For a couple of weeks now I’ve been down in the dumps because the plot outline of my next novel is not cooperating with me. Too many ideas, not enough coherence. So I thought I’d take a break and work on a project that would be fun.
That’s when the trouble started.
Over the past 25 years I’ve written and released seven or eight text adventure games (the more elevated term for this medium is “interactive fiction”). They’re all available for free, by the way. If you’re curious, ask for details and I’ll write a separate blog post about that. There’s a whole underground world of text games, complete with hundreds of free games, competitions, programming languages, and more.
Now, I could write a new game all by myself. I’ve done that a lot. I thought it might be a good idea to work with a collaborator this time, partly because it would keep me accountable and partly because kicking around various ideas can be fun.
At one time, back in 2008, I collaborated on a game with a fellow author named Eric Eve. That game was fun to write, and we worked well together. When I posted a message to the forum asking if anyone would like to collaborate on a game, Eric immediately volunteered. Way cool, right? Let’s roll up our sleeves and get started!
Our previous collaboration was done by emailing files of source code back and forth. Eric is in England and I’m in California, so typically I’m working on adding new material while he’s asleep and vice-versa. There’s not too much overlap. A program called a diff program (such as WinMerge) can easily be used to compare two versions of a file. I can look at my existing version, compare it to the version Eric has sent me, and copy the relevant bits into my file. Easy-peasy.
Ah, but that was then. Eric has since adapted himself to an online versioning system called github. The idea is, the master files are on github, in a place called a repository. Before starting work, you download the current state of the project from github. You then do your work and upload the results to the repository.
That would make sense, wouldn’t it? Unfortunately, everything in the preceding paragraph is wrong. You don’t download and upload. No, you Pull and Push. You can also Fetch, and don’t ask me what the difference is between Pulling and Fetching, because I don’t know. Before you can Push (or Pull?), you must Commit. You do all this in your Windows PC using a program called GitHub Desktop. And that’s where the nightmare begins.
I have been able to find no good, basic-level tutorials on how to use GitHub Desktop (hereinafter called GHD). The narrators of the tutorials on YouTube sling around various terms without explaining them. They assume things that aren’t the case, such as that you know what a branch is and what merging is. Nor is there a written, downloadable tutorial PDF for GHD. The assumption seems to be, you’re a professional programmer working with a large team to develop something complicated, so you’ve surely taken a class at the university that explains all that.
I’m not a professional programmer. I’m self-taught. And this isn’t a large project. It’s two people, and there will never be updates from 1.0 to 1.1, 1.5, or 2.0. We’re just writing a freakin’ game.
The problems with GHD are profoundly troubling. First, it will overwrite your current work files without notice, without asking you to confirm. It just does it. Fortunately, I’m smart enough to have created local backups of my work before I launched GHD. If I hadn’t, I would have lost creative work. On principle, however, I reject the idea of using an app that never asks for permission to overwrite a file because it thinks it knows what I want.
Second, it sometimes has its own ideas about what you want. This morning I told it to Push my latest work up to a branch in the repository, and instead it Fetched the out-of-date stuff in that branch and dumped it into my work folder, thereby erasing what was already in that folder. I don’t know why it Fetched when I told it to Push.
Third, the GHD user interface is full of stuff that I don’t understand, and there’s no easy way to find out what any of it means. Yesterday I tried asking chatgpt to explain what was going on. chatgpt is not always reliable, but it’s usually willing to churn out elaborate explanations (though they’re sometimes wrong). This time, chatgpt froze. No output at all. So that’s not a viable way to figure out what’s going on, or what I need to do.
None of this shit is even remotely fun.
I think the idea that Eric and I are working on could turn into a really good game. But the process needs to be fun. If it ain’t fun, I’ll have to find something else to do instead. I have now suggested to him that we go back to the old method, which worked fine when we were developing “Mrs. Pepper’s Nasty Secret.” He’s a lot more of a programmer than I am, so I hope he’ll understand why I want to do it the old-fashioned way.
I’m pretty sure this will be frustrating for him, and that’s not what I want. It should be fun for both of us.
Maybe I’m just getting old. Maybe the problem is just that it’s harder for me to learn new stuff than it used to be. But I don’t think that’s it. I’m good at learning new stuff when I can read a well-written manual, or when the new stuff builds on things I already understand. Neither of those factors is in play here. GitHub Desktop is just a fucking mess. You may quote me.
It’s been three or four years since I had a new book to shout about, but today’s the day! Feast your eyes on this:
I’ve always wanted to write a mystery series, but the two mystery novels I’ve published before didn’t seem to have the right ingredients for a series. I think I’ve got it right this time. The second book in the series, A Thief of Shining Lives, is complete, but because the company I use for cover art and interior layout is taking a break from late December until mid-January, it may be delayed by a couple of months. I’ve got an outline for the third book, but sometimes the story changes when you start to write it.
