Creative Computing Major Project - Semester 2, 2007
Creative Computing Major Project - Semester 2, 2007
The main interface..
The main interface..Well folks, here it is - The "Aqua Tower". Welcome to yet another revolution in sight and sound from DJD enterprises. I have included my analysis in the body of this blog, so there is no need to download it. The other documentation and audio of my performance with the Tower, can be accessed via the link at the bottom of this entry.
The mixing window..
Composition Analysis
Name David J Dowling
Course Creative Computing IIB
Initially this was to be a combination of five samplers, all in sync and cross-accessible on the fly. Due to programming difficulties and time constraints, I have had to settle for one but shall expand the MSP patch at a later date. That being said, I believe the restriction has proven itself worthwhile, as there was only one enormous set of parameters for me to control during a performance.
The Granulator control station..
The objective behind my program is to enable quick access to samples for manipulation via granular synthesis and standard effects such as delay/flanging. The interface is detailed, but straightforward and easy to use. I have used a theme of deep sea diving for the motivation behind my piece, as it seems to fit with the visual presentation of my program.
The Delay Controller..
The performance itself is of electro acoustic quality, with an emphasis on tension and release from intensity of sound, rather than utilising conventional tonal methods. The source of audio for the sampler is a batch of guitar tones that were recorded in Studio one and EMU space during the semester. I have chosen to use distorted or ‘dirty’ sounding tones and chords, as I have heard a lot of electro acoustic music that utilises clean guitar signals for audio (even made some myself last year), and I wanted to try something a little different.
The Autopan Controller..
Although there were a couple of ‘licks’ thrown into the recording batch, there is little in the way of conventional sounding guitar music of any kind in the final product. This was my intention all along, and even if it was not, the MSP patch I have created demands a lot of attention for controlling its numerous parameters, leaving little time for non-electronic experimentation.
The Speed Ramp Controller..
While Max/MSP is a wonderfully diverse and intuitive tool for software design, it can easily become too complicated for the user to control or understand. A problem one faces when building a program for creative purposes, is to continually invent and add new features that were never intended initially, simply because you can. At its stage of completion, my interface, when unlocked, reveals a bit of a rats nest of cables, which I ran out of time for organising. The bulk of the mess is from little things too, like loadbangs, keyboard shortcuts, and menu item access. I did my best to make the audio signal easy to follow.
The Flanger Controller..
Despite these confusing issues the program works, and the real task ahead is to become fluent in using the interface. Even though I have written all the shortcuts and tricks of the program, I have yet to produce a satisfying five-minute performance from the unit. As we all discovered during the hardware instrument building and improvising phase of this year’s forum, it is one thing to make the instrument, another thing entirely to play it.
Click here to link to online folder containing other documentation in PDF format and an MP3 of my performance.
Reference:
Haines, Christian. "Creative Computing 2 – Music and Sound Processing – Electroacoustic Performance." Lectures and tuition undertaken at Level 4 and Level 5, Schultz building, University of Adelaide, October/November 2007.
The mixing window..Composition Analysis
Name David J Dowling
Course Creative Computing IIB
Initially this was to be a combination of five samplers, all in sync and cross-accessible on the fly. Due to programming difficulties and time constraints, I have had to settle for one but shall expand the MSP patch at a later date. That being said, I believe the restriction has proven itself worthwhile, as there was only one enormous set of parameters for me to control during a performance.
The Granulator control station..The objective behind my program is to enable quick access to samples for manipulation via granular synthesis and standard effects such as delay/flanging. The interface is detailed, but straightforward and easy to use. I have used a theme of deep sea diving for the motivation behind my piece, as it seems to fit with the visual presentation of my program.
The Delay Controller..
The performance itself is of electro acoustic quality, with an emphasis on tension and release from intensity of sound, rather than utilising conventional tonal methods. The source of audio for the sampler is a batch of guitar tones that were recorded in Studio one and EMU space during the semester. I have chosen to use distorted or ‘dirty’ sounding tones and chords, as I have heard a lot of electro acoustic music that utilises clean guitar signals for audio (even made some myself last year), and I wanted to try something a little different.
The Autopan Controller..
Although there were a couple of ‘licks’ thrown into the recording batch, there is little in the way of conventional sounding guitar music of any kind in the final product. This was my intention all along, and even if it was not, the MSP patch I have created demands a lot of attention for controlling its numerous parameters, leaving little time for non-electronic experimentation.
The Speed Ramp Controller..
While Max/MSP is a wonderfully diverse and intuitive tool for software design, it can easily become too complicated for the user to control or understand. A problem one faces when building a program for creative purposes, is to continually invent and add new features that were never intended initially, simply because you can. At its stage of completion, my interface, when unlocked, reveals a bit of a rats nest of cables, which I ran out of time for organising. The bulk of the mess is from little things too, like loadbangs, keyboard shortcuts, and menu item access. I did my best to make the audio signal easy to follow.
