Sunday, December 20, 2009

Catching some Z's

Z-DSP that is.

Over the last couple months, I've been ever so fortunate as to have been a beta tester for the TipTop Audio Z-DSP voltage controlled digital signal processor.  It was an extremely exciting prospect when I first heard about it; processing any modular signal, AC or DC, with a DSP and extending that with external voltage control.  It basically meant that almost anything would be possible with this module, even if in the beginning, we would only get a taste.  And what a sweet taste it was.




Our beta units came supplied with the Dragonfly Delay algorithm cartridge, which consists of 8 different delay algorithms; some mono-mono, mono-stereo, time-ratio'ed multi-tap, etc, each of which having their own unique set of 3 controls which can be controlled with the manual controls and CV inputs.  But it's just the algorithm that makes the Z-DSP special.  If the DSP where the engine, the rest of module would be the drive train, chassis, steering wheel, etc.

The Z-DSP, in addition to any done in the algo, has a stereo, patchable, analog feedback path, with independent CV control over the feedback amount for each channel.  This allows you to patch any external processing in the feedback path of the processing, in place of, or in addition to any feedback done in the algo.  It also allows you have a CV controlled input, if you can sacrifice the use of the feedback loop (rather than use the normal inputs, patch your signal to the feedback in inputs and use the feedback CV as an 'input VCA').  Or, use it as a 2nd set of stereo inputs (again, sacrificing the use of the external feedback).  The Z-DSP also has CV selection, as well as directional sequencing of the algorithms stored on a cartridge.  With a simple pulse, you can sequence forward through the algos.  With a simple pulse and gate, you can sequence forward and backward.  And with any arbitrary CV you can scan through the algos.  The sounds that can result from this may be subtle and intentional, or wild and chaotic.  A very unique approach to signal processing.  Equally, if not more, unique is the fact that the Z-DSP offers the ability to inject your own clock for the DSP and override its own.  Almost everyone these days is familiar with sample rate reduction, and bit-depth reduction, but actually reducing/varying the processor clock is an entirely different sonic result.  Not only can it achieve aliasing and "crunchiness" similar to that of sample rate/bit-depth reduction, but it can lead to insanely glitched out and wild digital instability.  Hell, it's possible to crash the DSP (of course, no damage is done in this case).

One might be able to argue "but I can do all this stuff with my computer, why would I want this in a module?".  Well, I know that one of the main reasons I enjoy modulars (and the Z-DSP) is because I have real-time, no latency, tactile control over the processing.  A physical, "real" interface to play with; wether it's manually twiddling knobs, or using control voltages to replace my hands, having that "real" interface allows me to sculpt what I want sonically much more easily.  I'm finding that, even with just the delay (which I use a lot in my compositions) I'm able to get a much better sense of what my final composition will sound like because I'm playing with the delay in real time, and not adding it later as an after though.  In this way, the delay becomes part of the composition, not just an effect.  For example, the following piece was written after a week with the Z-DSP.  It's not exactly a great demo of the Z-DSP per se because it's not really doing anything extreme; no algo sequencing, heavy CV (there is a little CV of the feedback loop amount), or crazy clock mangling, but, without the Z-DSP, this patch sounded lifeless and dull.

<a href="http://felixinferious.bandcamp.com/track/life-in-minor-preview">Life In Minor - Preview by Felix Inferious</a>

I am head over heels for the Z-DSP.  And I've only ever experienced the Drangonfly Delay cart.  I'm eagerly anticipating not only the Bat Filter cart, but also the various other magic that is being cooked up down at TipTop Audio.

Shooting blanks

I never gave much thought to blank panels for modulars.  It always seemed silly to me to buy a functionless panel just to cover up open space.  I understood the fact that there was the possibility of shock by touching power pins when the system is on, but I felt pretty safe in my ability to not do something like that.  So, I never gave blank panels a second thought.

Until a little over a month ago when I sold some modules to Stephen @ pro-modular.com and he told me about these laser engraved acrylic blanks that he had started to make and asked if I'd like one.  I ended up getting two sizes and in terms of color and what was engraved, I said, "surprise me".  And I was surprised.




