Implementing forces in Unit Motion

[Atrik]

These are videos from a current work in progress to implement different forces into unit motions, like inertia, knockbacks and charge attacks! The values would be very customizable in unit templates for mods too!

    Edited June 18 by Atrik

[wowgetoffyourcellphone]

That ship one is sick.  

[Thalatta]

Indeed it looks good, although infantry seems to be moved less than it should on the first video, and the ship way more than it should. I’d reduce the "mass" of infantry (I guess it’s not really possible to throw some to the ground or send someone flying from time to time? :P), and increase it quite a lot for the ship, along with greatly increasing the "dampening" effect of water. Ships getting cramped and not able to break off was a thing, and a weakness for ramming ships, which needed some space to accelerate for ramming, and didn’t transport a lot of soldiers not to be slow, making them vulnerable to boarding in those situations.

[Emacz]

templates for mods too!

[Atrik]

3 hours ago, Thalatta said:

Indeed it looks good, although infantry seems to be moved less than it should on the first video, and the ship way more than it should. I’d reduce the "mass" of infantry [...], and increase it quite a lot for the ship, along with greatly increasing the "dampening" effect of water. Ships getting cramped and not able to break off was a thing, and a weakness for ramming ships, which needed some space to accelerate for ramming, and didn’t transport a lot of soldiers not to be slow, making them vulnerable to boarding in those situations.

Noted but for the videos, the unit template parameters are for the "demo". The balancing of it all will be a task on it's own , my focus at the moment is to put the features out there as most of them would be optional (so disabled/absent from unit motion).

3 hours ago, Thalatta said:

I guess it’s not really possible to throw some to the ground or send someone flying from time to time?

It could be possible with not too much work on the code side, now that a lot of the structure is already there. But it also require a proper animation, else would look too weird. So that's for another time.. 

You can note that the boat receiving the blow in the video almost seems animated here but really it's just "hacked" with some rolling motion.

[Lopess]

4 hours ago, Atrik said:

 You can note that the boat receiving the blow in the video almost seems animated here but really it's just "hacked" with some rolling motion.

When I saw this, it raised a real question for me! 

[Thalatta]

4 hours ago, Atrik said:

You can note that the boat receiving the blow in the video almost seems animated here but really it's just "hacked" with some rolling motion.

That effect is really well done, did you use a trigonometric function to obtain its positioning, or am I just, as usual, overthinking?

I was thinking, if the game has some choppy oceanic waters at some point, that could become handy. Besides, that could give different advantages to ships, since Mediterranean ships weren’t ocean-going, unlike northern Celtic ones.

Edited Friday at 18:44 by Thalatta

[Atrik]

45 minutes ago, Thalatta said:

That effect is really well done, did you use a trigonometric function to obtain its positioning, or am I just, as usual, overthinking?

We use one trigonometric function to determine the hit angle and deduct the amount transferred as angular velocity and add the force to right direction.

 

47 minutes ago, Thalatta said:

I was thinking, if the game has some choppy oceanic waters at some point, that could become handy. Besides, that could give different advantages to ships, since Mediterranean ships weren’t ocean-going, unlike northern Celtic ones.

It's a really nice suggestion that I like, I don't know how animating "choppy oceanic water", it would require to implement a terrain-dependent bonus or effects, then adding the force would be easy now. However i like it, not sure about what would be the timeline to even have the features enabling this.

[nifa]

The ship collision is cool. Could both units be affected, so the attacker gets a little backlash too?

With this maybe we can have naval capturing soon, with two ships dragging side by side for boarding.

[Atrik]

14 hours ago, nifa said:

The ship collision is cool. Could both units be affected, so the attacker gets a little backlash too?

It does in the video, but there are reasons why it end up just slightly slowing down the attacker. Now taking a step back from what I did empirically and reading the comments, I might just rethink some aspects that would address some defects like this one.

 

14 hours ago, nifa said:

With this maybe we can have naval capturing soon, with two ships dragging side by side for boarding.

Would be very nice, yes more realistic movements and collisions would make a boarding mechanic interesting.

[Thalatta]

2 hours ago, Atrik said:

Would be very nice, yes more realistic movements and collisions would make a boarding mechanic interesting.

Just to give you a heads up, I'm preparing a Thoughts on the Athenians post, with naval combat reserved for its Part II, where I’ll briefly explain how it really was, and what could be done about it to make it more realistic.

[LeJardindEpicure]

On 19/06/2026 at 9:42 PM, Atrik said:

We use one trigonometric function to determine the hit angle and deduct the amount transferred as angular velocity and add the force to right direction.

 

It's a really nice suggestion that I like, I don't know how animating "choppy oceanic water", it would require to implement a terrain-dependent bonus or effects, then adding the force would be easy now. However i like it, not sure about what would be the timeline to even have the features enabling this.

Is the computation heavy ?

[Atrik]

29 minutes ago, LeJardindEpicure said:

Is the computation heavy ?

So for the explanation again, we could now plug any force that could contribute to unit motion and we just sum up the vectors. So it's almost free there, when calculating unit motion. Maybe excepted we still have to check for collisions. The rest of what makes the unit motion looks realistic is inertia, which is also just a ""force"" vector that carry unit velocity over the game turns.
Calculating the forces themselves is generally making some dot products and sometimes maybe some trigonometry (kept as minimum for angle of impact etc) so also possible to be kept very cheap. For the current demo of the boat we use 1 trig function which seems reasonable.

[nifa]

Gotta say I like the inertia in these two videos: https://gitea.wildfiregames.com/0ad/0ad/pulls/8957#issuecomment-140218 

Especially the ships slowly drifting before stopping looks so much more natural than the abrupt stop now.

Edited 48 minutes ago by nifa