Simulating wind movement?
Simulating wind movement?
Hello, i was thinking about how to simulate wind movement in a good way. What i was thinking about was simulating wind direction and speed in a somewhat real way.
Anyone have any ideas, i have looked on Google but not found any good answers.
Maybe its possible to use some noise to simulate it?
Anyone have any ideas, i have looked on Google but not found any good answers.
Maybe its possible to use some noise to simulate it?
-
- Posts: 1186
- Joined: Fri Dec 29, 2006 12:04 am
Re: Simulating wind movement?
Have a look at the GrassPatchSceneNode in the irrExt repository. It has a wind generator class.
"Whoops..."
Re: Simulating wind movement?
Awsome, thanx : )
Re: Simulating wind movement?
if you want to simulate wind... I would advise simulating areas of high and low pressure. Then use proper equations to work out wind speed and the movement of these. You will get nice varying direction changing wind
Re: Simulating wind movement?
That would be the correct solution i guess, but way to advanced for my application : p
Re: Simulating wind movement?
for my wind generator I used 2 1D perlin noise generators, one for the wind magnitude and one to interpolate between a min & max direction vectors, i tuned the effect by "linking" the generator to a visible wind vane style helper object showing the direction and force (scale in wind direction) of the wind.
Re: Simulating wind movement?
That sounds like the way i thought about doing it.
Came in contact with noise generation when i played around with terrain generation and thought that it would work good for wind generation as well : )
Came in contact with noise generation when i played around with terrain generation and thought that it would work good for wind generation as well : )
Re: Simulating wind movement?
a quick note: the code I use can be found in the flag scene node i created btw.
http://irrlicht.sourceforge.net/forum/v ... =9&t=45093
http://irrlicht.sourceforge.net/forum/v ... =9&t=45093
Re: Simulating wind movement?
while we are at it is there any good paper or article on how to build a such simulation it's the kind of implementation i will likely use in my projectdevsh wrote:if you want to simulate wind... I would advise simulating areas of high and low pressure. Then use proper equations to work out wind speed and the movement of these. You will get nice varying direction changing wind
Re: Simulating wind movement?
I really wouldn't recommend that... Its how the weather service does it
(And it takes long and the look ain't worth it If you only want game wind effects)
(And it takes long and the look ain't worth it If you only want game wind effects)
Re: Simulating wind movement?
well obviously for a game you implement a very simplified version but i don'T only wanna implement a wind simulation i wanna implement a planetary wide weather simulation to obtain result that would look natural obviously it will need to be simplified to fit in a game but i still think it would be worth it in the type of game i'm working on
Re: Simulating wind movement?
Well then find some Meteorology books and start reading... it's seldom traveled Territory you're in!
I wish you Good Luck!
I wish you Good Luck!
Re: Simulating wind movement?
i will start by understanding how to build a dynamic lod that can generate a complete planet with irrlicht then ill look into this .....
Re: Simulating wind movement?
Even our greatest supercomputers don't model the planet's entire weather system. \studied meteorology many moons ago (no pun intended)
Modelling local conditions is more manageable and effects from outside the local system can be averaged. High pressure from north, low from west etc.
My suggestion would be to use a few vectors, you can add noise vectors for randomness, but essentially you want a master vector and a series of affecting vectors (from north, south, east etc). Then you would allow vectors to add or cancel each other giving you a local result.
Gah... sounds soooooo dull.
Modelling local conditions is more manageable and effects from outside the local system can be averaged. High pressure from north, low from west etc.
My suggestion would be to use a few vectors, you can add noise vectors for randomness, but essentially you want a master vector and a series of affecting vectors (from north, south, east etc). Then you would allow vectors to add or cancel each other giving you a local result.
Gah... sounds soooooo dull.
Re: Simulating wind movement?
well a basic map of high and low pressure zone could be simulated based on a height map ,and some data (i have no idea currently what kind of data would be usefull in a such an estimation)
for exemple and gpu pixel shader could be used to compute the wind vector on each point every X amount of time and updating pressure zones etc
the simulation only need to be accurate enought to fool the user localy not simulating everything uber accurately every where
for exemple and gpu pixel shader could be used to compute the wind vector on each point every X amount of time and updating pressure zones etc
the simulation only need to be accurate enought to fool the user localy not simulating everything uber accurately every where