I ended up with a 16HP clear one and the pictured 28HP black and gold one.  I'm still trying to get a really cool shot of the clear acrylic with some LED backlighting, but the black+gold really just jumps out.  The acrylic is a little less than 2x the thickness of a standard aluminum faceplate, so it's very sturdy, with very little flex even at 28HP.  I could easily see module panels being made the same way.

I've been super stoked on these blanks, especially the 28HP black+gold.  I was sad when I had to remove it while testing the Z-DSP as I didn't have room for both.  However, now that I've sold off a few more modules and made some room, it's back in the case!

Anyone looking for blanks (or even not looking for blanks), jump on over to pro-modular.com and check them out.  Full 19" rack blanks are now possible too!

Sunday, November 22, 2009

Back in the saddle

It's been 6 months since I last posted. I certainly seemed to have neglected things around here, but I'm hoping that will change. I've been spending much less time on the forum lately so I think it's necessary for me to at least continue contributing by way of the blog here.

So, what's happened over the last 6 months... It's hard to remember to be honest.
Buchla lust has come and gone again at least twice, the Harvestman Hertz Donut was finally released, to much jubilation by myself. I finally obtained a Flight of Harmony Plague Bearer (not sure why it took so long) and it's an amazing little module. If you like distortion and waveshaping, you owe it to yourself to try it out. There may have been something else before the HzD and the PB, but I'm drawing a blank. Budgets have been tight around the house here, so I've significantly cut back on expanding (that and I'm drawing near to filling up the Monster Case).

Speaking of filling up the Monster Case, I'm getting closer and closer to understanding the types of modules (and therefore the types of synthesis) that I am most drawn to. It's very clear that FM/AM and other audio rate modulation is key to the process; as is waveshaping of some kind. While I always love a good classic subtractive patch, they aren't the things that really drive my creativity. This is probably why the Hertz Donut has become such a favorite, and frequently-used, module since its arrival.
Low-pass gates seem to also be a strong standout over normal VCAs. The just-not-quite-subtle reduction in harmonics in concert with the reduction in amplitude (and vice-versa) really adds an sonic element that one normally finds with acoustic instruments. It feels like it adds a bit more "life" to the dynamics of the sound. The MakeNoise QMMG has been a fantastic addition here for that, as not only do you get 4 channels of LPGs, you also get a 4x1 mixer as well as LP and HP filters and 'normal' VCA operation. It's DC coupled as well, so it can be used for control voltages, however I must admit to never having exploited that bit of functionality. It sounds quite nice too, which is always a plus.
It's always been very clear that I prefer simple Attack-Decay envelopes over standard ADSR, and that has remained true. My Plan B M10s (I've had two for a while now) are definitely the go-to modules for that. The release of the MakeNoise MATHS module has got me very interested, but at this point I don't *need* more envelope generators, so for now I'm simply keeping a trained eye (and ear) on how folks are using it. I was able to play around shortly with a couple at the annual AHBA get together, and they certainly are very intriguing.
One of my favorite new (or rather, yet-to-be-released) module is the TipTop Audio Z-DSP. I'm cooking up a separate post for that, but to summarize, this will be an essential module in my setup going forward. I'm trying to convince myself that I don't need two.

On the creative side of things, the EP that I had planned for the end of the year has fallen behind. I'm still working on a couple of pieces, but I'm really trying to progress from what I did with Teratoma, so I'm being very careful and specific about which recordings I take and work on further into finished pieces. Speaking of Teratoma, I moved it off of VIRB (since they seemed to have removed the ability to download) and over to a Bandcamp page, where all my future releases will also be available. felixinferious.bandcamp.com/

That's it for now. More posts and more content coming soon.

Monday, May 11, 2009

Flame Clockwork Video Demos

On the unlikely chance that you haven't found these already through my twitter feed, the Muff forum, or even Matrixsynth, here are my latest video demos, featuring the Flame Clockwork Eurorack module.