The Flanger Controller..
Despite these confusing issues the program works, and the real task ahead is to become fluent in using the interface. Even though I have written all the shortcuts and tricks of the program, I have yet to produce a satisfying five-minute performance from the unit. As we all discovered during the hardware instrument building and improvising phase of this year’s forum, it is one thing to make the instrument, another thing entirely to play it.
Click here to link to online folder containing other documentation in PDF format and an MP3 of my performance.
Reference:
Haines, Christian. "Creative Computing 2 – Music and Sound Processing – Electroacoustic Performance." Lectures and tuition undertaken at Level 4 and Level 5, Schultz building, University of Adelaide, October/November 2007.
posted by David J Dowling @ 6:57 PM
6 Comments:
Quote from AquaTower dialog: "David J Dowling is a music technology super genius who also holds a 500 dan extra dark black belt in 12 karates so don't crack his software". Thanks for filling me in on that one!
I checked out your AquaTower and "Art of Diving" piece and enjoyed your distorted rhythmic figures in the piece, which provided an interesting backdrop for the soundscape material. I understand better now what you meant when describing that you were still quite "new" to your instrument (as in the application) and so the performance was maybe not representative of the AquaTower's capability. I thought I was going to hear some full-on dive bomb activity, or was that for your PIMT composition?
When running the AquaTower on the PC, I heard some pretty interesting textures being created. I ran my "Chromos" piece through the granulator and was most impressed with the result. You've clearly put a great deal of effort into implementing the DSP features of this application (ie. effects) not to mention the interface. In some ways the AquaTower is the anti-thesis to my ambienceFACTOR program, in that it presented me with a vast amount of options, rather than being "locked" into a specific mode. That said, given the many features you've incorporated, I wanted to record the output of what was happening, but found no WAVE/AIFF recorder option, although I'm sure you've thought of that, just a time issue there.
A few other thoughts - the application interface consists of a large number of bpatchers and jsui objects, and I'm not sure if this is an issue, but just out of interest, after I booted AquaTower, the Max Runtime started in around 3 seconds as per usual, but your application took nearly 35 seconds (the MSP Patch is the same). I assume this is due to the graphics-load. Although I used fpic and pictctrl objects for my patch, I guess there were fewer of them (excepting all the LEDs) but I'm not sure if this affects application snappiness for better or worse. Just an observation.
Aye, as bpatchers are my tool of choice for reducing clutter in a patch (rather than clever and concise programming) I suspect each one you introduce adds a lot more code for the program to process. It got very painful to work with at home, as the PC was taking around four times as long to do anything as the G5’s were. I’m interested as to how the ambienceFactor performed on your home PC when it was open for scrutiny. When I opened it up (the patch that is – not the App, which worked fine by the way) on my PC it was an equally painful wait for mouse clicks and other simple commands to register, as all of those signal connections were hogging the CPU…
I don’t know what the answer is to any of these issues when it comes to Max / MSP programming, maybe it’s just to wait until computers are so fast that they just deal with it, but I suspect the nature of Max and the way it implements its language is always going to be heavy on the CPU once things start getting complex. All the more reason to get stuck into Supercollider and keep learning lower level languages, the trade off being that these tools are heavy on one’s brain, rather than the computer processor. That being said though, I’ve spent just as much time staring blankly at Max as I have at Java, waiting for the right approach to a problem to present itself to me, so maybe I should stick with the program that gives the most visual feedback… Meh, computers blow, I’m goin’ shredd’n’.
Yeah I have to admit I get a bit lazy in that I tend to avoid bpatchers and sub-patches...I like it all to be there right in front of me at a glance. Nonetheless, I ran out of screen real estate to include comments about objects etc.
Typically, my ambienceFACTOR patch or app loads in around 10 seconds on my PC, being an older Pentium IV (slower than what you are using). And on the G5s, about 5-6 seconds wait, so your experience with it is just another of life's mysteries I guess.
Interesting comments regarding SuperCollider, and I agree totally. It obviously lends itself toward optimisation that might be less achievable with Max or other visually controlled environments.
Perhaps it would be worth running our patches under the MaxMSP 5 demo, when it is released, to see if the code has been better optimised, or contains CPU optimisation instructions (MME, SSE2, 3DNow etc).
It's a nice feeling knowing I don't have any more Max work in the immediate future. Would go grab a guitar also, but I don't have one. Yet.
dave-o,
i have to congratulate you on completing a project on an enormous scale, this one really is a monster! had a lot of fun playing around with it - sounding very autechre there bro - top marks.
Cheers bro, that must mean it runs okay on the Macbook..
Cool stuff Davo. Haven't had a play yet. If it runs crap on your computer.. then I doubt there is much hope for mine. The piece was pretty cool though.
I agree with your comment on supercollider though. I'm looking forward to being able to do things more directly.
I'm also glad to here someone else spends as much time as me staring blankly at the max screen waiting for the answer to come.
Post a Comment
<< Home