Thank you for sharing. I particularly love Blender when it tells us something about life and its organisation; congratulations for combining art, science and technology in such a healthy environment. Let's hope that Blender Apps will soon come out to let you design your own interface. If you ever need somebody to develop an animated human embryo or several types of cells'inner organization... Do you use the simulation provided by the recording to test the validity of your settings and equations or do you aim to produce some kind of 3D version of the game of life?
I can't see why you'd want to include so much unnecessary detail in the simulation... it's going to be hugely expensive to simulate each cell's precise shape (and then remesh it for stability!!!) like that, even before you come to cell division... Wouldn't it be easier to have a primitive model of the shape and location of the cells and then a code step that converts each time step in the simple model into some mesh-based topology?
I could totally see how this could be used in scientific papers for example, as some kind of diagram. Sometimes you need a somehow accurate simulation, rather than a pretty and optimised one. I don't think these technique would be used in large productions as they are largely unoptimised, this is more likely to be used as some kind of visualisation.
Well, not being funny, but if you want an accurate simulation then you probably don't want to be mucking about in Blender with Python... C, maybe... or even running it on a GPU @@frzfrz6556
The cell's precise shape is one key trait determining its interactions at each time step, you can't simplify too much without losing explanatory power. The motivation here is plausible biological simulation for science, not convenience of visualization (while of course still having to work within what is computationally feasible).
Thank you for sharing. I particularly love Blender when it tells us something about life and its organisation; congratulations for combining art, science and technology in such a healthy environment. Let's hope that Blender Apps will soon come out to let you design your own interface. If you ever need somebody to develop an animated human embryo or several types of cells'inner organization... Do you use the simulation provided by the recording to test the validity of your settings and equations or do you aim to produce some kind of 3D version of the game of life?
I want to see this in Unity with DOTS!
Would be an awesome game!
Where is the addon
Awsome!
no
it's a great achievement but still not full potential is unleased when the add-on is downloaded into blender
good start of the lecture, and then it kinda trailed off to insanity; killer idea tho
I can't see why you'd want to include so much unnecessary detail in the simulation... it's going to be hugely expensive to simulate each cell's precise shape (and then remesh it for stability!!!) like that, even before you come to cell division...
Wouldn't it be easier to have a primitive model of the shape and location of the cells and then a code step that converts each time step in the simple model into some mesh-based topology?
I could totally see how this could be used in scientific papers for example, as some kind of diagram. Sometimes you need a somehow accurate simulation, rather than a pretty and optimised one. I don't think these technique would be used in large productions as they are largely unoptimised, this is more likely to be used as some kind of visualisation.
Well, not being funny, but if you want an accurate simulation then you probably don't want to be mucking about in Blender with Python... C, maybe... or even running it on a GPU @@frzfrz6556
Because it's obviously not for vfx
The cell's precise shape is one key trait determining its interactions at each time step, you can't simplify too much without losing explanatory power. The motivation here is plausible biological simulation for science, not convenience of visualization (while of course still having to work within what is computationally feasible).
meh, the biopy approach still fine
first