Help Does both Obi Cloth and ObiRobe are needed to make a football net?

[josemendez]

Ahh, I see that I got the use of the Obi Rope wrong, I thought that the net was literally build using small ropes or something like that, not that was just used as tensors. Thinking well, a net properly build like a real one with collisions would require a ridiculous amount of processing, my bad.

Yep, it's far simpler and faster to use a mesh with a net texture on it

Well, comparing to my skills on the subjects you quoted, I'm a hell of a programmer lol. But I'm trying to improve my math and physics related to gamedev, cause build a football game already relies on these 2 (my dream is to build a complete football game, but I'm far of that level I'm afraid). Not much choice here, I don't believe there's a better solution than Obi for this case, the alternative is try to build a good net system myself, which I already tried and failed.

Another doubt came up, how easier is to build this system with Obi compared to keep trying to build it myself? Asking this cause a friend of mine already suggested to go through this physics simulation path to make this system. His suggestion was to use the built-in cloth component and build/find an equation to predict when and where the ball will collide with the net, adding the respective physical reaction to both ball and net accordingly. While his theory is clear to me, in practice, things are really hard to come up with. I just want to know how much Obi will put me (or doesn't) in the same, miserable, direction lol...

I've been in exactly the same situation as you. A few years ago I needed a net simulation for a soccer game (not a fifa-like realtime game, more like tactic/turn based stuff). Unity had just removed cloth-rigidbody interaction from their built-in cloth, so there was no way to have the net and the ball react to each other. The ball would just pass trough it. So I implemented my own cloth, and over time it became Obi:

https://www.youtube.com/watch?v=aY0IEIJek9U
(Note this video is quite old, things have improved considerably since then).

It took me like 4-6 weeks to get the basic stuff working. However, getting things like performance and user experience to acceptable levels took me around 2.5 years. If you have the time, resources and drive, by all means try to do it yourself. It's a lot of fun and you'll learn heaps . I've gathered all material anyone can need to build their own engine, a list of all research articles I used is available on the webpage: http://obi.virtualmethodstudio.com/references.html

If you decide to use Obi, it will do most of the work for you. You do need to understand some basic stuff about 3D: how meshes and transforms work, what's the difference between local and world space, and a few other things, but I think you'll be fine.

What about the ball X net collision considering the specifications from the previous message? Is there no risk the ball to pass through the net with the mentioned speeds?

No problem. This is called tunneling (https://www.youtube.com/watch?v=ms0Z35GY6pk).
Obi implements continuous collision detection (via speculative contacts), it should be pretty robust. I've also recently added surface collisions to prevent small objects from slipping trough gaps in between cloth particles, so collision detection is continuous both temporally and spatially: https://www.youtube.com/watch?v=9LXeIem4gvM


You can take a look at Obi's manual here:
http://obi.virtualmethodstudio.com/tutorials/

Take your time reading it and poke around, I can answer any questions you might have.

cheers,