Angled Water Areas

[xiphirx]

Hello,

I noticed that angled water areas are not possible.. if you rotate a water area, it acts as if it is a large cube instead of a sloped plane...

This allows things to float in the air...

Is there a way to override this or get around this?

[Russ Money]

Hello,

I noticed that angled water areas are not possible.. if you rotate a water area, it acts as if it is a large cube instead of a sloped plane...

This allows things to float in the air...

Is there a way to override this or get around this?

Maybe turn off the collide for the plane? Don't know if that'll solve it.

[Streetboat]

What are you trying to accomplish? Perhaps some trickery could be applied.

[xiphirx]

I have water running down a slope... so my water area needs to be sloped.

Russ: It's an area...

[palistov]

Select your area and press 'B'. This creates a compound which is rotated independently of the water area. This should work as long as you don't break down the compound afterwards.


EDIT: Wait actually I just tried and I'm able to rotate the area without making the compound...

[nemesis567]

He meant that if he rotates the area, in the game it behaves like a cube and that's how the engine works, so no fix for that I believe

[xiphirx]

Select your area and press 'B'. This creates a compound which is rotated independently of the water area. This should work as long as you don't break down the compound afterwards.


EDIT: Wait actually I just tried and I'm able to rotate the area without making the compound...

Yea, I'm able to rotate it too, but in game it acts as if its a giant cube instead of like how its supposed to...

[Apjjm]

Yea, I'm able to rotate it too, but in game it acts as if its a giant cube instead of like how its supposed to...

I just tried making "steps" of water areas to emulate a slope. However, this doesn't seem to work.

I think the solution to this is going to require scripting buoyancy to a degree. Obviously we can't clip the shape by the area and calculate the submerged volume using current methods provided by the engine, so we are going to have to assume all objects are spherical - which unless you require perfect physics here is a fair sacrifice.

The idea is thus:

1. Create a water plane as normal. This will be the surface.
   
2. Create a thin script area running parrallel to the water plane. Call it liquid_1. This area should have a collide callback triggering on something like "object_*" with all floating objects prefixed with "object_". Have this call a function in your hps file on both entrance and exit.
   
3. Duplicate the script area multiple times so we have various "slices" beneath the water area.
   
4. Write some code to flag how many "tiles" each object is and store this information.
   
5. For each object that is flagged apply a force away from the water based on how many tiles the object touches.
   

I am working on slapping some code together now

Edit Done!
http://www.mediafire.com/?62ywej36g5w887w

Notes:
The values of the vector * volume displaced are VERY approximate. There is no reason why you shouldn't properly work out these values using some trig, and multiply them against volume. You could also produce a lookup table for objects and their volumes (remember to divide by the maximum no. depth objects they can flag), for much more precise forces when dealing with different sized objects. I am also not going to say right now this is bug free, it was put together in just under an hour (including writing this post), but it seems to behave fairly realistically considering. Finally, this WILL require some tweaking to get to work with saves, thanks to the use of the List class which wont be saved.

Edit 2: You can also apply the effects of current by changing the values of the vector*volume.

[palistov]

Oh I get what you mean! It behaves in-game like a giant cube. That's disappointing. Sorry mate

[xiphirx]

Yea, I'm able to rotate it too, but in game it acts as if its a giant cube instead of like how its supposed to...

I just tried making "steps" of water areas to emulate a slope. However, this doesn't seem to work.

I think the solution to this is going to require scripting buoyancy to a degree. Obviously we can't clip the shape by the area and calculate the submerged volume using current methods provided by the engine, so we are going to have to assume all objects are spherical - which unless you require perfect physics here is a fair sacrifice.

The idea is thus:

1. Create a water plane as normal. This will be the surface.
   
2. Create a thin script area running parrallel to the water plane. Call it liquid_1. This area should have a collide callback triggering on something like "object_*" with all floating objects prefixed with "object_". Have this call a function in your hps file on both entrance and exit.
   
3. Duplicate the script area multiple times so we have various "slices" beneath the water area.
   
4. Write some code to flag how many "tiles" each object is and store this information.
   
5. For each object that is flagged apply a force away from the water based on how many tiles the object touches.
   

I am working on slapping some code together now

Edit Done!
http://www.mediafire.com/?62ywej36g5w887w

Notes:
The values of the vector * volume displaced are VERY approximate. There is no reason why you shouldn't properly work out these values using some trig, and multiply them against volume. You could also produce a lookup table for objects and their volumes (remember to divide by the maximum no. depth objects they can flag), for much more precise forces when dealing with different sized objects. I am also not going to say right now this is bug free, it was put together in just under an hour (including writing this post), but it seems to behave fairly realistically considering. Finally, this WILL require some tweaking to get to work with saves, thanks to the use of the List class which wont be saved.

Edit 2: You can also apply the effects of current by changing the values of the vector*volume.

Cool, I was hoping that I didn't have to go in and script it, but that's fine as well. I'll look at the code and report back. I'm not too advanced in physics, but I should be able to understand it. Thanks man!