I learned to reduce light bounces total from 12 to 3 max, reduces drastically the time to render glass and refractions/reflections in general, I haven’t seen any difference besides it being 100 times faster
this is awesome. Can you throw a tutorial please im trying to understand the thought process behind node tree, copy paste is easy but understanding your vision to set it up is important.
Thank you for the response. I meant the method to the maths behind nodes for movement of the spheres based on radius, mass and probably some more things i'm unable to understand. I followed and remade the whole thing but im still quite unclear. However i wanna learn how i can make something similar in future, maybe a simulation with more bodies or a solar system as it also involves multiple bodies.
Oh. I studied physics and simulations in university. I don't really.know if I can easily explain it in a reddit post. But in short, I used the gravity equation to figure out the acceleration.
Let's say you're designing a transmission system for a car. How do you know what kind of conditions it will survive? How do you know how long it will last? You could make a prototype and test it until it breaks, but that's incredibly expensive. What you would do is model it with CAD software and simulate it using FEA. Almost all highly technical physical products these days are designed with simulation software. You want to know how your product performs before you spend a lot of money on a prototype. FEA is used to test stresses, forces on objects, etc. CFD is used for fluids, airflow, etc.
Ah okay. Thanks for explaining. So is this generally a one-man job? Like companies don't seek "simulators" to help designers, they hire designers with the ability to simulate themselves? Also, is this still how it works for simulating things like chemistry or astrophysics?
I'm sure every company is different, but yeah, usually companies hire an engineer since they are experienced in design and simulation. I don't know much about chemistry, but astrophysics, yes. That's what's known as aerospace engineering.
Another option could be solver iterations or biases. Let's take divergence for example. If you're enforcing incompressibility from the top of the screen to the bottom, cell by cell, your result will be skewed if you don't have enough solver iterations. It will also be skewed if you have biases in your equations. Hope this helps.
You're probably confused because your transparent is probably at 1 and yet it's not transparent. That's because if you have a metallic value that isn't 0, your material will lose its transparency. Metal isn't transparent. Turn your metallic down to 0.
It's really just a particle sim with a lot of particles, then I used the lifetime and the frame count together with a color ramp and some math to make the colors.
Honestly, this is a very simple particle sim. The only thing that's cool about it is that there are a lot of particles. No tutorial, but here's a video that heavily inspired the render.
There's a fluid shader here on Gumroad, available for free. It should help improve the greyness of your fluid. You can also use it on the glass and ice.
Part of the reason is probably because this is actually in 3D, so it kinda looks weird if you're expecting it to be 2D. But yeah, I can agree with those other points. In simulation nodes, the time step is locked to the frame rate which is annoying, but I'll admit that I only used a first order divided difference scheme. It's my first time doing simulations in simulation nodes, so I just wanted it to be as simple as possible so I could learn the workflow. Thanks for the feedback though, I'll keep that in mind for future simulations.
I learned to reduce light bounces total from 12 to 3 max, reduces drastically the time to render glass and refractions/reflections in general, I haven’t seen any difference besides it being 100 times faster
It was already reduced.
this is awesome. Can you throw a tutorial please im trying to understand the thought process behind node tree, copy paste is easy but understanding your vision to set it up is important.
I don't know of any tutorials for this. No method. Just trial and error, but not massively. It takes a few tries.
Thank you for the response. I meant the method to the maths behind nodes for movement of the spheres based on radius, mass and probably some more things i'm unable to understand. I followed and remade the whole thing but im still quite unclear. However i wanna learn how i can make something similar in future, maybe a simulation with more bodies or a solar system as it also involves multiple bodies.
Oh. I studied physics and simulations in university. I don't really.know if I can easily explain it in a reddit post. But in short, I used the gravity equation to figure out the acceleration.
Would you be able to explain exactly what you mean? I'm not familar with that title in this context. How exactly does it connect to product design?
Let's say you're designing a transmission system for a car. How do you know what kind of conditions it will survive? How do you know how long it will last? You could make a prototype and test it until it breaks, but that's incredibly expensive. What you would do is model it with CAD software and simulate it using FEA. Almost all highly technical physical products these days are designed with simulation software. You want to know how your product performs before you spend a lot of money on a prototype. FEA is used to test stresses, forces on objects, etc. CFD is used for fluids, airflow, etc.
Ah okay. Thanks for explaining. So is this generally a one-man job? Like companies don't seek "simulators" to help designers, they hire designers with the ability to simulate themselves? Also, is this still how it works for simulating things like chemistry or astrophysics?
I'm sure every company is different, but yeah, usually companies hire an engineer since they are experienced in design and simulation. I don't know much about chemistry, but astrophysics, yes. That's what's known as aerospace engineering.
Thanks, I will relook my equations.
Another option could be solver iterations or biases. Let's take divergence for example. If you're enforcing incompressibility from the top of the screen to the bottom, cell by cell, your result will be skewed if you don't have enough solver iterations. It will also be skewed if you have biases in your equations. Hope this helps.
Reminder, this is
I'm sorry.
You're probably confused because your transparent is probably at 1 and yet it's not transparent. That's because if you have a metallic value that isn't 0, your material will lose its transparency. Metal isn't transparent. Turn your metallic down to 0.
https://wethaney.gumroad.com/l/dpkjv
Woah! That's amazing! How did you do it!? Willing to share nodes at all?
It's really just a particle sim with a lot of particles, then I used the lifetime and the frame count together with a color ramp and some math to make the colors.
Any chance you could share the blend file? Or did you follow a tutorial? I'm still fairly new to particle sims
Honestly, this is a very simple particle sim. The only thing that's cool about it is that there are a lot of particles. No tutorial, but here's a video that heavily inspired the render.
I want to marry you bro. Please send me your address
Oh. No thank you. ❤️
There's a fluid shader here on Gumroad, available for free. It should help improve the greyness of your fluid. You can also use it on the glass and ice.
Yo
Did you actually....
[удалено]
Part of the reason is probably because this is actually in 3D, so it kinda looks weird if you're expecting it to be 2D. But yeah, I can agree with those other points. In simulation nodes, the time step is locked to the frame rate which is annoying, but I'll admit that I only used a first order divided difference scheme. It's my first time doing simulations in simulation nodes, so I just wanted it to be as simple as possible so I could learn the workflow. Thanks for the feedback though, I'll keep that in mind for future simulations.
https://wethaney.gumroad.com/l/eqyciu
https://wethaney.gumroad.com/l/eqyciu
I've been working on my water lately and it is not good
I'm sorry to hear that.
Can we get a wireframe or clay view of the water? It really looks like a 2d plane.
Wireframe.