81–90 of 132 posts
Sadly it doesn't seem to generate UVs for the models. But using the cube projection in Blender seems to do an OK job. Add a skybox and a couple of simple particle effects in PlayCanvas. And we have some WebGL spaceships :-) https://playcanv.as/p/kZtPZpnH/
I don't really like most procedural generation. It has no meaning and the results are not intellectually stimulating. At best you can't spot the pattern and parameters but you usually can after a few examples. An idea I am more interested in is that you generate requirements and use an optimiser to solve the actual design. This way, there is a hidden "why". With some study, a human might be able to discern why x is s…
I don't really like most procedural generation. It has no meaning and the results are not intellectually stimulating. At best you can't spot the pattern and parameters but you usually can after a few examples. An idea I am more interested in is that you generate requirements and use an optimiser to solve the actual design. This way, there is a hidden "why". With some study, a human might be able to discern why x is s…
The selected examples got me thinking: Those eight are presumably examples the author felt turned out particularly well. Could one build a neural network trained on generated ships that "turned out well" to automatically generate better-looking random ships? Could this sort of process be used in games where procedural generation might otherwise be rejected because it looks "too random"?
Short answer: yes.
Long answer: yes, but you'd have to retrain it for each style/form/concept that "well" was defined for.
But yes, it's possible. Would be a lot of work though.
Sadly it doesn't seem to generate UVs for the models. But using the cube projection in Blender seems to do an OK job. Add a skybox and a couple of simple particle effects in PlayCanvas. And we have some WebGL spaceships :-) https://playcanv.as/p/kZtPZpnH/
Pardon my ignorance, but what are UVs?
Sadly it doesn't seem to generate UVs for the models. But using the cube projection in Blender seems to do an OK job. Add a skybox and a couple of simple particle effects in PlayCanvas. And we have some WebGL spaceships :-) https://playcanv.as/p/kZtPZpnH/
Pardon my ignorance, but what are UVs?
Though daredevildave may have meant environment mapping instead.
Sadly it doesn't seem to generate UVs for the models. But using the cube projection in Blender seems to do an OK job. Add a skybox and a couple of simple particle effects in PlayCanvas. And we have some WebGL spaceships :-) https://playcanv.as/p/kZtPZpnH/
My physics knowledge is pretty weak. Wouldn't you want a space ship to be closer to something like an oblate spheroid? Less surface area volume ratio builds cheaper, lighter space ships presumably? The primary thing I'm unsure of is steering, but how much of a problem can that be?
You probably would have boxy starships for the same reason you have boxy houses: more usable space. The same arguments apply for spherish houses, but those never caught on. Given the distance that the starship is going to travel, the main cost is probably not the materials so much as the time, so you'd want to maximize the amount of stuff you could ship back and forth.
Sadly it doesn't seem to generate UVs for the models. But using the cube projection in Blender seems to do an OK job. Add a skybox and a couple of simple particle effects in PlayCanvas. And we have some WebGL spaceships :-) https://playcanv.as/p/kZtPZpnH/
Holy crap that site basically stops my computer.
Sadly it doesn't seem to generate UVs for the models. But using the cube projection in Blender seems to do an OK job. Add a skybox and a couple of simple particle effects in PlayCanvas. And we have some WebGL spaceships :-) https://playcanv.as/p/kZtPZpnH/
Very cool -- what about sliders to feed parameters for new ship generation :)