[Community Project] WIP Weather & Ocean Water Shader

[DIFTOW]

It's my own code what you see at that video but that doesn't mean you can't make something similar with the current buoyancy code, because you can!

I never said it would work perfectly right off the bat with the instructions I gave you, I was just trying to help you get started since it sounded like you don't know how to use the buoyancy component.

I don't think anyone here said that everything works perfectly without some tinkering around

I get your frustration but no need to be rude! eventually it will be more simplified and I also plan to share my buoyancy code and other shader improvements for those who are interested.

-------------------

@DotCam

Would it be okay if I make my own fork of the project? that way I could share some of the experimental stuff I am working on, and if there is something you like, you are more than welcome to add it to the "main" branch

Sorry if I was rude. I am however justified in feeling bitter. I wasted a considerable amount of my time and got nowhere.
I'm already stretched thin with my project, because the 2 artists I have are not helping me at all with anything. I'm doing everything myself, on top of managing the team, on top of learning an entire new engine and editor and the workflows for everything. So I'm already feeling like there isn't enough hours in the day, and only so much my brain can absorb in a day before the headaches come and before I'm just completely exhausted.

[Komodoman]

Sorry if I was rude. I am however justified in feeling bitter. I wasted a considerable amount of my time and got nowhere.
I'm already stretched thin with my project, because the 2 artists I have are not helping me at all with anything. I'm doing everything myself, on top of managing the team, on top of learning an entire new engine and editor and the workflows for everything. So I'm already feeling like there isn't enough hours in the day, and only so much my brain can absorb in a day before the headaches come and before I'm just completely exhausted.

Diftow,

I think anyone on these forums, or in game development in general who has had to deal with creating any sort of finished product has dealt with the frustration and time commitment that comes with it. That is not to say that this work isn't incredibly rewarding for those that enjoy overcoming these challenges. Most of us have had to deal with some sort of lousy co-worker or teammate in the past, but that doesn't give anyone the right to feel bitter or lash out at work that is entirely free, whose creator is not paid, and whose purpose is betterment of community content. I'm not trying to be a jerk, but please consider the amount of hours and frustrations that DotCam has gone through for people like YOU before posting about feeling cheated out of YOUR time if not everything is polished and documented perfectly.

Komo

[DIFTOW]

Diftow,

I think anyone on these forums, or in game development in general who has had to deal with creating any sort of finished product has dealt with the frustration and time commitment that comes with it. That is not to say that this work isn't incredibly rewarding for those that enjoy overcoming these challenges. Most of us have had to deal with some sort of lousy co-worker or teammate in the past, but that doesn't give anyone the right to feel bitter or lash out at work that is entirely free, whose creator is not paid, and whose purpose is betterment of community content. I'm not trying to be a jerk, but please consider the amount of hours and frustrations that DotCam has gone through, and will continue to go through before you post about feeling cheated out of YOUR time if not everything is polished and documented perfectly.

Komo

The only criticism I gave was directed at TKMaster for failing to inform me that his video was done on a different system and code, when I asked him how to do it in private message.
This resulted in me trying to achieve the same results when I otherwise would not have, if he hadn't omitted that very crucial information.
I have not criticised DotCam nor have I blamed him for my challenges. I've apologised to TKMaster. Please do not distort the situation.

[Nsomnia]

The new fish code or the old code? Do you remember the error your were getting?

FishCode.zip dl'ed on march 6th. I had some weird issues adding the new c++ files in the editor and then replacing them and then recompiling so its probly an issue on my end. I cant remmeber the error sorry. Fish isnt a major thing in our game, I just wanted to test it out.

[Komodoman]

FishCode.zip dl'ed on march 6th. I had some weird issues adding the new c++ files in the editor and then replacing them and then recompiling so its probly an issue on my end. I cant remmeber the error sorry. Fish isnt a major thing in our game, I just wanted to test it out.