The world in which Thaddo and Nielle live is not our Earth. The culture and technology are similar to Los Angeles or New York in the 1930s, with streetcars, typewriters, telephones, revolvers, all that stuff, but the social structure is very different. There’s a rigid caste system. Thaddo is one of the Best. He’s a nobleman. Nielle is his faithful office assistant. and she’s not Best. She’s a Scribe. Cross-caste relationships are difficult, but if you’re guessing the two of them may become romantically involved, you’re a good guesser. This is not a romance novel, though. The story is about murder. Well, actually, three murders, one after another. A beautiful young woman is found strangled in a reception room in the opera house. When the Local Authority gets nowhere tracking down the killer, her younger sister comes to Thaddo and begs him to help. After that, things get complicated.
When you’ve written a novel or short story, what do you put at the top of the first page?
I consider myself pretty good at titles. Back in the day, I sold eight or ten stories to the top science fiction magazines, and I never had an editor change a title. (They’re allowed to do that.) Possibly the fact that I was writing headlines every month for a music magazine helped sharpen my insight into what makes a title work.
I’m not sure I could explain to anybody, though, what makes a title work. Some combination of impact, sound, implications, and freshness.
When it comes to naming my own novels, I find that I gravitate toward longish phrases. All of the single-word titles have been used, and most two-word titles just fit together two of those over-used, worn-out words. Longer phrases give you more chance of doing something fresh. The names of my fiction books, to date, are Walk the Moons Road,The Wall at the Edge of the World, While Caesar Sang of Hercules, Woven of Death and Starlight, and The House of Broken Dolls. The Leafstone books followed their own format (The Leafstone Shield, The Rainbow Tree, The Heartsong Fountain, and The Firepearl Chalice). I should have kept my original title for the second one, but I chickened out for a dumb reason and substituted Rainbow for the much better Ribbonglass. Shame on me.
The first book in my new mystery series, What Must Remain Unspoken, may hit Amazon as soon as next week. The second book in the series is now ready to send to the formatting service so they can start on the cover art and the interior layout and design — but I can’t ask them to do the cover until I have a crackerjack title, and I’m having a hard time coming up with one. This sucks.
I don’t want to fall back on a short title; I need something that has meat on its bones. After kicking around a few other ideas, I thought I had decided on Fall from a Great Height, but I’ve lost confidence in it. It seems kind of stale. Something with more pizazz is needed.
A Fallen Star fits the story, and it’s not actually bad, but it’s too short to fit with my ongoing title schtick, and that’s not even the big problem. Here’s a tip: While contemplating the title of your upcoming book, you really need to go over to Amazon and search the Books category for your title. This can be a discouraging exercise. It turns out there are at least nine books with “fallen star” as the main element in their title. Okay, cross that one off the list.
By the way, if you’re thinking of writing under a pen name, try the same thing. If your real name is a common one, you may be in even worse trouble. The other day I was searching for some fiction by William Alexander, and while he has written several middle-grade and YA novels, the Amazon search engine thought I was looking for books by a different William Alexander, a biblical scholar.
I’m not going to reveal the likely title of my next book, because I may change my mind yet again. But I’ll tell you, I did a search for it on Amazon, and it’s safe. There are no other books that even come close.
Someone once said of Don Buchla that he based his instrument designs on the features he wanted for his own music. If a feature would have been useful for lots of musicians but not for his own purposes, he didn’t include it.
Possibly something similar could be said of the software in the Glitchmachines lineup created by Ivo Ivanov. These VST3/AU plugins are amazing, but they’re not for everybody. You want beefy basses and silky pads? Look elsewhere. The Glitchmachines devices are strictly about making glitch stuff: animated bursts of almost-noise that form catchy rhythms or unstable textures. Subtleties are possible, but heavy jolts of sound mangling are more likely.
As of two years ago, when I wrote this review, there were nine plugins in the lineup — four synths and five effects. Rather than choose one or two to review and ignore the others, I wrote a roundup. I’ve worked with them all and noted a few strengths and limitations, but this article won’t be a feature-by-feature description of any of them. Instead I’ll focus on three that I found the most intriguing, and then mention the rest only briefly.
These are deep, complex devices, so even if I tried to explain every detail, the explanations might not make perfect sense. The documentation is good, but half the fun is clicking on things to see what they do. The results you’ll get will depend on your ingenuity and your willingness to experiment. I experimented with the effects plugins using a variety of sources — spoken word, drum loops, and bass lines, for instance. Some of the effects presets work better with some inputs than others.
One caveat to mention up front is that there are no downloadable demos of this software. However, you can hear and see what they’re all about by checking out the excellent videos on the Glitchmachines website. You’ll also find more than a dozen sample packs, including two free ones, with unusual sounds that would be ideal for use with the Glitchmachines synths.
Cataract2. The first device that caught my eye was Cataract2. It’s a wildly configurable groovebox. You can load up to four stereo samples, preferably long ones with interesting details, and then cut back and forth among them using several dozen independent step sequencers, tempo-synced LFOs, and randomizers. The program ships with a large sample library and more than a hundred wild presets.
Two limitations should be noted at the outset. First, Cataract is always going to play either in no time signature at all (free-floating) or in 4/4 time. The step sequencers can be set to one, two, four, or eight measures in length, but they’re always going to be 16 steps long. I mentioned this to Ivanov, and he said they may consider supporting other time signatures in the future, but for now, your choices are tempo-synced 4/4 or non-synced LFO chaos. Since 98% of pop music is in 4/4, this is not a huge problem.
