As a quick and dirty it works very well. From an engineering perspective, that's an impressive artistic accomplishment. ;) I would characterize such a fluid as follows: 1) infinitely compressible - fluid doesn't push fluid out of the way 2) zero attractivity / self attractivity i.e. zero surface tension, viscosity, 3) perfectly dampened. It can't slow down or speed up, with the exception of hitting the surface. This is because you're using the set position nodes to be the dictator of how the fluid reacts, whereas a real fluid needs to be altered at the velocity level, and incorporating a terminal velocity aspect. I think you'll end up with even more interesting simulations with eddying and calculated turbulence. (And you would probably still use a noise texture to help it along into its turbulence depiction, but you would perform a check against the velocity to determine how much noise to apply to it so that you can have laminar and turbulent flow.)
@@Hecker9974 no. Note that as air moves over and under an airfoil (plane wing) it has to speed up, then slow down. Because it speeds up faster on the top side than the bottom, the bottom has higher pressure than top and delivers lift. The two streams recombine after the foil. This is like a controlled raindroplet hitting the car simulator that ignores wind and coalescence as a contributors. This is why we're convinced of its realism.
for some reason for me at 13:29 when i copied the nodes i had to change the killer to ''less than or equal'' instead of ''greater than'' otherwise it would not start, i think it had do so something with the position of the killer because when i placed it behind the emitter it worked but when placed in front of it , it did not until i changed that . also great tutorial! thank you so much!
a few weeks ago I needed something along the same lines but the collission object would redirect the colliders to another direction. couldn't figure another way other than BRUTE FORCING lol
for the material, instead of a principled bsdf, i wonder if plugging an emission shader node in the Material Output node's Volume socket wouldn't give a good enough result, but possibly better perfs and at the very least removing the need to play with light paths...
"it'll build up anger" -CGMatter lmao. Hey, I program procedural systems and tools in Unity3D and I happened upon this video after I started a search for procedurally creating wind patterns - I'm thinking about trying to simulate the movement of wind across the surface of a heightmap..I want to distribute moisture and heat more accurately across a terrain so that I can determine where rain shadows are and what kind of weather an area on the map is most likely to have at any given time of the year. I think this video has some good starting points to work from. Thanks. SUB'D
#1 Why is the addon so inexpensive? Thank you? #2 Why the transparency in the particles? I'll edit this comment once I find a solution to have the particles be more visible in the renders. Thanks again for this great addon. I was looking for something like this just a few months ago.
what should i do when there is like buggyness by particles going in and out of the object really quickly. It looks really ugly. How can I fix this? Please help ...😅
This is a cool effect but not how real air works. Air creates pressure when it hits an object and interacts with other air. Please don't use this to design stuff. If you want a free aerodynamic simulator, Freecad is your best bet.
@@sicfxmusic The interesting part of this video is that he sets it up the same way as you'd set up a normal fluids analysis. Air in, air hits the thing, air out.
That is if you consider the thin film that gets created around the analyzed surface as a separate domain with different conditions that make the equations more complicated, so you end up having an external domain where you can simplify the equations into Euler equations and an internal domain where you can't make the assumption. But for a simple subsonic fluid in a visual demonstration, you can just work by assuming that you can simplify the entire domain with Euler equations, speed of simulation is a priority in vfx
You have clearly made geo-nodes your btch, I am wondering, why you haven't delved into a software like Houdini, which provides even more potential and interactivity between nodes? Is it related to your audience, are you attached to Blender for other reasons? It confuses me a bit
now teach us how to make a procedural lobster
procedural lobster
.
Much was revealed to humanity upon the discovery of lobster aerodynamics
Lobter 🤤
Yes ye
Blender really has come a long way... I still remember nodes being limited to textures and the render engines being Cycles and Internal.
Thank you, Ed Sheeran! That’s an awesome effect
Hes not Ed Sheeran. He is our Blender Ed Shaman
@@peterlaverty9397 for real, dudes a mathmagician
As a quick and dirty it works very well. From an engineering perspective, that's an impressive artistic accomplishment. ;)
I would characterize such a fluid as follows: 1) infinitely compressible - fluid doesn't push fluid out of the way 2) zero attractivity / self attractivity i.e. zero surface tension, viscosity, 3) perfectly dampened. It can't slow down or speed up, with the exception of hitting the surface. This is because you're using the set position nodes to be the dictator of how the fluid reacts, whereas a real fluid needs to be altered at the velocity level, and incorporating a terminal velocity aspect. I think you'll end up with even more interesting simulations with eddying and calculated turbulence. (And you would probably still use a noise texture to help it along into its turbulence depiction, but you would perform a check against the velocity to determine how much noise to apply to it so that you can have laminar and turbulent flow.)