Ok cool,

I forked DotCam's repo on GitHub and will be implementing them into my own fork today. If that goes well we should be able to easily add it to the plugin. Thanks for testing it out by the way!

Komo

[DIFTOW]

So apparently the issue was the other package. I tried your pre-built and it works properly on that.

But I do have a question about what I had to enable to get this working properly.

Right now the Galleon mesh has physics enabled, and the collision is set to "use complex as simple".
Without that, it just simply doesn't use the buoyancy points.

What I'm curious is does that mean that 1000s of polygons on my mesh are being calculated during the collision/buoyancy with the waves?
Or does your system ignore the geometry of the mesh and just require "use complex as simple" for the buoyancy points to register?

EDIT: Okay its getting confusing again <_<
You see those red spheres? Those are just component shapes, they represent where the buoyancy points are. If I remove those, the ship doesn't float anymore, even though the buoyancy test points are still there! O_O
I'm really confused now.

I'm going to make a video about it cause its just better to show than to try and explain. I'll post it here when its ready.

[TK-Master]

Don't quote me on this as I haven't tested it that much, but the buoyancy forces do not work well with complex collision, you should make a simplified collision mesh (that's what I'm doing at least).

My guess would be that the buoyancy is applying the forces to the component spheres since it sounds like the galleon does not have a collision mesh.

And by the way.. yes, "use complex as simple" means per-poly collision and it's expensive (avoid it whenever possible!).

[DIFTOW]

Don't quote me on this as I haven't tested it that much, but the buoyancy forces do not work well with complex collision, you should make a simplified collision mesh (that's what I'm doing at least).

My guess would be that the buoyancy is applying the forces to the component spheres since it sounds like the galleon does not have a collision mesh.

And by the way.. yes, "use complex as simple" means per-poly collision and it's expensive (avoid it whenever possible!).

I just don't understand why collision matters if I am specifying buoyancy points. I'm trying to understand how the system works, and the fact that it needs the buoyancy points in 4 locations with collision data is like having the same collision data twice. I don't get it.

But that aside, I don't want the collision that I use for the ocean to be the same as the collision I use for players walking on the ship, or ships colliding with other ships :P

Is there a way I can make it so the only collision used for the ship and ocean is just 6-8 spheres like Assassins Creed? And then I can use different collision for other stuff?

EDIT: I tried adding 6 simplified sphere colliders and using "use simple as complex", and this is what happens O_O



If that isn't weird enough, wait till you see this! So I tried using complex collision again with these shape components (red spheres) cause I accidentally discovered that this made the ship buoyant properly.





The red spheres are 'shape' components, not colliders (cause apparently colliders do nothing <_<). These shapes not only seem to suddenly make buoyancy work, even though they are nowhere near the waterline, but when physics is tested against the landscape, the red spheres collide, not the ship.. even though collision is set "use complex as simple" in this case. <_<
It should be trying to collide the entire mesh, but for some unexplained reason, the spheres are the only thing colliding with the terrain. For some explained reason, these sphere 'shapes' make buoyancy work, when sphere colliders do not. It is all very confusing.

If you need more information, let me know :P

[Avolanty]

I just don't understand why collision matters if I am specifying buoyancy points. I'm trying to understand how the system works, and the fact that it needs the buoyancy points in 4 locations with collision data is like having the same collision data twice. I don't get it.

But that aside, I don't want the collision that I use for the ocean to be the same as the collision I use for players walking on the ship, or ships colliding with other ships :P

Is there a way I can make it so the only collision used for the ship and ocean is just 6-8 spheres like Assassins Creed? And then I can use different collision for other stuff?

EDIT: I tried adding 6 simplified sphere colliders and using "use simple as complex", and this is what happens O_O
[ATTACH=CONFIG]30139[/ATTACH]
[ATTACH=CONFIG]30141[/ATTACH]

If that isn't weird enough, wait till you see this! So I tried using complex collision again with these shape components (red spheres) cause I accidentally discovered that this made the ship buoyant properly.