Second, Cataract is not an instrument you can play from your MIDI keyboard or sequencer track. It’s linked to your DAW’s transport, so whenever the transport is running, Cataract will be doing its thing. Assuming you don’t want it to grind on and on from the start of your track to the end, the recommended way to work is to capture maybe a few four-bar or eight-bar loops as audio. Take note of which Cataract presets you used (and save your edits, if any, as a new preset), then delete the plugin from the project and copy and paste your audio clips as needed.
The audio source material resides in two scanners. Each scanner holds a pair of samples, which can be mono or stereo. You can control the pitch and loudness of each sample and the blend between them, and each scanner has a compressor and basic three-band EQ. There is also an active cross-panner that can flip back and forth between the two scanners under the control of a pair of LFOs.
You can scan through segments of the samples using either an LFO or a dedicated step sequencer. And here’s a detail that may not make sense unless you’re actually running the software: The gate length of each segment can be shorter or longer, and the gate can either be on/off or a repeating loop. If it’s a loop, you’ll get something like a pitched tone. If the loop switch is off, the gate always starts or ends abruptly, usually with a little click. For glitch music, little clicks are not a problem, and may even be appreciated, but they’re sort of unavoidable unless you switch gate looping on. Each scanner has a low/highpass filter, but lowering the cutoff of the lowpass filter doesn’t eliminate the clicking.
To add to the fun, each scanner has a couple of extra LFOs for secondary modulation. These modulate the LFOs that control pitch of both samples and the rate of the scanning, and/or the waveform selections used by the primary LFOs. Things can get pretty crazy when you start messing with this feature.
Each scanner also has a selection of effects (crusher, ringmod, chorus, phaser, and formant filter) and a dedicated step sequencer for switching amongst them. Only one effect per scanner will be heard at any given moment. Beyond that, every knob in Cataract can be animated by its own step sequencer. Add this to the hundreds of samples in the Cataract library, and you’ll be sorting your way through the possibilities for hundreds of years.
This overview by no means exhausts the list of Cataract2 features, but it’s time to move on.
Polygon2. What? You actually want to play a keyboard? Then maybe Polygon2 is for you. It can do exotic groovebox beats thanks to its four step sequencers, but it’s equally adept at one-shot rasping, grumbling, and explosions.
Polygon has four slots for loading samples. The samples can be used either in the normal way, complete with loop playback, or as sources for granular synthesis. The output of each sample slot passes through its own pair of processors, which can be a filter (low/band/highpass), EQ, wavefolder, overdrive, bitcrusher, or ringmod. Each sample can either track the MIDI keyboard input for normal chromatic playing, or remain fixed. In either case, the playback pitch can be modulated.
The granular synthesis implementation is not fancy, but it works well. For each sample slot, you can trigger from one to 16 grains per note. They can be spaced out through the sample. The size of the grains and the amount of randomness applied to size, spacing, and pitch can be adjusted. I was disappointed to hear pitch randomness being applied simultaneously to all of the grains rather than individually to each of them. Individual pitch randomness results in a richer tone.
In addition to the samplers, Polygon boasts a couple of oscillators. One of them is suited for adding a solid sub-octave, but it can be tuned to any pitch. The other is a modulator, and can modulate the main oscillator for basic two-operator FM and/or any of the samples. Applying FM to a sample tends to result in thick noise, because the frequency modulation is applied to each and every overtone in the sample. Using the two oscillators, on the other hand, you can coax Polygon into doing more or less normal synth tones. Note, however, that it lacks ADSR envelopes. Attack and release times are the only envelope parameters for any of the sound generators.
A couple of global filters are available, and the output of the filters can be sent to either or both of the global effects (delay, reverb, or stutter). The topology is a little odd, but some cool results are possible.
The step sequencers and the eight syncable LFOs are where the possibilities get wild. Each sequencer can be any length up to 16 steps, and can proceed at any tempo (synced to your DAW or free-rolling). Each step can be a rising or falling segment, a fixed value, or random. There’s also a tidy little set of CV processors and random number generators.
MIDI key number and velocity can be used as modulators. Conspicuously missing, however, are MIDI mod wheel and pitchbend. I hope they’ll add these in an update.
Other Synthesizers. Tactic is a groovebox. You can load up to eight mono samples and build a beat using a 16-step sequencer. Unlike Cataract2, Tactic is not limited to 4/4; you can shorten the sequencer pattern. It then complicates the picture with probabilities and per-step modulations. The per-step modulation sequences can be different lengths from the main sequence, so longer output patterns are possible. Tactic also has a window full of randomization options.
Palindrome isn’t about rhythms, it’s about sustaining textures. You can load up to four samples, each with its own filter and some basic effects, and then build a two-dimensional vector that moves smoothly back and forth among them. Like Polygon2, Palindrome is playable from a MIDI keyboard. But also like Polygon, it lacks control inputs from the mod wheel and pitchbend. This is kind of a bummer, actually.
Quadrant. I’ve never seen an effects processor like Quadrant, and I’ll bet you haven’t either. It’s very, very patchable. There are eight slots for mini-modules, and into each slot you can insert any of 36 different processors. Many of Quadrant’s factory patches are self-generating sound makers: They don’t require and indeed won’t use the audio input. Others are a bit more conventional.