Flame Clockwork Demo Pt. 1


Flame Clockwork Demo Pt. 2


Enjoy.

Wednesday, April 29, 2009

Monomachine Couples Counseling

As I mentioned back in September, I had traded both my MKI MachineDrum and MonoMachine for a MKII MachineDrum. Then, in December I decided to pickup a MonoMachine again, this time a MKII to match the MD. The initial reason was to try and fend off the severe Buchla lust I was enduring at that time (which I'm happy/sad/confused to say has not fully gone away).

But, in the months since then (particularly, the new year and onward), I had been once again using the MnM less and less, paying more attention to the modular than anything else and working on ideas for a new album. I was coming very close to selling it, I even made a half-hearted post on twitter about it, but there were no takers.

So there it sat on my desk, next to it's sibling, which does still get regular use, unlit and uninviting. It wasn't until I started thinking about putting together performance material that it crossed my mind to fire up the MnM again. I was again considering a small modular performance system, but I tend to like a lot of complex sounds, swirling in stereo, and I like the performance to be somewhat consistent from performance to performance, otherwise I get frustrated with it not sounding right. Both things that the MnM can handle...and there's the whole recall and syncing with the MD thing.

So, for the last couple of days, the MnM and I have been in "couples counseling". i call it that because that's sort of what it feels like. This on again off again relationship that I would really like to make work (then again, I've never been to couples counseling, so I really have no idea).

The first night was spent with just the manual. Just reading over the feature set and mentally re-aquainting myself with what it can do. I purposefully didn't want to sit down in front of it, tweaking randomly with an empty head and getting stuck in a "sound rut" like I had many times before with it. That approach seems to be no problem with the modular, but doesn't really work for me with the MnM for some reason. The manual quickly reminded me of the arpeggiator and more importantly, the independent trigger tracks and how para-locks work with "trigless trigs", a implementation concept which is fundamentally different from the MD in that you can set param-locks on steps which do not trigger any other aspect to the sound (PITCH, AMP ENV, FILTER ENV, or LFO TRIG).

Building ideas on these concepts *before* actually laying hands seemed to pay off quite well. The following evening I had a long session experimenting with sequencing out lots of param-locks with trigless trigs, separate FILTER and LFO trigs, and never actually laying down pitches; simply playing the sequence of "timbre automation" and holding down notes. It made some very nice drone-y noise-y stuff. I'd love to share, but I did this from the comfort of my couch, away from all the accoutrements of the studio.

Another concept that never initially crossed my mind before was using the arpeggiator to play the DBOX drum "machine" (basically the drum sample sound engine). While it was a little time consuming to search up and down the octaves of sounds (each octave of the keyboard is essentially a different sounding collection of sounds, grouped along the lines of a typical sample kit) to find a "chord" of the right samples for the arpeggiator, the results were very good, especially since the arp can run at different rates than the main sequence and can have shorter than 16 steps in length, including odd numbered steps. Using two different tracks with two different arpeggiators was that much more time consuming, but that much more rewarding as well.

Last evening was spent dealing with "the sound". I dare say I've become quite spoiled by the modular and that much more picky about the "quality" of the sound, and I certainly lean more analogue in my preference, but I certainly don't hate the sound of digital. But, many aspects of sound of the MnM just don't sit well with me and in writing down what those were, they were all ones that either attempt to sound analogue, or ones that I expected to sound analogue. Chief among these are the SWAVE machine and the filters. So, I decided to avoid those as much as possible (easy enough to avoid the SWAVE, not so much the filters) and focus specifically on the most digital aspects of this digital synth - the DPRO and the FM+. Now, the FM+ had long been a favorite since it offers pretty complex FM algos with a very simple (if sometimes unclear) set of controls, but I had not really explored the DPRO wavetable machine that much. Once again, diligence to constrain my scope paid off and I found myself getting sounds that I really liked and did not find myself relying on the filters to dynamically alter the sound and instead only found myself using the filters to shape it a little better, more like an EQ. Now, these most certainly did not resemble anything analogue, but it was digital being as digital as it can and I really liked that.

