I’ve been playing with the neuron addon a bit, tryin to make an Insaniquarium copy for fun, but i missed the beehave debugger… so i made a really junky version of it :).
Its bad and it needs a lot more work but i wanted to share the progress here
I’ve been playing with the neuron addon a bit, tryin to make an Insaniquarium copy for fun, but i missed the beehave debugger… so i made a really junky version of it :).
Its bad and it needs a lot more work but i wanted to share the progress here
Holy shit, what, this is insane, nice work.
A debugger is a great addition. Unfortunately, I don’t think of the neuron API as being particularly stable yet and I have a few changes I want to make to it right now, so I don’t think I’m ready to make a debugger for yet. Like, for example, I’m not terribly happy with how index works and I think it could be simplified or made more sensible somehow. When you get TL2 (and thus quota on the gitplace) I would love to check out the code ^^ (…or maybe I should just give everyone a small default quota)
Also, Insaniquarium!!! Wow, I forgot about that game. I used to play that a lot when I was younger.
thank you. cant wait to see what you’ll do!
btw, can i throw this as a public repo on github instead? i already my setup for that and for now its a bit more nice as I have my other projects there
well, if that includes the neuron library, i would prefer that the neuron library code stay off of github because i don’t want it to be scraped by an llm.
but… the license doesn’t forbid you from doing that and i can’t really stop anyone from uploading whatever they want to github or stop the llms from scraping data anywhere they can, so it’s probably fine.
ill keep it local for now, maybe ill make it a merge/pull request later
woah!! this looks awesome even in its early beginnings! ![]()
Oh that’s super cool!
I love that you can see which parts of the tree worked or not, that’s really useful.
As an aside, I think I just figured out how to fix the cumbersome index design in a nice elegant-ish way. Instead of maintaining an index, I think it would be more useful for the brain to track when each node was last executed. This allows me to maintain the design of keeping the nodes “state-less”. This also lets me implement a cooldown node, which is not currently possible within the current design.
I think I’ll play around with this idea a bit more, but I’d be happy to hear what you think.