The preset list is long but not organized into categories, and the names are in tiny letters. If you want to process an incoming signal, look for presets that don’t have “Gen” (short for generator) in the name. When auditioning the Gen presets, you may want to put a limiter after Quadrant in your DAW’s signal chain, as a few of the presets can be quite loud and shrill.
The mini-modules include simple items such as a VCA, a noise generator, and MIDI reception, along with more sophisticated options such as a 16-step sequencer, a quad lag processor, a dual FM oscillator, a grain delay, a multimode filter, a clock generator, the inevitable LFO, and so on. There’s even a sequential switch. Up to four inputs and outputs may be available per module, depending on what you’ve chosen.
Patching is not quite as simple in Quadrant as it could be, because all of the graphic patch cords flow through a central area, so it’s generally impossible to follow the signal flow with your eyes. To make the patching more awkward, each input and output on each module has an attenuator, which is great — but the attenuators show up on-screen only when you’ve alt/option-clicked on an input or output connection, they’re really small, and you don’t get a numerical readout of the value. Unless you read the manual, you’ll never know this function is available. This is maybe the worst implementation of control knobs I’ve ever seen.
I’d have to spend a lot more time with Quadrant to understand what it’s capable of or what it’s best at. The learning process might not be all fun, but I can see there are some great discoveries to be made.
Other Effects. A few DAWs have edit windows in which you can patch effects together in various ways, but I haven’t seen any effects processors as thoroughly modular as Glitchmachines Quadrant and Subvert2. By comparison, Convex, Cryogen, and FractureXT are more normal, but “normal” may not be the best word to describe them. “Mad” would be closer to the truth. With all of these effects, you need to be aware that stopping the input audio (by stopping the DAW’s transport, for instance) won’t necessarily stop the output. Depending on the patch, the effect may continue to do its own thing for a while.
If you’re familiar with patchable modular synthesizers, you’ll feel right at home with the user interface of Subvert2. The signal flow among its modules is entirely configurable using graphic patch cords. There are two each of distortion modules, filters, basic multimode filters, ring modulators, digitizers (essentially, bitcrushers), metalizers (fast delay lines), and dynamics processors. Also on tap are two modulation mixers, four audio mixers, four macro knobs, and three-band output EQ.
Convex has a dual signal path. Each path has a filter, a pitch shifter, and a delay line, and in each case you can crossfade between the two paths. The modulation options include four LFOs, two envelope followers, and some mod mixers, multipliers, and inverters. The filters are multimode, but very basic.
FractureXT is a granular effect. It has controls for the input storage buffer, a delay line, and the granular process. These modules can be modulated by four LFOs using a simple patch cord interface. This is probably the simplest of the Glitchmachines machines.
Cryogen is not entirely unlike FractureXT, but it has dual looping input buffers, dual filters, and dual bitcrushers. There are four LFOs and some modulation mixers, which are hooked up to the sound processors using a large modulation matrix.
About the Audio Clip. I’m not really a glitch music specialist, so I combined several of the Glitchmachines plugins with some more normal sounds in Reason 12.6 to produce a sketch I call “Season of the Glitch.” (No, there are no actual quotes from Donovan in the mix.) I don’t claim this track is typical of what people will want to do with the Glitchmachines plugins; it’s just me doing whatever I thought sounded interesting.
The opening sound is from Palindrome. A groove from Cataract then starts, augmented with a kick drum that’s a stock Reason sound. I bounced the groove out of Cataract as an audio track, after which I muted Cataract. The next 12 bars are the Cataract groove with a pad, a bass, and lead flourishes from Reason devices, along with a melody line from the Imagine synth from Expressive. At the end of this phrase I chopped the Cataract audio apart for a percussion build-up.
The second section starts with a dual sound effect from Palindrome and Polygon and a brief flourish in which the bass sound is mangled by Subvert. A different groove from Cataract then starts. The chords in the second section are the same as in the first, but now a new bass sound is being distorted by Convex. The two Palindrome patches return to provide another transitional sound effect.
To wrap up the encounter, I grabbed a Tactic preset (both the sounds and the groove are in the preset). I added my own kick and snare, because the Tactic preset happened to have no downbeat, so a kick was needed. My snare is being processed by Reason’s Polar pitch-shifter to make it a little less normal.
Your Mileage May Vary. I like these plugins a lot, not necessarily because I’ll use them a lot but because I like musical devices that are complex and mysterious. They each have lots of creative presets, so you can use them just by picking a preset and going to town with it. Or you can dive in, learn how every feature works, and craft your own sound library. Loading your own samples into any of the instruments (Cataract2, Palindrome, Polygon2, and Tactic) may prove especially rewarding, because a given patch will sound completely different depending on what samples are loaded. Consider these devices an instant cure for boring mixes.
Here’s a video I spotted this morning, which serves up some quick tips for fiction writers.
It’s self-explanatory. I would insert only one comment. In tip 5 he suggests that moments of great emotion or tension call for short, punchy sentences. That’s certainly true, but once in a great while you can go the other way. A trick I’ve used a few times (never more than once per story!) is to stop using sentences entirely at the moment of greatest action/tension. Instead, show the viewpoint character’s thoughts, feelings, and sensory input in one long spew of words, maybe with a comma or two here or there but basically written as James Joyce or Gertrude Stein might have written it, just an endless flow of interiority. And then when you stop, when there’s a period or a paragraph break, bang! You get a second dramatic effect on top of the first one.