Now, it's not like I'm totally in love with the MnM again, but I have a new found appreciate for devoting time to it and getting better results from focusing on very specific features/functionality of the instrument. I think the MnM, moreso of the MD, requires more mastery of the device to really extract really unique and great stuff from it...at least for me anyway.

Another interesting rule I set for myself in this process was to NOT SAVE ANYTHING. I know, that's weird right, but I starting thinking about how I get in these "ruts" with this thing and a lot of these ruts start because I'm doing more or less the same process over and over again, and a lot of the time it's because I'm starting by revisiting something I worked on last time, when I stopped because I felt in a rut. It's also a practice that's not possible with the modular, and I wondered if the fact that you cannot save or fully and immediately recall your work on modular makes you think that much more creatively and retain more important bits about what you're doing as your explore the instrument. My early conclusions are that, yes it does.

One of the main downsides I'm still hung up on this point is the lack of a common reverb for all tracks. For example the MD has a global reverb and delay, and a reverb and delay send amount for each track. The MnM conversely has an independent delay for each track (which is fantastic for sound design reasons), but to get a reverb you purpose one of the six tracks as an FX track, and even in that case, it's not a send-return configuration like on the MD. You can route multiple tracks to the FX track running the reverb, but you do not have control over the volume level of the signal from each track going to the reverb, so there's no way to control the amount of reverb for each track. :( It's just as well I guess, upon actually recording it's likely one would just take each track out of the 6 independent mono outputs of the unit and mix in the box.

In the end (end of this post anyway, I'm certainly not through with this whole revisiting the monomachine process), I'm glad I didn't sell it and I've taken the time to explore it again. I may just have some Elektron-based live sets worked out after all.

Sunday, February 22, 2009

Mr. Felix Visits A Buchla

So after witnessing my endless back-and-forth Buchla lust over the last several months (and really, it's gone on longer than that), I was extended an invitation to visit the venerable Chris Muir and his Buchla for an afternoon of enlightenment.

Since the Buchla lust had kicked up again as of late, I was bit worried about being exposed directly to a Buchla, but at the same time, I'm practically broke, so I wasn't that worried that I would phone up Don in the morning to place an order. I was anticipating it though.

What I got was something else.

First, the system is as visually striking in person as it is in pictures, this I knew, but after Chris turned up the speakers, I heard something I wasn't expecting. There is definitely something lost when listening to MP3s and possibly even WAV files (I can't remember how many of the demos that I have listened to have been WAVs). There is much more dimension to the sound when you're listening to it through speakers and the details of the sound are much more complex. What came across as almost noise in some MP3s, was a complex harmonic evolution of sound in real life. I'm not trying to be over dramatic, I was just surprised at the very quantifiable difference in hearing the Buchla in real life.

As Chris gave me a tour around his system, it was very interesting to see how he utilized modules in ways that never crossed my mind in the hours I had spent staring at the Buchla web pages and module images. It was equally as interesting to find out that many aspects of the modules worked differently than I understood them to. One thing that excited me was learning that the 291e filter was able to sequence through it's various stages on its own, without an external clock pulse. Each stage of the 291e has a editable time constant which determines the time until it switches to the next stage.

Speaking of the 291e, it was quite impressive. It's definitely not designed to be used as a "standard" synth filter; it's practically half a synthesizer+sequencer all its own. While it's bandpass only, it sounds incredible. It's very clean, clear, and "vocal" sounding, and with some applied audio-rate modulation, has an incredible growl. Listening to Chris make a quick sequence of the filter's stages, with different amounts of FM per stage, and some external CV to the overall frequency, was just stunning. It was like an alien spaceship was somehow singing to me. Once he started using additional nodes of the filter (this was just using node A initially!) it became even more complex. Since the amplitude of each stage is also variable per-stage, it seems to me that a 291e paired up with some kind of source for filtering, is all you need for serious experimental composition.

