Yes, a CSV spline or node builder would be awesome. With a spreadsheet you could do all sorts of calculations and logic that can be more cumbersome in Audulus. Being able to translate that to splines would be awesome.
Thank you, I updated the video to include the link to your new repository.
Yes, JSON is the defacto storage method for app structures unlike xml which is very verbose, JSON is compact. I am definitely investing time into learning how to understand and manipulate JSON structures. Very glad you opened this thread.
@jjthrash definitely opened the door to being able to extend the capabilities of Audulus with his wavetable patch script.
One of my TODOs is to break the patch-building code out into its own library so you can do other things. Hopefully that will pave the way for making other interesting things. I’ve found it super handy for wiring the 8x8 Conway’s Game of Life patch, which has hundreds of connections.
Another TODO is possibly to document the JSON format, but that’s a little harder to do.
I’ve started writing up changes I want to make here: https://github.com/jjthrash/audulus_wave_table/issues
Feel free to add your own ideas and discuss them there.
Would this be a library of objects for coding in Ruby?
It can be. As I mentioned, I’d like to split that out, but for now since I haven’t, adding ideas for a general library would be great.
@jjthrash there’s a rubypico app for iOS but I must be doing something wrong as I can’t get new code to run.
Here’s the instructions for adding GitHub access to it.
I’ve created a GitHub repository for Audulus 3 SVG Icons and graphics issues. It includes some svg icons I’ve created and Audulus patches for them. Not all of the icons are useful as some are for exploring the issues surrounding the impact on performance with svg files in Audulus patches.
We’re fortunate to have an app developer in @taylor who can straddle both vector graphics and the music/DSP worlds so the issues I raise are directed at my fellow users who might not be as familiar with them and their effects plus for my own edification.
@robertsyrett mentioned in one of his videos that he might do something on vector graphics. I would definitely be interested in that. This winter I sewed some custom cases for my instruments. I like the idea of custom graphics for my favourite synth patches. I also mentioned in the AUv3 discussion thread that finding a way to encourage people to have some control over graphics without sacrificing the aesthetic theme of Audulus would be a very cool feature for the AU release.
I think the Audulus developers have to strike a balance between giving users access to being able to add vector graphics and any extra support headaches associated with doing so.
If some users have unrealistic expectations and find the process of custom icons too daunting or they create svg files that are too complex which leads to poor patch performance and disgruntled users who give bad App Store reviews, it will undermine their attempts to foster positive growth for Audulus.
Perhaps the easiest solution would be a separate app which could take users custom graphics and convert it using various templates into viable SVG graphics for Audulus?
This would facilitate ease of use, compatibility, independence, self expression, and optimize compatibility with Audulus.
The how long is the coast of England question is a good way to conceptualize the complexity and “size” of vector graphics.
One thing I notice with the custom built blocks in Reaktor, is I tend not to bother with the ugly ones (for better or worse). Ugly designs don’t seem to proliferate. It is like there is a natural curating process that happens, where the modules with the prettiest lights (and intelligent algorithms) seem to do well. On the other hand, it takes quite a bit of knowledge to produce a proper low CPU complex/useful module, so anyone that makes graphics for these already has enough good sense to produce a nice object.
GUI design and the inner workings of apps have equal weight now more than ever. On development teams there are frequently people who focus on one of these aspects of development exclusively.
Cool stuff people! I’m excited to see what everyone does with the sample node in Audulus 4 - you’ll be able to store data like with the spline node but do it much easier.
That said, one use that people haven’t talked about much so far is using the spline node for arp tables, and lookup tables in general?
You could make these little discrete steps that get referenced on
sample/totalSamples intervals. You could even use it to make a more efficient quantizer probably? Just spitballing.
John Conway’s Game of Life! I looked at the old forum posts you had on this and tried to find some of the patches in the zipped file of old forum patches, because I can’t download them directly off of the thread anymore. But I can’t find anything.
I got into Conway’s ideas through Manuel DeLanda’s Philosophy and Simulation: The Emergence of Synthetic Reason.
If anyone is interested, I would say DeLanda’s ideas are systematic. By that I mean to say that if you try to isolate a few of them and integrate their conclusions into your own thought, there will be definite limitations. However, over the course of his body of work, having addressed and overturned key ideas, a different way of dealing with categories or categorization unfolds. Vast possibilities open up and a path diverges from the stuck, received ontologies into what DeLanda and others refer to as a new materialism, through concepts like “multiplicity,” borrowed from Gilles DeLeuze.
Where did we get to with the Game of Life concept? Have we been able to seed an environment that evolves unique independent entities that interact in ways that codefine their identities, which, when taken together, produce multi-organized superstructures?
For grins, I’ve attached the 8x8 grid and the 32x32 grid, both assembled by a script.
conway64.audulus (1.0 MB)
conway1024.audulus.zip (2.6 MB)
I did a bunch of working trying to figure out ways to make these things useful, but it was a challenge. I built a 64-slot interlace node (like a sort of clocked mux node) inspired by @biminiroad’s work, but it’s still a challenge to do anything with it. It might be cool as a sort of procedural randomization source, or possibly a step sequencer.
Would it make sense to use these to break the signal out. I was able to generate something that is floating around – a shape is now moving about the grid on its own without “dying.”
This is pretty cool:
I guess the thing is you can generate patterns, or “life” out of certain initial conditions. So it is not just a “randomization” device.
It is news to me that, although Conway’s Game is a type of cellular automata, alot of the more grandiose colonies are the result of similar rule based grids such as “brain,” for example. Having a look at the video below, there are some inspiring ideas for sequencing.
I’ve never played with rubypico, but I suspect if anything is going to cause problems it’s the FFT library I’m using, which requires native compilation. I may take a look at a pure-ruby alternative, but it would require some rewriting. However, it only impacts the wavetable node, which I think maybe people aren’t using as much as the spline and midi node stuff.
This may be naive to ask, but I will anyway! Could this lead to a “scripted element” node in the Audulus UI that has a property that points to an external file (!!) or embedded code?
…and speaking of UIs, is the node or grouped element appearance saved in a manner similar to stylesheets? Any chance that could be brought out into the light?