This is an example of how llm's break when you ask them "what should I build?" No current AI can provide good judgement.
To be clear: gaussian splat is very cool, but "in the terminal" is just nerdbait. The project includes some explanations, but given the llm source, I'm not sure it's accurate.
Here is a good video describing the tech in general, if you're wondering why there's so much buzz about it:
The interest isn't artificial; A lot of people (especially me) are just highly interested in this tech. I myself am working on a 2D variant meant for low-bandwidth environments to replace low-res pixel grid images.
You personally may not feel it, but there is a real, genuine interest, especially given how good it is at replicating photorealistic environments in realtime, relatively cheap (in comparison to NeRF). It's awesome.
Also, what would the ulterior motive be for someone to 'pump' this? There isn't a financial or clout-based motivation to do so. It's just interesting tech, which I believe we are all at least partly here for.
It's a really cool technique, and this particular article, LLM-written or not, is the best tutorial I've seen for it.
It's been decades at this point since I did any 3D graphics work, so I've completely lost touch with the field. I didn't feel like I was missing much until I saw the literal holy SHIT-class demo at https://superspl.at/scene/c67edb74 (which was also linked on HN a while back.) I don't care about terminal-based rendering specifically, but this page did a good job of bringing me up to speed on what a Gaussian-splat pipeline looks like and how it differs from polygon-based renderers.
Those who aren't interested can just move on to the next distraction, instead of trying to police a site they don't own.
Second commit is an ask to an llm “pickup things that go viral”. That is a weird thing to ask. :/
> Works over SSH. Literally the pitch: "view a 3D scene over SSH". Huge hook for HN.
https://github.com/darshanmakwana412/tsplat/commit/be98af413...
This is an example of how llm's break when you ask them "what should I build?" No current AI can provide good judgement.
To be clear: gaussian splat is very cool, but "in the terminal" is just nerdbait. The project includes some explanations, but given the llm source, I'm not sure it's accurate.
Here is a good video describing the tech in general, if you're wondering why there's so much buzz about it:
https://m.youtube.com/watch?v=X8yRlA7jqEQ
Who’s pumping all this gaussian splatting content onto hn, and why?
The interest isn't artificial; A lot of people (especially me) are just highly interested in this tech. I myself am working on a 2D variant meant for low-bandwidth environments to replace low-res pixel grid images.
You personally may not feel it, but there is a real, genuine interest, especially given how good it is at replicating photorealistic environments in realtime, relatively cheap (in comparison to NeRF). It's awesome.
Also, what would the ulterior motive be for someone to 'pump' this? There isn't a financial or clout-based motivation to do so. It's just interesting tech, which I believe we are all at least partly here for.
It's a really cool technique, and this particular article, LLM-written or not, is the best tutorial I've seen for it.
It's been decades at this point since I did any 3D graphics work, so I've completely lost touch with the field. I didn't feel like I was missing much until I saw the literal holy SHIT-class demo at https://superspl.at/scene/c67edb74 (which was also linked on HN a while back.) I don't care about terminal-based rendering specifically, but this page did a good job of bringing me up to speed on what a Gaussian-splat pipeline looks like and how it differs from polygon-based renderers.
Those who aren't interested can just move on to the next distraction, instead of trying to police a site they don't own.
huh