and wasted a week waiting for it. When it arrived, I had to hack the plastic packaging open with a box cutter. God forbid they should put these things in packages you can open with your fingers. Now the package is ripped to shreds, so there’s no point in trying to return it.

One might legitimately ask, shouldn’t the Yamaha tech have said, “Does your new computer use PCI or PCI Express?” But no, that would be passing the buck. It’s my fault.

Hewlett-Packard — reinventing the wheel, one spoke at a time.

It’s … Blue

Creating music with Csound is a slow and fiddly process, granted.  But I’m still playing with it. The essential thing, it seem to me, is this: Conventional music software makes certain assumptions about your music — that it will have a time signature, for instance, and that notes will be arranged in a scale with 12 notes per octave.

These assumptions are valid for many, many people. That’s why manufacturers make the assumptions: If people don’t buy the software, the assumptions are wrong, and need to be changed. But nobody is buying Csound — it’s free.

If you sit down to play the piano, or to write piano music, the nature of the piano will steer your creative effort in certain directions. That’s not a bad thing, it’s just the nature of the instrument you’ve chosen. You won’t be writing tones that glide smoothly from one pitch to another, for instance.

Conventional music software is rather like a huge, complex version of a piano. The options you can use go much further, but ultimately, you’re still dealing with piano-like notes. And because you can buy synthesizers that have hundreds or thousands of sounds that are ready to play, you’ll often be able to craft finished pieces of music without needing to think in much detail about the sounds you’re employing. You can just grab a preset and play it from a MIDI keyboard.

Your synthesizer’s browser probably has categories for bass, lead, and pad sounds, for instance. This makes perfect sense, but it embodies an assumption, namely, that your music will probably include a part that is functionally a bass.

Csound tends not to make that sort of assumption. It doesn’t even require that your music be constructed using a series of notes. Well, you need to put at least one note in your score, but it can be used to turn on an automated or interactive process that will run for hours.

Csound events are arranged in linear order on a timeline, because it’s the nature of music that it is heard over some span of time. But there’s no built-in concept of meter.

Nor will you find a large palette of pre-assembled sounds that you can use. Sure, there are lots of Csound instruments available online, and they’re worth study, but mostly they were created by a composer for one specific piece. They’re not likely to be something you’ll want to use without modifications. Or at least, that has been my experience so far.

In any event, Csound instruments are by definition almost completely modifiable by the user. With Csound, you really do have to think about every detail of the sound you want to create. That’s why it’s a slow, fiddly process. If you want a string pad, you start by listening to the sound in your head and noticing details. Then you use Csound’s tool palette to craft that sound. There are usually several ways to do whatever you’re wanting to do, but none of them is nearly as easy as grabbing a preset in a browser.

Csound is a blank canvas. Conventional, commercial software is like a paint-by-numbers set. It’s an amazingly complex and versatile paint-by-numbers set, but it’s not a blank canvas.

I’ve been learning the basics of blue, a feature-rich “front end” for Csound. Blue doesn’t eliminate the need to write your own code line by line, but among other amenities, it has a timeline with multiple tracks, much like what you’d see in a conventional sequencer. This speeds up the composition process, but without introducing too many assumptions about your music. At a basic level, blue is simply a multi-windowed interface with easy block copying.

There’s a lot more to it than that, which I haven’t explored yet. Blue allows you to make a few conventional assumptions about music if you want to. It has a piano-roll editor and even a tracker for building patterns and phrases the easy way. But I doubt I’ll find those very useful — and they’re not central to the interface, the way they would be in a conventional sequencer.

On the back end, blue hosts Python code, which can be used to generate Csound scores. Python is supported in native Csound as well. I have no idea how I would use it — yet. But using Python just seems a lot more interesting to me than inserting notes in a piano-roll editor.

Don’t ask me why. Something to do with brain chemistry, I suppose. Some people think it’s fun jumping out of airplanes, or watching football on TV. You gotta go with the juice.

Keeping Track

Update on the “swirly shit” mystery track conundrum: A Yamaha tech support professional explained to me that there are actually two different Motif XS editors for Windows 7. One is stand-alone and the other is a VST plug-in. (In fact, there are two versions of the latter, one for a 32-bit OS and the other for 64-bit, making three editors in all.)

When I download and install the stand-alone editor, I can load my two-year-old song files and see all of the correct sound programs in Sound Manager, which is a utility shell that forms a hub between Cubase, the XS Editor, and the XS hardware unit. (If all this sounds convoluted, that’s because it is convoluted.) The swirly shit is Preset Bank 7, program 93, “Clearing.”

I still can’t use the Motif XS with this setup. I’ll have to wait until the PCI Firewire card I ordered arrives, because I get dropouts with the built-in Firewire in the HP computer. But even if the card never arrives, I’m not up a creek, because now I can create the correct multi-timbral templates in the Motif by hand and then send MIDI data to the Motif through a different interface. (If all this sounds convoluted, that’s because it is convoluted. And yes, I meant to repeat that sentence.)

I’m still missing a couple of plug-in synths that were on the old system, but I’m much closer now to being able to replicate the arrangements of various tunes. That’s the point of the exercise: After two years away from these mixes, I can hear many, many things that need editing. Add an extra bar before the chorus, need more variety in the drum part, that string pad is starting to annoy me. Little things and big things. I’m not actually a bad arranger, especially when I can pilfer a little inspiration from the Motif’s arpeggiator patterns, but I’m kind of spotty. I don’t always make quite the right decisions every time.

