It really makes the holodeck seem WAY more amazing when you think about it. Just being able to build a cohesive environment that exists in euclidean space, from a vague vocal prompt, is an achievement.
IRC I think it’s only the holodecks on the Enterprise D that were advanced enough to create environments from scratch in real time using the ship’s computer. Other holodecks we’ve seen only had saved presets and the ones on DS9 had to be loaded with holodeck programs that were created elsewhere by programmers.
I’ve been rewatching DS9, and the way that the baseball holodeck program is described gave me this impression too. It makes me smile to imagine Sisko iterating towards the end goal, with his dumbass nerd grin on his face
Some of the results are pretty crazy detailed based on a simple prompt. But even something simple like “person walking out of a room” and the image is the person walking into a room…that’s what makes it frustrating. Even simple thing like “John is wearing a purple shirt and brown shorts. Sally is wearing a green shirt and grey skirt.”. Yet half the results are John is wearing a purple skirt and a gray shirt 🤷♂️
Trying to use an AI image generator it still blows my mind how difficult the simplest of prompts are.
It really makes the holodeck seem WAY more amazing when you think about it. Just being able to build a cohesive environment that exists in euclidean space, from a vague vocal prompt, is an achievement.
IRC I think it’s only the holodecks on the Enterprise D that were advanced enough to create environments from scratch in real time using the ship’s computer. Other holodecks we’ve seen only had saved presets and the ones on DS9 had to be loaded with holodeck programs that were created elsewhere by programmers.
From my understanding you could go into any holodeck and begin creating a world through prompts, but it could take some time to get things just right.
Or you could aquire a pre-made program for a curated experience.
Turn out, all the cheap holodeck stuff is just cobbled together asset flips and lazy prompts
I’ve been rewatching DS9, and the way that the baseball holodeck program is described gave me this impression too. It makes me smile to imagine Sisko iterating towards the end goal, with his dumbass nerd grin on his face
Same with LLMs. I hate how you have to constantly refine your prompts to get it to answer even a simple question sometimes.
Some of the results are pretty crazy detailed based on a simple prompt. But even something simple like “person walking out of a room” and the image is the person walking into a room…that’s what makes it frustrating. Even simple thing like “John is wearing a purple shirt and brown shorts. Sally is wearing a green shirt and grey skirt.”. Yet half the results are John is wearing a purple skirt and a gray shirt 🤷♂️