Uber, I figured out how to get Subs to work... (I think)

Its a hell of a lot harder to create a layer in a spherical world than it is in a flat one.

This won't work for several reasons:

• Shallow water. There is no current "minimum depth" attribute, so its quite likely the sub will clip into the ground in shallow water. Besides being hard to see, this will also mess with targeting and projectiles (both inbound and outbound)
• The hitbox extends out of the water. So bombers and other surface vessels can still hit them, which mostly negates the point (layers only dictate automatic targeting; manual attack ground commands can still be used).
• The center of the unit is not determined via the base bone, but rather the physical center of the model.
• Building the sub in the factory would just look... weird

And probably many other technical reasons.

I should point out that in TA the subs weren't in their own layer. They were ships made to appear below the waterline. This has many disadvantages, such as subs blocking surface ships, as well as pathing issues, since it depends on each unit being able to specify min water depths. Doing that means new pathing nav meshes for each unit, which is not feasible for flowfields, especially when they are dynamically generated like in PA (TA uses heightmaps, PA doesn't).

I know, I'm griping that 17 years later naval is still busted and it looks like it'll stay that way for a long time. That naval stretch goal in the kickstarter was by far what I was most excited about, and since Uber has pretty much ditched that goal I gripe about it whenever I get the chance.

GUESS WHO WAS RIGHT EVERYBODY!?

Now now, no-one likes a smart-arse

Well that's counteracted by everyones unconditional love of me. <3

No one likes a gloater...

Although I admit I thought of your suggestion the minute I saw the subs

tl;dr squishy wants subs to be able to clip through the ground

Well then so does Uber as this is pretty much exactly what they did.

I'm going to count this as a win on my part.

no, thanks, I'd rather wait until it's done properly

Who's to argue what the proper way is? Did you know subs in TA weren't actually underwater either? That they just had a blue filter over them basically?

This mainly being because _there was no depth_ or no under water in TA.

So? TA was made more than 15 years ago. SupCom had an underwater layer, AND subs actually moved around like they were underwater - they couldn't clip through the ground in shallow areas. I'd rather avoid the clipping through the ground bit and find the 100% solution the first time.

You don't understand, these subs shouldn't clip underground, they'd bounce against the shallow bit. The models are physically there, the model is the collision model essentially.

that is the opposite of what you said the first time

No what I said then is that the base armature would be on the surface, as far as I can tell from modelling on PA the models themselves are their own collision boxes. BBesides Uber put the sub barely below the surface, they're not as deep as the bottom of the old ships!

"yet move as if on the surface, and this means that there is no need for a layer below the surface."
That means it moves as if it were a surface ship - it would be able to go places a sub wouldn't be able to, and it would clip through the ground.

Is there not control for minimum depth?

Yes it would move as if on the surface because the base armature is where are movement and turning comes from, what I meant by this is that the base armature would turn or move on the surface while the sub stays below the water.

Anyway none of this matters anymore, as the sub is too close to the surface to clip through the ground anyway, it's just submerged to the periscope. My original suggestion had it all the way to the bottom of the original ship AND the sub models are probably smaller now.

No, it doesn't. As I said in my earlier post, the proposed method wouldn't have worked before because the game lacked either a "minimum depth" attribute or the ability to automatically work one out based on the bounding box of the model.

If either of those two things have been implemented, then @squishypon3 is absolutely correct, and there wouldn't be any clipping or inappropriate movement.

idk, but oceans aren't nearly as deep as they probably should be.