[ATTACH=CONFIG]30148[/ATTACH]
[ATTACH=CONFIG]30147[/ATTACH]
[ATTACH=CONFIG]30149[/ATTACH]

The red spheres are 'shape' components, not colliders (cause apparently colliders do nothing <_<). These shapes not only seem to suddenly make buoyancy work, even though they are nowhere near the waterline, but when physics is tested against the landscape, the red spheres collide, not the ship.. even though collision is set "use complex as simple" in this case. <_<
It should be trying to collide the entire mesh, but for some unexplained reason, the spheres are the only thing colliding with the terrain. For some explained reason, these sphere 'shapes' make buoyancy work, when sphere colliders do not. It is all very confusing.

If you need more information, let me know :P

Can you show my your physics asset for the ship, I had to do something funky to get my AI to collide with another while using the Buoyancy on them. This also may help, if your Collision for your mesh is not set correctly http://i.imgur.com/FgrAbrb.png

[Avolanty]

Dotcam I really appreciate your hard work. I am like a kid leaning over the store shelf excited to get his sweets from the store clerk with lips drooling. I want to play with your Buoyancy structs setup as soon as you release it. Thank you so much for all that you do.

[Komodoman]

Thought I'd post this here, some visible improvements to the fish schooling

[SpykeRel04D]

Hi guys,

With the permission of DotCam, I want to post here, the "NOOB" pluguin Version for UE 4.6.1 :P

I'm sure, that more than one, tried to use the pluguin in this version, and they can't..

If it can help someone, I'll be satisfied! If not, no problem, more rubbish.. hah

https://mega.co.nz/#!SYpSzZJI!xbOAvy...t4CqvZl-7j3zUc

[DotCam]

Hi guys,

With the permission of DotCam, I want to post here, the "NOOB" pluguin Version for UE 4.6.1 :P

I'm sure, that more than one, tried to use the pluguin in this version, and they can't..

If it can help someone, I'll be satisfied! If not, no problem, more rubbish.. hah

I appreciate you trying to help, but that link is for a different ocean system (VaOcean). If anyone needs an earlier version please use one of these links until the GitHub build is updated:

4.6.1 - https://drive.google.com/open?id=0B3...R28&authuser=0
4.7.2 - https://drive.google.com/open?id=0B3...czA&authuser=0

Thank you.

[DIFTOW]

Hey DotCam, I posted some of my bug testing results yesterday with images on the end of Page 10.

I have a feature request, assuming this isn't already easily possible with UE4. I am still getting used to the editor :P

Basically, make it so that the buoyancy test points act solely as the ocean surface collision, no collision model necessary, or allow the usage of a separate collision set, so what the players collide with is not the same as what the ocean collides with. I'd ultimately like to use a special collision set for the Ocean to keep it cheap. The collision I would have to use for players is way more data than the ocean surface needs :P

Some of the issues I experienced on the previous page though, the buoyancy would not accept simple sphere collision on my galleon. The galleon would be unresponsive to the waves, and very slowly descend along the Z-axis for about 10 seconds, then ascend very slowly up the Z-axis for about 10 seconds, then it would start to descend again much faster on the Z-axis, and completely sink to the bottom of the ocean. That was my results with simple sphere colliders.

[SpykeRel04D]

I appreciate you trying to help, but that link is for a different ocean system (VaOcean). If anyone needs an earlier version please use one of these links until the GitHub build is updated:

4.6.1 - https://drive.google.com/open?id=0B3...R28&authuser=0
4.7.2 - https://drive.google.com/open?id=0B3...czA&authuser=0

Thank you.

Oh!
I think that I made a MIX of your Ocean with VaOcean.. OhGOD!
Because there are the two versions :P

But also, thank you very much for the actualized links and excuseme for my Noobness :P