I just read the recent dev log about how capital ships were a little unwieldy, I was a little curious about that.

Are you simply assigning a mass value to the ships or is the system automatically calculating based on the size of the ship? In what way did you have to cheat the system, did you simply decrease the mass of capitals? If the system is automatically calculating the ship mass, you could add in an exponential falloff so that as the ship gets larger, the amount of mass assigned is reduced (much more corridors/open spaces for occupants to walk) or simply retool the weight assignment for all ships slightly lower so the much larger ships would have a large reduction while fighters would see almost no gain.

This should be in Technical or Suggestions I think.. Could a mod be so kind, I'll pay more attention to the other forums in the future


[The kind mod can't decide, either, so General fits best. - Gazz =P]

Mass is automatically calculated from size, and then inertia (which controls maneuverability) is automatically calculated from mass and the size of the object, using the approximation that the object is a solid block.

See: http://en.wikipedia.org/wiki/List_of_mo ... ia_tensors

The true inertial tensor of a block scales as the square of the dimensions times the mass. For a block of uniform density, this effectively means that the inertia is, roughly, scaling as the fifth power of the object's size (where "size" is taken to mean the length of one axis of the object, assuming we scale the whole thing uniformly). I have found this to make large ships too slow, so I've essentially removed the second power from the formula, making it scale roughly as the fourth power of the object's size.

Honestly, the reason why I have to do this is probably because I'm not calculating torque correctly. If I were, I would take into account the length of the axis, which would effectively counteract one of those powers, bringing it back down to 4th power anyway.

So this means maneuverability scales roughly inverse-linearly. You have a battleship that is 100x as long as a fighter, you can expect about 1/100th of the maneuverability of the fighter. Makes sense I think

Honestly, the reason why I have to do this is probably because I'm not calculating torque correctly. If I were, I would take into account the length of the axis

Torque = length of the lever arm multiplied with the force. I think this takes us into the game's rules for engine power. Assuming

• the torque is delivered by small maneouvring thrusters whose thrust follows the same general principles as the thrust of the main engines (scaling with the n_engine-th power of size)
• and the distance of maneouvring thrusters from the center of mass (lever arm) is proportional to the size of the ship

the torque would scale as the (n_engine +1)-th power of size.

As an example, if n_engine is 2 (the thrust scales with the square of the size), the torque would grow with the 3rd power of size.

A related question is how drag will scale with power of size. Real life air drag, for instance, follows the drag equation http://en.wikipedia.org/wiki/Drag_equation where A scales with the square of the size. If you want big ships to be slower than small ones, LT drag needs to scale with a bit more than the n_engine-th power of size. It's all interconnected .

JoshParnell wrote: For a block of uniform density, this effectively means that the inertia is, roughly, scaling as the fifth power of the object's size (where "size" is taken to mean the length of one axis of the object, assuming we scale the whole thing uniformly)

Surely it's proportional to the square of the (axial) size? The multiplicative constant will vary depending on the shape and which axis it's about, though.

http://hyperphysics.phy-astr.gsu.edu/hbase/mi.html#cmi

Edit===

Ah, I see what you're saying - you're factoring in mass assuming it's proportional to the cube of the length. While that's valid for equal lengths (e.g. spheres), aren't caps more likely to be long and lean, i.e. they're a closer approximation to a rod?

*Head explodes*

HowSerendipitous wrote:*Head explodes*

Physics considered harmful

Since ships are planned to be customizable, wouldn't it be advantageous to have a more accurate size approximation.

This does bring up the idea that different shapes could matter, and be advantageous for different purposes.

If I have a bombardment cap that can fire massive ordnance at faraway slow/stationary targets, then I probably need to guard against similar incoming. That means it's sensible to have a smallest possible cross-sectional area, i.e. be like a long, thin pencil. Probably with a massive shield at the front. This means I will have very poor rotational manoeuvrability, but that doesn't matter as long as I stay far from the enemy as I can rotate to face any threat in plenty of time. If enemy fighters get close, then my manoeuvrability is irrelevant anyway as it's my point defence ability that matters. Thus a pencil is the optimal shape for such a purpose. Similarly other types of ships may have other optimal shapes for their purposes.

Yes, a long slender ship with a huge rail gun down the spine for the purpose of long range bombardment of planetary instalations or orbital facilities. You could accelerate the slug to C frational velocities. I like it!

Ring design

Call it the Ouroboros. =)


Or build an Escher staircase. Confuse the hell out of attackers.

I want a space station with a moebius strip for an entrance pathway. "Treasury this way!" It'll confuse the hell out of any invaders.

I want a space station in the shape of the sentence "YOU SUCK" covered in anti ship weapons. Laugh as they realize how much they suck as they are incinerated.

The Hedge Knight wrote:I want a space station in the shape of the sentence "YOU SUCK" covered in anti ship weapons.

In which language, and using which alphabet?

You can easily argue that every single space station you come across has the shape of the sentence "YOU SUCK" in the local hieroglyphs.