Continuing my series of product reviews that Synth & Software paid for but never published….
Some vintage synthesizers, such as the Minimoog and the Prophet-5, became iconic because they were the first to do something new, and everybody had to have one. Others are legendary because they were unusually powerful, and are maybe even more sought-after today because they didn’t sell well. The Rhodes Chroma is in the latter category. It was a next-generation analog polyphonic synth with what was, for the time, a strong and flexible design. But it appeared in 1982, a year before MIDI and a year before the Yamaha DX7 took the keyboard world by storm. As great as the Chroma was, it couldn’t compete.
Cherry Audio has obviously put a lot of effort into their software recreation of the Rhodes Chroma, and the results are impressive. The original voice design of the Chroma is mostly unchanged except for the addition of some very nice digital effects and the possibility of seamlessly adding more sound layers via the built-in Expander module. Chroma’s original user interface was difficult to work with, and while Cherry Audio has kept the basic appearance of the Chroma panel, it’s enormously easier to use.
The sounds of the presets remind me strongly of the synth-pop hits of the ’80s. A few special effects patches are part of the sound set, but this plugin may not be of much interest to musicians who crave cutting-edge wavetables or physical modeling. Vintage synth lovers, however, will find a lot to get excited about. There are hundreds of presets to choose from, in the usual categories.
Overview. The Chroma voice has two oscillators with a choice of pulse or sawtooth waves. Each oscillator has its own LFO (called a Sweep), a filter with a choice of lowpass or highpass operation, and two envelope generators. However, the LFOs and envelopes of the A and B layers are available for modulating either the A or the B oscillator or filter. In addition, the audio signal path can be configured in various ways. You can run the filters in series or parallel, use one oscillator to FM or ring-modulate either the other oscillator or a filter, and so on. The diagram above the parameter “button” will change to show the currently selected audio path. (Try doing that in hardware!)
As exciting as that sounds, there are limitations. The envelope generators are not ADSRs, they’re ADRs, with no sustain level parameter. If the decay segment is set to 100, they turn into ASRs with full sustain and no decay. This was a sensible enough design decision in 1982, but it’s not what we’ve come to expect today. Multiple filter modes, which are found in many modern instruments, are not included. What you get is an authentic vintage synth design with only a few nods to 21st century tastes.
Also, some of the modulation routings are on/off, with no depth control. For instance, you can modulate LFO rate from velocity, the keyboard, the mod wheel, or an envelope, and any of those souces can be inverted if desired, but there is no way to attenuate or taper the amount of LFO rate modulation from your chosen source. Either it’s on, or it’s off. The modulation routings that have depth control are those that control oscillator pitch, oscillator waveshape, filter cutoff and resonance, and amplitude.
If you’re planning to use the Cherry Chroma in live performance, or even while recording music that has lots of expressive movement, you’ll be happy to learn that it responds to both mono and poly aftertouch. In addition to the standard pitch-bend and mod wheel inputs, there are two “left-hand” control levers, which you’ll want to assign to MIDI sliders. What’s more interesting, but perhaps slightly problematic, is that the hardware Chroma had a spring-loaded bidirectional mod lever. (It wasn’t a wheel.) A modern MIDI mod wheel will only take this lever from the center up to the top, so depending on how the preset is programmed, you may be missing some of its expressive potential unless you record its moves using the mouse.
The pitch glide works in a way I haven’t heard for many years. Because the original Chroma had a discrete circuit for each voice in a polyphonic preset, each voice “knows” which note it played most recently. If you replay the same note again, you won’t hear any glide, because the pitch of that voice hasn’t changed. But if you play some new notes while playing chords, individual voices may glide up or down in a way that’s not directly controllable, because the assignment of voices to notes is effectively random. The result can be dramatic, but it’s not predictable. If you want to glide chords, it helps that the new Chroma gives you control over the number of active voices (2, 4, 6, 8, 10, 12, 14, or 16, plus of course monophonic mode). The original hardware instrument would do sixteen single-oscillator voices or eight dual-oscillator voices, so the software doubles the polyphony of dual-oscillator sounds.
About the programming interface: The hardware Chroma had rows of buttons with nothing on them but numbers. You would push a button and then use a data slider to adjust the value of that parameter. Unless you constantly consulted a laminated sheet with the list of parameters, you would have no idea what you were doing. Cherry Audio has kept the rows of buttons, but replaced the numbers of the parameters with pop-up menus, brief word descriptions, and numerical values for parameters that have a continuous range. Editing is still not as intuitive as one might wish, but it’s very manageable.
Effects. Each of the Chroma modules (the basic instrument and the Expander) has its own distortion, phaser, flange/chorus, delay, and reverb. Instead you can route both modules through the main effects, but I had more fun doing things like adding a delay to the Expander only. At the global level are a three-band EQ and a limiter. Changing the order of the effects is not possible, but Chroma covers the basics, and goes a tiny bit beyond. The delay can do a tape echo simulation, for instance.