So I can't see if my plane will fly with this simulation? 😞
@@Hecker9974 no. Note that as air moves over and under an airfoil (plane wing) it has to speed up, then slow down. Because it speeds up faster on the top side than the bottom, the bottom has higher pressure than top and delivers lift. The two streams recombine after the foil. This is like a controlled raindroplet hitting the car simulator that ignores wind and coalescence as a contributors. This is why we're convinced of its realism.
That is insanely impressive work, and quite easy to follow. Thanks for sharing.
wdm
Easy to follow, but impossible to learn 😂😂
You, my friend, are BRILLIANT! Thanks for this tutorial!
Looks dope specialyl with that guitar riff in the back xD
Thks a Lot ! ... very useful to schematise winds for architects !
This is so cool. I always wanted to do a formula one in a wind tunnel scene.
for some reason for me at 13:29 when i copied the nodes i had to change the killer to ''less than or equal'' instead of ''greater than'' otherwise it would not start, i think it had do so something with the position of the killer because when i placed it behind the emitter it worked but when placed in front of it , it did not until i changed that .
also great tutorial! thank you so much!
MAGICIAN !
The price was worth the purchase.
Thank you for your work and for sharing.
Im already giving this a thumbs up based on the intro
I love blender and geo nodes, but I had no idea this was possible. Thank you for making this video.
Wow... please do the double slit experiment to see if the results change if you observe the render or not!
I am going to use this to simulate my room airflow and decide on a best course of actions for heating and air quality.
a few weeks ago I needed something along the same lines but the collission object would redirect the colliders to another direction. couldn't figure another way other than BRUTE FORCING lol
for the material, instead of a principled bsdf, i wonder if plugging an emission shader node in the Material Output node's Volume socket wouldn't give a good enough result, but possibly better perfs and at the very least removing the need to play with light paths...
haha i was about to buy that as an addon at blendermarket hehe :D
"it'll build up anger" -CGMatter lmao. Hey, I program procedural systems and tools in Unity3D and I happened upon this video after I started a search for procedurally creating wind patterns - I'm thinking about trying to simulate the movement of wind across the surface of a heightmap..I want to distribute moisture and heat more accurately across a terrain so that I can determine where rain shadows are and what kind of weather an area on the map is most likely to have at any given time of the year. I think this video has some good starting points to work from. Thanks. SUB'D
Thanks DC ! One of the best tutorial i've seen at your YT chanel! Thanks! Greate Gob!
Nice work! I bought this blend file at blender market.
Watching in awe!
please add a 💥 after the intro guitar riff.
also thanks for the turorial
#1 Why is the addon so inexpensive?
Thank you?
#2 Why the transparency in the particles?
I'll edit this comment once I find a solution to have the particles be more visible in the renders.
Thanks again for this great addon. I was looking for something like this just a few months ago.
Masterpiece
i got the issiue that it dosent draw the dots but rather some glitchy surfaces and i dont know how to fix it :/
Im hyped 😍
Sweet tutorial man.
what should i do when there is like buggyness by particles going in and out of the object really quickly. It looks really ugly. How can I fix this? Please help ...😅
5:57
can make a tutorial with smoke effect?
This is a cool effect but not how real air works. Air creates pressure when it hits an object and interacts with other air. Please don't use this to design stuff. If you want a free aerodynamic simulator, Freecad is your best bet.
Hold on. You mean air doesn't have an "anger" parameter?
Calm the freecad down, scientist!
@@sicfxmusic The interesting part of this video is that he sets it up the same way as you'd set up a normal fluids analysis. Air in, air hits the thing, air out.
@@mealroyale that's cool
That is if you consider the thin film that gets created around the analyzed surface as a separate domain with different conditions that make the equations more complicated, so you end up having an external domain where you can simplify the equations into Euler equations and an internal domain where you can't make the assumption.
But for a simple subsonic fluid in a visual demonstration, you can just work by assuming that you can simplify the entire domain with Euler equations, speed of simulation is a priority in vfx
👑
daaaaamn that inntro rips
Does this work with skinned characters?
wtf bruh i was looking for this 4 days ago
We getting aero software for free with this one
Why CFD, this is much more accurate!
You should probably make your own vfx company
The goat 🐐
big
You have clearly made geo-nodes your btch, I am wondering, why you haven't delved into a software like Houdini, which provides even more potential and interactivity between nodes? Is it related to your audience, are you attached to Blender for other reasons? It confuses me a bit
1st
Noice