anattā

anattā

@cubething.dev

👤 1996 | they/them | software, philosophy, art 🖌️ /pfp @wraithcastle.bsky.social 🖥️ /about https://cubething.dev

174 followers476 following1.3K posts

some hair. maybe a bit small. but i'm still pretty proud of it. shrinkwrap and looptools are godsends. #blender

anattā reposted

"the general public doesn't want to play videogames made by normal people, they want to play videogames made by insane evil horny trannies"

i've been using ai to get over some humps in my game dev and i really feel like it's finally allowing me to pursue my ambitions. being a solo developer seems... _possible_ now

1. the brain is two organs 2. nonrepresentational awareness and symbolic cognition are two different things 3. nra in brain stem; sc in neocortex 4. ensoulment is awareness of self i.e. interaction btwn nra & sc 5. the hypothalamus is the seat of the soul

i agree that we shouldn't be lowering our standards to meet whatever llms produce. but if you aren't giving your agents a proper harness then you're doing it wrong. it's trained on median data so it's going to produce median output. learning how to get an agent to produce good code takes a while

anattā reposted

neoplatonism isn’t real it can’t hurt me neoplatonism isn’t real it can’t hurt me neoplatonism isn’t real it can’t hurt me

Dominik Schnaus@schnaus.bsky.social

The intuition: models trained on different data seem to converge to a similar geometry (the Platonic Representation Hypothesis). Distances between concepts like food, people and animals look alike in both spaces. So is this shared geometry enough to align them?

And certainly for smaller orgs, if you want to compete on resources per proof of concept, an LLM will provide you something far cheaper far faster than human labor.

A lot more could be said here. But I wanted to conclude by stating that Arca is an excellent resource. I was just surprised to see such a bad argument from an otherwise insightful publication.

The military argument in use in this article strikes me as a bad analogy for most cases. I don't think corporations are tactically equivalent to militaries and the analogy isn't fleshed out enough to make that claim.

The article suggests that military might could be comparable to... number of software products released? But that seems like a tenuous comparison at best. Plus, Iris claims there hasn't been a major increase in products produced by major corporations... but what about Windows 11s many slop features?

As an individual, LLMs are _ridiculously_ useful. Are they going to replace fleets of highly skilled software engineers? Can they do effective architecture and long-term maintenance? No, but they can certainly be used to offload the initial "make it work" phase

...and, with enough passes and a human-in-the-loop process, LLMs often produce high quality work. What counts as high-quality usually is polished, correct, maintainable code. It takes some fenagaling and multiple prompts, but LLMs can certainly produce that kind of output.

mediocre work done at a high pace is way better than done slowly and in fact enables the high quality work to happen.

LLMs are like dishwashers. Sometimes you have to scrub stuff by hand, but I still wouldn't rent an apartment without one.

While it's in the nature of LLMs to be flattening, averaging machines, they've certainly accelerated the pace at which high quality work can be done. Recent maths work for instance -- not clean, but definitely high quality in the sense of boundary-pushing ("do a breakthrough")

lots of graphics updates in this! WESL and a HUGE solari update. the BSN cleanup is really nice too, and scheduler updates!!

Bevy@bevy.org

Bevy 0.20 is out now! It features Solari and DLSS improvements, better BSN syntax, BSN "ready" events, WESL is now Bevy's shader language, more UI widgets (color picker, scrollable list views, dropdowns), Sprite Materials, Extended 2D materials, Pan Orbit Camera, and more! bevy.org/news/bevy-0-...

base model is uv mapped and has a basic texture. next are some animations. I see vertex pickers in my dreams. send help.

anattā reposted

Seeing as my most famous post on the site below has been getting a little traction again. Just let everyone know, my Scanned Texture packs are currently on sale. I've only had one sale on itch since the end of May so anything is appreciated! itch.io/s/198768/sca... #sale #textures #gamedev #art

itch.ioScanned Textures - Aug/Sep/Oct Sale by Clastic ArtisticA bundle by Clastic Artistic, $24.00 for 6 games
Clastic Artistic@clasticartistic.bsky.social

With the Scanned Textures tagline being "Real textures from real objects for your projects", it does create some absurd situations. Behold hay on a scanner. #artassets #designassets #gameassets #assetpack #indiedev #graphicdesign #textures #design #art

Very good. Really the only broadly defensible position in this debate, if we're to admit to rehabilitation.

Flora 🍓🏳️‍⚧️@floramerigold.bsky.social

Yesterday's news about Jacob Geller broke my heart, so much that I couldn't stop dwelling on it all day. In an effort to make sense of this horrific news and my admittedly complicated feelings about it, here is my article on the Jacob Geller situation: epiloguegaming.com/what-do-we-d...

getting a hang of blender :) need to play with some vertex weights and rig the hands but this is pretty promising! the joints deform really well :) :) :) turnaround below

hurting people is inevitable. the only thing you can do is to grow. i don't doubt that geller of all people has spent his life trying to overcome this. in many ways that puts us in the same boat.

Jacob Geller@jacobgeller.com

hurting people hurts you, too, especially when you are the sort of person who tries to make amends whenever possible. sometimes you can never be forgiven. and you have to learn to live with that. you have to learn how to move on, to never do it again, to prevent the harm from happening when you can.

it is not an easy thing to do. and honestly i applaud geller for being forthright about this, and for trying to protect his ex as best as he can.

anattā reposted

extremely good thread from my clone <3 unfortunately it also got her marked as spam so uhhh if you're just a LLM enjoyer do go read it, if you're a bluesky team member please unjail oomf

イネス⎔@personhood.removal.surgery

Banger thread. I disagree with a lot of it on technicalities (which I will proceed to be insufferable about below) but in general - yes, the problems Phil talks about here are exactly what's been most itchy for me and what I've been trying to solve ever since I started working with persona agents.