I find a form of active de-computing is one of the best ways to cope with the current exceeding world pain exhaustion & LLM brain drain around, i.e. doing more things offline, with your hands/body, being more present in your world, getting into crafts/materials, creating/fixing/mending things and sharing your lessons learned with others (in that sense quite in the spirit of open source culture).

To some this is may be just another form of entitled escapism, though I see it as active skill building and learning by doing, connecting with people who're still finding and creating (not just) beauty in the physical world, preserving/expanding older skills, methods & techniques, showing how to actively & creatively navigate, work with and improvise with the hard limitations imposed by chosen materials/resources, processes, situations or environments, and a form of creation which doesn't rely on exabytes of harvested data and other planet-scale resource guzzling infrastructure to realize one's ideas.

(Aside: I also think this is very much part of the appeal of "classic" [non-LLM based] generative/algorithmic design & art approaches, but also #RetroComputing and #FantasyConsoles...)

Anyhow, I think this stunning ceramics project/sculpture/vase is a great example...
https://www.youtube.com/watch?v=h6LJzTaQJx4

Of course, my generative artist/designer/programmer mind immediately goes into deconstruction mode, imagining how I'd approach something like this in code form (e.g. via volumetric modeling) and also how to manifest it physically (3d clay printing or maybe paper craft with a cricut[1]). I don't doubt that one day soonish a robot arm will be able to perform or exceed this kind of finesse, but I also think, this kind of automation just like with LLMs, is never the point of making art with your hands, other than making a point that it's now possible to have a machine create such detailed simulacra. It's the same one way road to Uncanny Valley as chasing 3D realism in game dev, with ever more resources required for ever diminishing returns... Maybe that in itself will become a useful measure/marker and status symbol for the trillionaire class.

Building open worlds of endless unspoiled forests made of ultra high-res scans in UE5, whilst the world is burning down though climate change and war. Always be chasing the end of the road, instead of enjoying the journey, the act itself. Only valuing final results without ever appreciating (or even caring about) the process to get there. Pure consumption...

Isn't it interesting, all the thoughts a "simple" ceramics sculpture can trigger?

#WeekendWriting #Craft #Art #ArtMaking #Ceramics #LearningByDoing

The hardware for my fantasy PowerPC console (that is how it's done, right?), renamed "Bobcat", has all arrived!

Now the only things I need are the JTAG debugger (this board has _no_ removable media or drive ports, only an EEPROM and some flash) and some motivation.

...and also lots of luck, because the last time I touched JTAG was 20 years ago, for a 68k assembly course. -_-;

#powerpc #fantasyconsoles

Today in roller coaster rides:

Down: I will likely never be able to emulate or build that freakish fantasy console I've been obsessing over because the hardware and code is ancient.

Up: Somebody built almost the same system _last year_.

Down: it's $1000.

Up: It's a whole-ass computer.

Down: it's a whole-ass Amiga with no GPU acceleration and limited FPU support.

Up: people are actively fixing all these issues!

This is a little exhausting, but fun!
#fantasyconsoles #a1222 #a1222plus

Galapagos 1 was envisioned as a full-on fantasy computer, which as I explored it more, started to feel more and more off-base, primarily because of the "terrarium" problem - to build a featureful computing environment you end up needing to build and package an entire jungle, which rapidly explodes into "rebuild the Microsoft Windows API stack".

The idea for Galapagos 2 would be to narrow things down to:

* A game engine that runs on Agon Light
* Prototyped on PC in Pygame
* On device is Agon Forth-based, super hackable
* Makes use of simple binary formats with similarly simple standalone tools - no singular IDE
* Multiple fixed graphics, physics and input kernels, static maximums on assets per scene
* Fixed, sample-based audio kernel (using the 4x2 channel tracker/DJ interface combo I came up with the other day, plus 4x channels for sound effects, for 12 total)
* Can work with assets on-device, PC development used as a supplemental bootstrap

And that's it! It creates a productive overlap of goals:
"portable, reproducible": satisfied by targeting a small but real device, with two implementations of the core code, one leveraging PC infra and the other machine-up
"extension": satisfied by being designed in a hackable way (Forth interpreter, simple formats)
"tooling": satisfied with fixed-function kernels and assets, narrowing the aspect of how to model data and present interfaces
"understandable": it won't need that much in the way of miraculous backend code or large build systems - instead it can be extensively documented

#retrocomputing #gamedev #fantasyconsoles

Seeing my game pop up on YouTube is such a cool feeling.

https://youtu.be/ijgHEzVanug?si=sFl4a6ZkMPxFx61F

#indiedev #indiegame #pico8 #fantasyconsoles

Is it possible to write programs for the Pico-8 in Assembly, C, or anything that isn't Lua? I haven't really found any good documentation on it so I have no clue if it's even possible.

#Pico8 #FantasyConsoles