When we starting talking about how Chris uses his setup again and I was surprised to find that he uses his 210e router almost exclusively as an attenuator for one of his 281e modules and as a audio matrix mixer for FX (delay, reverb, etc) sends. Chris has a basic "base patch" which he builds on and modifies slightly as necessary, but also makes frequent use of the preset abilities of the 200e. While it's always interesting too see how other people use a particular module, it was becoming extra interesting to see how Chris utilized his setup and how much it differed from how I imagined using a very similar assortment of modules. Previously, I believed the Buchla was tailored for a fairly specific use of the system, or at least more specific than say combining your favorite euro/frac/modcan-format modules. Listening to Chris describe how he works with his system really changed my perspective on that. I was clearly underestimating the Buchla.

We focused on the 256e next, which I found out was one of the few modules I had fully understood going in, but it was great to see how much, and to what extent, that Chris used it. Not only does it function as a way to alter and scale incoming CV (including inverting and scaled inversion), it also allows for mixing/cross-fading multiple CV signals as well as VCA for CV.

Over the next hour (or who knows how long it was, time was hard to keep track of), we bounced around to various aspects of the system, and finally settled into the 261e oscillators...or I should say Complex Waveform Generators. To call it an oscillator really is not doing the module justice. Being two oscillators, at least one VCA, multiple waveshapers all rolled into one, the thing is practically a voice all on it's own. While most modular folks balk at the fact that the waveform generation is digital, there are several advantages that come along with that. Chief of those for me was the fact that when Range is set to "Track", the modulation osc tuning determines a pitch ratio, which is maintained as you tune the principal osc. This was absolutely amazing to use in practice. I simple tuned the modulation osc to get the desired FM side bands, and then simply use the primary osc tune to control the overall pitch. Another interesting aspect of the 261e that cannot be discerned by staring at pictures is that there are several modulation modes possible. It's not just pitch, timbe, or amplitude modulation, but rather those individually plus combinations of all of them! Pitch + Timbre modulation was quite possibly a favorite, although amplitude modulation with independent control over Timbre and Symmetry was very cool too. Even more interesting is that the Pitch modulation is always done using sine waves, where as the other modes follow the waveshape setting of the modulation osc. This seems like a limitation, but being able to have Pitch+Timbre modulation, and vary the waveshape of the modulation osc and have that only affect the Timbre was fucking awesome. Limitation my ass! Speaking of which, the variable waveshaping didn't sound that bad at all! Based on various opinions that I'd read already, I expected an almost "unusable" PWM sound, but it wasn't that bad at all. It was no AFG, but it had a very unique and special sound. Along those same lines, the Hard Sync sound is not the ripping harmonic sound that one typically associates with Hard Sync, but it had it's own endearing character. The interesting thing was that Sync on it's own was not that impressive, but Sync mixed with FM+Timbre modulation made sounds like I had never heard before. Like circuits being tortured.

I knew I was going to like the 261e going in, but I had no idea to the extent at which I would love it. The waveshaping (timbre, symmetry, and high-order) combined with FM/AM is just absolutely incredible. The lack of additional "unwanted" side-bands that digital FM offers really brings an amazing pallete of sound to the table when paired up with waveshaping. I literally could have sat there for hours turning those 4 controls (modulation index, timbre, symmetry, and high-order). The ability to go from a simple sine wave to a sound that I can't even begin to describe was...well...fucking incredible! I was also surprised how much the controls interacted with each other. Depending on how much timbre and high-order I had adjusted, Symmetry would yield very different timbrel results. Likewise, depending on how much of the Modulation Index and Symmetry I had set, the Timbre would yield wildly different results. And, I'm no stranger to teaming up FM and waveshaping. Since I picked up the Doepfer A-137, I've been feeding it a steady diet of FM and while the results are fantastic as well, I've never had the control and range of sound with so few controls. I'm really starting to run out of words to explain how awesome I thought the 261e was...and this is 1,000x more than I said to Chris after turing the knobs for who knows how long (Chris, correct me if I'm wrong, but I think I simply said "Oh...wow!")