Having access to the right tools is only half the battle. You also have to make the right decisions. And then you also have to keep your system functional for an extended period of time (years), in case you want to make new decisions about older pieces. This is not easy at all.

In the Chips

In my dream, I plug some stuff into a computer and it just works. Sometimes reality is similar to the dream; sometimes not.

I’ve been trying to get my Yamaha Motif XS to play nicely with my new Windows 7 PC. They’re both swell pieces of gear, and after only a couple of missteps I managed to get all of the driver software installed.

On listening to the computer’s audio output playing back through the Motif (which is part of the point, though far from the only thing the gear is going to do), I hear an audio dropout of a half-second or so, about once a minute. Yamaha tells me that the chipset Hewlett Packard uses for their Firewire bus, a chipset made by VIA, has some sort of clocking problem. Yamaha recommends the Texas Instruments chipset.

So okay, I’ve ordered a new Belkin 3-output PCI Firewire card from Amazon — only $35, with shipping. I can deal with that, always assuming it solves the problem.

But I’m left with a lingering question, namely: Why wasn’t this problem solved years ago? If the VIA Firewire chipset truly doesn’t work for sustained real-time communications (which is what you need for music), has Yamaha alerted VIA to the problem? Do the VIA engineers just not care? That’s entirely possible. Pro audio has to be a vanishingly small portion of their market.

I’m told that one of the reasons Yamaha’s Firewire technology doesn’t work with VIA when audio interfaces from other vendors do work is that Yamaha is sending more channels of data back and forth. That makes sense, up to a point. Of course, it would theoretically be possible to write a driver that would look at the chipset and say, “Oh, I’m going to have to limit the user to four channels of data.” But that would mean extra work in order to produce a less capable device, and Yamaha would still get complaints, and who knows what all.

I’m just hoping the Belkin card is compatible with the HP form factor. I’m not going to try to research that; I’m just going to hope for the best. Wish me luck.

Losing Track

Always take good notes on your music projects, kids! As you go along. Every day. If your sequencer doesn’t have a notepad, put a nice text file in the same folder as the sequence file, keep a text editor open on the desktop, and jot down everything. When you back up the sequence file, back up the notes too.

Today I’m trying to do some fresh edits on a piece I recorded two years ago. Fortunately, I saved the basic multi-timbral template in the Motif, and it’s named after the song, so it’s still available … mostly. For some reason, tracks 1 through 4 play the correct sounds, while the programs for tracks 5 through 7 weren’t stored in the template.

One of the tracks is named “swirly shit.” That is just so awesomely helpful, Jim. I could have named it after the patch I used — but no. I have a mixdown, so I can hear exactly what it sounds like. It sounds like swirly shit. But what patch created the sound? On what instrument? Is it even an instrument I still have? Your guess is as good as mine.

I’ve got about 20 mixes that I may be wanting to re-edit. This is number 2. I can see it’s going to be tough sledding.

I should really put in a plug for Csound here. Not that you could do a piece like this in Csound, because … well, you could, because Csound can do anything, but it would take at least ten times as long as if you were using commercial synthesizers. The reason to bring up Csound is because it’s open-source. If you’ve saved a .csd file (and they’re just small text files), you can open it up to edit it ten years from now, and it will still sound exactly the way it does today. I’m using Cubase and various other things, in a setup that’s constantly mutating. I go to a 64-bit OS, and maybe some of the plug-ins that I used on that old mix are not compatible with the new OS. Taking notes won’t help with that.

Abstract

What better way to ring in the new year than by listening to a bunch of extremely abstract computer music? I can’t think of one.

My copy of The Csound Book is ten years old. (It was a review copy sent to Keyboard when I was still on staff.) Tonight I copied the contents of the book’s two bind-in CD-ROMs to my hard drive. It’s packed with Csound orchestra and score files, most of which I’ll be able to open and render without trouble. (Some may have been rendered using audio files that aren’t included.)

Also on the disk are .mov audio files, which I guess predate the mp3 format. I can listen to lots of extremely abstract music. Now playing: Jean Piche’s haunting ”Incantation,” which features vocal synthesis and bells.

Much of the music is slow-moving, spacious, and rather thin. A cloud of tone may hang motionless for several seconds before you realize that something is slowly changing. Above all, this is not pop music! It’s not hyped up to keep your pulse pounding. It’s not compressed to jump out of the speakers. There is no beat.

I’ve checked out a few of the new mp3′s on the Csound website — music created by students in Richard Boulanger’s classes. Most of it is loosely in the dance instrumental category: 4/4 percussion and aggressive timbres, with a few exotic colors that say, “Yeah, I’m studying computer music. You got a problem with that?”

If I were Dr. Boulanger, I’d give my students one simple instruction: If your piece is in 4/4, you get an F. No exceptions. But that’s just me. I’m sure they’re great kids, and they’re doing what they care about. I’m just an old grouch. I can’t help thinking they’re missing something, though.

When I was a kid (younger than that), my father took the family on excursions to art museums. I grew up on abstract expressionist painting. So I have no trouble processing and appreciating abstract music. I do it visually.

These recordings remind me, also, of my very first experiments in electronic music. I borrowed Tom Darter’s ARP 2600 synthesizer and Vic Trigger’s TEAC 4-track tape deck, and did some pieces. I wish I still had those tapes! (Not that I have anything to play them on.) When MIDI came in, I started making music that had, you know, chords and beats and bass lines and melodies.

I think maybe I lost track of something important, somewhere along in there. Abstraction is good. It’s clean. It doesn’t impose any particular expectations on the listener. And it isn’t in a hurry. The older I get, the less interested I am in hurry. Where are you going? Why not be where you are?