Random Notes. I’m told the Cherry Chroma can import not only sys-ex patch dumps from the hardware but also .wav-format audio of patch dumps stored on cassette tape in the pre-MIDI days. I didn’t have a chance to test this. The Rhodes Chroma was pre-MIDI, but a company called Syntech made a MIDI retrofit kit and created a sys-ex standard for it. There are lots of Chroma patches floating around on the Internet, and the Cherry Chroma can load them.
By right-clicking on any of the panel sliders or any of the parameter buttons that display numerical values, you can access MIDI Learn to assign the hardware slider or knob of your choice. This feature is not an innovation, but it adds to the expressive power of the instrument. A single slider can control multiple parameters, but there’s no provision for scaling or offset, such as you might find on a modern instrument.
The Chroma manual lacks a table of contents and section heads. I’m told this is because it’s an auto-generated PDF version of the online manual. If you consult the manual online rather than downloading it, you’ll always have the latest and greatest information, so this is not a significant issue.
Looks like I forgot to mention the arpeggiator. It’s okay.
True Colors. Generally I’m a fan of adding cool modern widgets to classic synths, and the more the better. (That’s why I liked the Cherry Audio PS-20 and was lukewarm on their PS-3300.) But I feel they made the right decision in keeping the original voice of the Chroma largely intact. It has a distinctive character that really evokes the sounds of the ’80s. As I checked out the presets I kept channeling riffs that would have worked in an LP by Gary Numan, New Order, or Depeche Mode. All I need are some LinnDrum samples, a clever haircut, and maybe a little black eyeliner and I’ll be ready for MTV.
I’ve always been fascinated by board games. Not sure why; it’s just the way my brain is wired up. I’m a lousy chess player, but I’ve invented a few chess variants. I wish I had a regular opponent to play against, but I don’t. Played a couple of variant games by email, many years ago. There are literally hundreds of chess variants, some of them brilliant and some of them very silly.
Go (the Japanese name for the game — the Korean name is baduk and the Chinese name is weiqi) is as complex as chess and has even more history. I learned to play go when I was a sophomore in high school. As with chess, I’ve never had a regular opponent or belonged to a go club, and that’s a shame.
Once in a while I’ve mused about variants for go, but the rules of the game are so very simple I never came up with anything that looked like a good game design. Last week, though, I ran across a YouTube video that covered some go variants. Way cool! I still don’t have an opponent to try playing any of them, but it’s clear there are possibilities I missed.
One easy variant is to play on a cylindrical board — that is, one whose left and right edges are conceptually joined. There are no corners on such a board, and there are just two sides. Corners are important in the opening of every go game, so this change throws the whole process up in the air. If you want to take it a step further, you can also “join” the top and bottom edges. At this point, the whole board is middle, with no corners or sides. Constructing a living shape may be quite a challenge! I think I’d want to try this on a 13×13 board rather than on the full 19×19 board.
Another simple variant goes by the Japanese term miai, a word that’s used when two points (that’s the go term for the intersections of lines on the board, nothing to do with scoring) are equivalent in some way. In miai go, in each turn each player selects two possible plays, and their opponent gets to select which one they like (presumably because it’s weaker or easier to defend against). The other stone goes back in the bowl.
The video suggested using a deck of cards with multi-stone shapes, let’s say of three stones. Draw a card and then play all three of the stones shown, retaining the shape but playing them in any available location. This set me thinking about a commercial game called Onitama. Onitama is a vaguely chess-related game in which cards dictate how the pieces can be moved. Each player has a “hand” of three cards. You choose one of your cards, move a piece according to one of the allowed moves on the card, and then pass the card to your opponent. The mechanic is a bit more complicated than that; you can look it up on YouTube if you’re interested.
It seems to me that playing go using a set of cards and the Onitama mechanic would be a brilliant possibility. A card might display something like this:
Shapes cannot be rotated or reflected; you have to do what’s shown. I’d suggest one supplemental rule: You don’t have to play all three stones if you already have stones in one or two of the positions shown. You can simply play the missing stone or stones to complete the shape shown in the diagram.
Now all I have to do is print out a set of cards and find someone to play against.
If you don’t have time to read That Book Is Dangerous by Adam Szetela, the video I’ve linked to below will give you the essentials. If you’re a writer, prepare to be pissed off. The title shot is definitely misleading, as there’s nothing in the video about sensitivity readers, except implicitly, and nothing about white male writers.
Szetela’s topic is a culture that has taken root in college courses on creative writing — a culture that is, frankly, destroying the ability of writers to write. Through peer pressure, students’ writing is being meticulously scrubbed of content that would challenge readers. Such content would not be considered “safe.” It would cause “harm.”
I’ve been aware of the absurdity of the “harm” concept since the incident seven or eight years ago in which a bunch of Unitarian Universalist (UU) ministers ganged up on a fellow minister, Rev. Todd Eklof, for daring to self-publish and distribute a book that questioned some of the shibboleths that “woke” UUs worship.
The other ministers published an open letter excoriating Eklof. In the letter they said they would name the harm that the book caused — yet they never named it! They never specified the harm that they alleged the book caused. This level of intellectual dishonesty would shame an eighth grader, but a bunch of ministers, all of whom, we must suppose, graduated from seminary, were unable to spot the problem.
