Helm's a-lee!


Mainsail haul!


Let go and haul!

Two-sided sails after I had turned off backface culling in the shader. They shimmer gloriously when they move.

I've used essentially the same thing as Boris said, but with a simplification that should work as long as the bullet moves much faster than the target.

From the speed of the bullet and the distance to the target at the moment you fire, you can get the approximate time t that the bullet will take to reach the target. Then aim at where the target will be at time t.

A competition! <glasses fall off>

It may be time to dust off my idea for a croquet+Victoriana game.

I found this nice pdf about the F-16 physics where there is a graph of the CL shown. I just tried to approximate it using Mathematica and ended at this equation (which is pretty accurate):

Code:

CL = sin(AOA/22)*1.9

where AOA is angle of attack and the sin takes radians. I'm not going for complete realism here, so this should be just fine.

That graph on page 7 for the F-16 looks weird. As I understand it, the graphs here http://www.aerospaceweb.org/question/airfoils/q0150b.shtml are more typical; the lift looks like a sine curve, going up to about 1.2, with a bump at the start. The bump is the useful part, since you get reasonable lift while the drag is still low. The stall is the dropoff after the bump.

The F-16 is maybe an unusual case. For example it isn't designed to be inherently stable like an old-fashioned aircraft, where the natural relation of forces between the wings and the tail tends to stabilize the angle of attack. On the F-16 it's done with computers.

For the parasitic drag, you could just add a small constant factor (I can't remember whether I used 0.2 or 0.02), independent of the angles, to the drag coefficient.

As eigenbom said, the straightforward thing is just to treat the velocity as a scalar and get the lift as a scalar out of it. But you want the full velocity of the wind, not the component parallel to the wing.

(Possibly there's some fancy way of sneaking vectors into the term for wing-area that makes the whole thing into an authentic vector product.)

How are you calculating your lift coefficient? Are you going to have stalls?

Some useful sources:

http://www.av8n.com/how/
http://www.aerospaceweb.org/question/list.shtml

In Crusader Kings 1 you have to find husbands/wives for yourself and the members of your family and court, and the children's stats are influenced by their parents'. I gather this was possible to game to breed superbeings over several generations, and I don't know if it's still there in Crusader Kings 2.

Cheers.

I imagine I should drop the damping as it surfaces. That should give it a nice bob as it comes up.

Yes; and have the buoyancy proportional to something like the fraction of the object that is underwater.

If it were me, I would also be tempted to put in a more realistic drag explicitly (but maybe still keep the damping, for general good behaviour):

velocity = velocity*damping - velocity*velocity*drag + gravity + buoyancy

It's in 2D.

I looked at the Wiki but that's my point - it's not simple. It's not as basic as I need. I'm not using force equations anywhere and I don't want to.

Here's my gravity:

Code:

gravity = 1;
damping = 0.99;

velocity = velocity * damping + gravity;

Then buoyancy should be the same, a constant upwards acceleration (depending on the density of the object). However there is a lot more drag underwater, so I imagine things reach a constant terminal velocity fairly quickly. I guess you will get something like this from your equation too, if you increase the damping.

Is this 3d or 2d?

In 3d, I did this: for each triangle in the mesh, calculate the area, the normal (pointing inwards) and the central point. Then work out the water pressure at the central point (proportional to depth, or 0 if above the water), multiply that by the area to get the scalar force, then apply the force at the centre in the direction of the normal. You can throw away the non-vertical components of the final force, since in a perfect world those would all cancel out.

This, together with some drag, worked fine for me even on fairly low-poly shapes.

Real buoyancy works the same way: if an object is fully submerged, the water pressure is higher at the bottom than the top of the object, so the resultant force pushes it upwards.

As far as shadows go, I would be tempted to have one, vertical one to help with positioning, and leave it at that.

Luckily I won't need to rely on them for that. I long since added a kind of targetting frame that extends downwards from the object. It's more precise than a shadow and prevents annoyance with dropping objects in the wrong places.

A shadow would also help understand the positions of objects that have already been placed.

Could we see screenshots of some complicated scenes?

I've never used them, but as far as I understand it those depth-buffer based shadows shouldn't slow down with multiple objects. Does the game really need multiple shadow-casting lights? (I'm assuming it's King Machine.) As far as shadows go, I would be tempted to have one, vertical one to help with positioning, and leave it at that.

I can't really answer the question, though, as I don't know what techniques scale well and are available easily in Unity. I would like to know too.

Hold the Line

Thanks for the feedback. I started with blueprints and then tweaked it by photo references found on Google images. Is there any particular parts that are especially noticeable to you?

I tweaked it a fair bit and got some different angles.

I think the main thing that bugs me is that you can't make out the shape of the cockpit, although I understand that this is something you are planning to leave to the texture.

I've seen guys in lowpoly threads on polycount do it like this:

...

That's a nice trick. I don't think I'm allowed it, though, as I'm trying to stick to untextured and flat-shaded models.

this is really nice. You could save a lot of tris on the wheel ( I assume when you say 154 polys you mean 308 tris?)

It would probably work out at around 308 tris, since most of the polygons are quads. How would you do the wheels?

Speaking of which I have already started another project; which is a Spitfire plane. C&C very welcome.

158 poly's.

I confess that that doesn't look a lot like a Spitfire to me, although it may be the angle. What are you using as a reference?

I saw the car and couldn't resist having a crack myself. 154 polys:

FFFFFFFFFFFFFFFF

Shot of my troubles: http://scr.hu/4yr/b3m90

As you can see, backfaces are rendering just fine (woooo dont have to manually duplicate faces in modelling _NISE_) but there's also massive sorting issues. The trunk renders through the leaves from multiple angles.

...

You could make the trunks and the leaves into separate objects, and draw the trunks as solid and the leaves as transparent. This would leave sorting problems on the leaves, but if they are all more-or-less the same colour that might be acceptable.

I really like this; and it's the kind of look I would like to go for on my current project. Can you say a bit about how it is done?