We wrapped up talking about various system configurations, thoughts on the 222e, various methods of control, portability, setup time, all sorts of little things.

At this point, you're probably thinking, "ok, so here's where he starts talking about selling all his euro modules and how the Buchla is the greatest thing ever". Oddly enough, I don't feel that way at all. I thought for sure that after visiting Chris, listening to him talk about his system, and getting quality some hands on time, that I would know, one way or the other, if a Buchla is in my future. Well, I don't.

I certainly know a lot more about the system now, and feel much more educated on its use and its strengths and weaknesses, particularly from an artist's perspective. I feel like I now can make a better, more educated, decision on wether or not to turn my attentions toward a Buchla, and I feel really great about that. I know that if I do decide to walk down that path, I know what to expect along the way. For now though, I'm going to stand here at the cross-roads thinking about both roads.

....

But... I do foresee a 261e and 291e in my future. ;)

Tuesday, January 27, 2009

NAMM 2009 - MaxMSP - Math

Once again, long time no post.

With my Serge lust more or less in check these days, I headed off to NAMM '09. I saw many cool things, particularly the new Plan B Zero cases (if you read this blog, you no doubt have already seen pictures), but I was surprised with what caught my eye the most.

The first was Max for Live. Most folks who pay attention to the Ableton community knew that something like this was coming after Ableton and Cycling '74 "joined forces" last year, however I didn't really expect such a simple, yet elegant implementation.

The other thing that really stuck we was the JazzMutant Lemur. I finally got to see one up close and personal and I was very impressed with what I saw. Not only the quality of the hardware, but with the programming environment and the new v2 software.

Lots of interesting new things, and the one thing that it all spurred was for me to take a good look at MaxMSP for once.

I used to own Reaktor and stumbled through that ok enough, but never really pleased with what I came up with. It was probably part lack of knowledge and part lack of direction, but it always seemed like the result wasn't worth the amount of time necessary to achieve that result. I had similar expectation with Max going in, and when I tried diving right in, deconstructing some of the example patchers, I was immediately confused. Rather than get discouraged, which I'm used to doing, I decided to take the less exciting route and go through the tutorials one by one. And boy am I glad I did.

The tutorials are very well thought out, don't move forward in too big of steps, and the overall documentation is excellent. Being able to select an object and hit a quick keycommand to pull up the full documentation on that object is wonderful. My favorite though is being able to just hit 'n' to start creating a new object...just start typing the name of the object that you want it will start auto-completing the various possible objects. Too slick! Max 5 is looking great too, I'm a bit of a picky bastard when it comes to UI and the older generations were always a bit of a turn off.

Another interesting turn is that I've become more interested in the math behind digital synthesis. I was never very good at math, although it wasn't do so much to lack of skill as it was a lack of motivation. Like many more in my generation, I suffer from a severe lack of motivation for things I'm not interested in, and during high school and college, I couldn't have been less interested in math. I could never see a practical use for what I was learning, and being a total smart ass, I would pose that question to my math teachers frequently. I really did want to know though, and I never did get a good answer. Always the "you need to know it" or the slightly more specific "you need to know it for calculus". The last math I actually remember studying was quadratic equations and polynomials. What do know, that's the basis behind much of digital synthesis, particularly waveshaping and filter design. The last few days have been filled with wonderful exploration into the world of Chebyshev polynomials, and their very pleasing results on sine waves.

So, where am I going with this? It's a basic plea to math teachers, should any read this blog (or even if you know some math teachers). PLEASE, please, tell your students some interesting or at least practical uses for the math that you are teaching them. If some student asks "what do we need to know this for", they may not be trying to be a smart ass, they very well would just like to know. I really wish someone had done this for me during my education. I might have come to really enjoy and like math, rather than hate it and, later in life regret, not putting in more effort.

Oh well.

PS. I have a few euro modules for sale currently, if any of you are interested:
http://www.muffwiggler.com/forum/viewtopic.php?t=2935

Cheers everyone and Happy New Year!