If you’re writing popular genre fiction, you may never run afoul of the MFA culture police Szetela talks about in the book and video. At least, not directly. But your attempt to find a literary agent may fail because the agents who might otherwise like your book may be operating with the same culture police mindset. Even if they’re not, they may be using poorly paid interns to screen their massive slush pile, and those interns may be enrolled in or graduates of the type of insanely destructive MFA program Szetela describes.
Censorship is alive and well on college campuses. Students’ opportunities to learn are being sabotaged by the culture of safetyism. I don’t know whether this shit is happening in Europe, Asia, Africa, and Latin America, or whether it’s specifically a problem in the United States. Other nations probably have their own cultural biases. Here in the good old USA, we’re raising generations of young people who are, frankly, weak and stupid. What’s worse, they’re pugnaciously proud of being weak and stupid.
Just a word to the wise. If you’re an author, and specifically if you’re a self-publishing author, you need to watch the video I’ve linked to below. I’ve always self-published exclusively through Amazon, because it’s easy. Nonetheless, Amazon is evil. As the video explains, if you’re reading ebooks on your Kindle, you don’t actually own the book. You only own a license. And Amazon can, in their sole discretion, yank the license and delete the book from your device. Or edit the book without your knowledge or consent.
Will I continue to publish through Amazon? Probably. I’m lazy. I’d rather spend my time writing than wrangling with the details of electronic distribution. But I’m not going into it blind. You shouldn’t either.
After mulling it over, I thought it might be wise (and courteous to various authors and publishers) to buy paper copies of some of the books I have on Kindle. But I’m not sure how practical that would be. My Kindle library includes the collected works of Somerset Maugham, Nathaniel Hawthorne, Anthony Trollope, W. M. Thackeray, Sir Walter Scott, E. M. Forster, Edith Wharton, and Henry James. Do I really need any of those books?
On the other hand, I have Volume 1 of the collected fantasies of Clark Ashton Smith. Who? He was a contemporary of H. P. Lovecraft, and wrote in somewhat the same vein. Turns out there are five volumes of his stories. I’m not sure I would ever read them, as I’m not really into vintage horror, but it’s a temptation.
If you’re looking for a powerful percussion plugin at a reasonable price, you just found one. Triaz, from UK developers Wave Alchemy (wavealchemy.co.uk), is so good it’s scary. It combines a huge and versatile sound library with a brilliant step sequencer and a very respectable set of sound shaping parameters, all for £119.
Overview. The core library of samples that downloads with Triaz is about 5.7GB in size, and if the website is to be believed, it includes about 15,000 separate samples. (I didn’t count them.) These are organized into about 700 factory presets, each of which includes 12 sounds and a tasty sequence that plays them. Both the sounds and the sequences are completely editable, and not just editable: Triaz has lots of Randomize buttons, so if you aren’t sure what you want, you can just keep clicking until you hear something that floats your boat.
The plugin is called Triaz because a single sound can and often will include three separate samples, which can be blended. Each sample has its own envelope, filter, and other parameters, and like all of the other parameters, the blend can be modulated.
Some other percussion plugins just give you sounds, on the assumption that you’ll be sequencing the sounds in your DAW. You can certainly do that with Triaz if you prefer; each of the 12 sounds in a preset can be triggered via MIDI. However, Triaz’s own sequencer has 12 tracks, one for each sound, and each track provides control for note-on/velocity, repeat (ratcheting), the probability that a given note will play, the note onset time (which can be either early or late if desired), pitch, and swing amount. Beyond this, there are 12 automation tracks that can record what Wave Alchemy calls motion. The motion tracks don’t have to be assigned one-for-one with the sounds; if you want to assign several or all of the motion tracks to different parameters of a single sound, go right ahead.
Given the sheer complexity of this plugin, you might imagine the user interface would be intimidating. Wrong! I found it very easy indeed to navigate. I needed to check the manual only at one or two points. The PDF manual is not easy to find on the Wave Alchemy site; it’s there, but it’s hidden. A good video is also on the site. The video will walk you through many, though by no means all, of the features. I wouldn’t have minded more high-contrast lettering in the user interface, but I’m old and my eyesight isn’t good. The dark interface is probably perfect for folks who like to work in the studio late at night.
Triaz is available in the usual formats (VST3, AU, and AAX), and there’s a 14-day trial mode download that includes some, but not all, of the content.
Sounds. Each of the 12 sounds in a Triaz preset can contain up to three samples, as already noted. For each sample you can control parameters in three categories: Drum (which basically means sample playback), Pitch, and Filter. The Drum parameters are amplitude envelope attack, hold, and decay time; velocity response; the start point of playback within the sample data; and also drive, slop, and random panning amount. There’s also a reverse playback button. I’ve always detested random panning, but it’s there if you need it. If you dial up the slop amount, it will randomize the sample start time, volume, pitch, and filter cutoff by a little, or by more than a little.
In the Pitch panel are coarse and fine tune, an LFO with waveform choices and a rate knob, a pitch envelope with adjustable amount and attack and decay knobs, and a knob for routing some amount of velocity to pitch. Over on the Filter panel you can choose lowpass, bandpass, or highpass and dial in the cutoff frequency and resonance amount. Other than that, the filter parameters are the same as for pitch.
But that’s just the start of the fun. Each sound channel has its own EQ, compressor, and waveshaper. The EQ has an unusual design, which I’ll leave you to discover for yourself. Compression ratios go from 1.5:1 all the way up to brick-wall limiting. Useful features in the shaper include a stereo width control, which sounds terrific; the usual sample-rate and bit depth reduction; four different overdrive algorithms; and a subtle and very unusual knob labeled Repitch, which emulates the primitive pitch-shift method employed in the E-mu SP-1200 sample playback drum machine. If I remember correctly, the SP-1200 used drop-sample pitch changing (also known as zero-order interpolation), which added interesting harmonic content to cymbals when they were played at something other than the original pitch.
For send effects, Triaz has two reverbs and a delay. One of the reverbs is the familiar algorithmic type; the other is a convolution reverb, and naturally it ships with a variety of impulse responses. For master effects, Triaz has a four-band parametric EQ, a compressor, and a maximizer. If you need that vintage vibe, you’ll be happy to find that the maximizer can add vinyl turntable crackling or other types of background noise.
When you start designing your own sounds using the included samples, Triaz will try to find its sample packs in the default location. Because my C: drive is close to full, I had moved the huge Triaz sample and preset library to an external hard drive. Triaz had no trouble loading its own presets from that drive once I used the utility menu to tell it where to find the data, but in order to load single samples I also had to click the Refresh button in the sample browser. This button is not labeled, and the manual doesn’t mention my use case, so it was a bit confusing. When you want to add your own samples to the sample browser, clicking the Refresh button is how you do it. Alternatively, you can just drag and drop samples into Triaz from your computer desktop.
Speaking of the browser, it’s well organized. In the preset browser you can select a style from among 18 categories, including not just R&B, rap, tech house, and disco but cinematic and experimental. There’s even a selection for “top loops,” in case you already have something else playing the kick. Each preset is listed with its BPM, so you can find the items that are most likely to work. It’s also possible to lock the rhythm pattern while loading a new set of sounds, or vice versa. The sample browser provides buttons for 20 different sound types — for instance, both “kick electronic” and “kick acoustic,” not just “kick drum.” And yes, you can even click the favorite button for individual samples.
The icing on the cake is Wave Alchemy’s long list of add-on sample packs. I didn’t have any of these for review, so I can’t comment on them.
The Sequencer. You can make beats using Triaz’s samples by sequencing its drum sounds in your DAW, but having the beat stored as a user preset in Triaz is easier to manage, and Triaz’s sequencing features are quite robust.
There are 12 sequencer lanes, one for each sound. Individual lanes can be any number of steps from 4 to 32, and the playback rate of individual lanes is also adjustable, from 1/1 (a whole-note per step) up through 1/48 (sixteenth-note triplets). Within each lane you can control not only the note triggers with velocity, but also ratcheting, probability, early/late start time, swing amount, and pitch.
In addition, there are 12 motion lanes, which can be assigned to just about any parameter of any sound. This is a really strong feature.
At any given time Triaz can hold as many as 12 multi-track patterns, only one of which will be actively playing. You can switch patterns from your DAW using MIDI keys C6 through B6 (notes 96 to 107) in order to create a song form. The factory presets I looked at have only one pre-programmed pattern, so there’s plenty of room to try your own variations without messing anything up. Just copy the factory pattern to a new location and start clicking on stuff.
A handy Export button can be used to drag either patterns or the playback audio into your DAW. When you export a MIDI pattern, the motion-lane parameter changes will be exported as MIDI controller data. However, your DAW may not support drag-and-drop import of pattern controller data. I tested this feature in Reason, FL Studio, and Ableton Live. The notes themselves were exported correctly in all three DAWs, but only Ableton was able to correctly import the controller data into a MIDI clip. I had no luck at all exporting the MIDI CC values in patterns from Triaz into either Reason or FL Studio. But don’t despair: This is not as big a problem as it may seem. Any Triaz knob can be automated by drawing controller data directly into your DAW’s MIDI editor. Sixty-four channels for external automation are provided.
In a modular synth, I often set up step sequencing so that the number of notes is different from the number of modulation steps, so that the modulation (filter cutoff, for instance, or attack time) moves across the note sequence, resulting in a much longer combined pattern. Triaz won’t do this in its internal sequencer. Everything in the sequence for a given sound will have the same number of steps. This is a limitation, but not a huge one. If you want that kind of pattern shaping, you can do it in your DAW’s MIDI tracks. And just to be clear, each of the 12 sound tracks in Triaz can have its own length. If you choose prime number lengths for the various tracks (31, 29, 23, 19, and so on), a combined Triaz pattern will run for several years without repeating.
As a footnote, Triaz provides multiple audio outputs. Assuming your DAW can be set up to use them, each of the 12 percussion sounds can be sent to a separate stereo output for external processing and mixing.
The Wrap-Up. Triaz would be a good value at twice the price. At £119 it’s pretty much a no-brainer. If you make any kind of music that involves percussion patterns, I’ll go out on a limb and say Triaz may be the only drum plugin you’ll ever need.
Jim Aikin's Oblong Blob 