Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[GlyphGryph]

Like with cars! Turning would also impart forward momentum. Which mean's you'd have to add brakes, realistically, and who wants to deal with those?

So engine pairs are definitely better.

[ank]

Am i the only one thinking this could make an excellent drone turret system?

No drive engines, only rotational.

BRB building...

[GlyphGryph]

Technically, the turn thrusters could give it pretty weak overall propulsion... enough to get it into position anyway.

[Ivefan]

So... Is it just my ship or is reverse thrust breaking slower than not doing anything at all? atleast according to the speedmeter it is.

[USEC_OFFICER]

Technically, the turn thrusters could give it pretty weak overall propulsion... enough to get it into position anyway.

Technically all you would need for forward propulsion would be to hold the turn left/right or up/down etc. buttons. Unless the amount of thrust is completely unbalanced or they're on different parts of the ship or something, their rotational forces should cancel out and you should be able to go forward. So if you can turn quickly, then you're also able to move quickly as well. Unless you haven't balanced the thing, of course, in which case I'm just rambling as per usual.

EDIT:

So... Is it just my ship or is reverse thrust breaking slower than not doing anything at all? atleast according to the speedmeter it is.

If the thrust amounts are the same on both sides then everything should be the same. Well, unless the game includes relativistic mass, but for some reason I doubt it.

[GlyphGryph]

Except we are presuming paired opposed engines - this generates propulsion-less turning. Literally "on a dime". Holding left and right would just keep the ship exactly where it is, since you'd have an equal amount of force pushing it backwards and forwards.

You would have to decouple them to get propulsion, and the reason it turns so quickly is the lever action - there would be a lot more weight involved in propelling the ship, where that doesn't provide a benefit, meaning that turning quickly is basically the exact opposite of moving quickly in this instance.


Primary engines for an actual high speed version would probably be the heavier types with more power and whatnot, though you wouldn't need breaks, as I said, since you can just turn around and use the same engines, so you'd save some effort there. That's the way most ships should be breaking, though, and is the big advantage of opposite paired turn engines.

[USEC_OFFICER]

Ah, I misunderstood the diagram then. Everything makes perfect sense now.

[Ivefan]

If the thrust amounts are the same on both sides then everything should be the same. Well, unless the game includes relativistic mass, but for some reason I doubt it.

No, really, I go forward with my main engines in a straight line and then let the ship slow down by it self while watching the speed meter. If i then enable the reverse thrusters the speed meter decreases much more slowly than doing nothing at all.

[Girlinhat]

The game handles the speedometer a little weird.  You'll notice this most often when you're drifting slowly and attempt to turn, and notice your speed changing, or staying the same, when it should be dropping, because you're not applying any forward push.  The same applies to reverse thrust, it sorta wonks out when you attempt "the engines are firing but the speed is lowering" because it sorta works off the idea that speed rises with engine thrust.  In a way though, the speedometer really measures thrust output, not actual movement.

If you have an engine mounted on the front of your ship, used for turning, then it will turn the ship at a slight angle because it's not paired.  It will give a little forward push, but not really.  It's caused by the off-center-of-mass and the unequal push.  If you fired another engine opposite it, like your left and your right, then they would totally cancel and nothing would move.  To get thrust like that, you'd need forward engines as well, so you might be able to save on keybinding by making your left and right thrusters also linked to your main thrusters.  However, you should be able to get some very awkward, very slow thrust by alternating left and right quickly, leading your ship to wobble as it slithers forward, very slowly.

[GlyphGryph]

Like how you skateboard on flat ground.

[Girlinhat]

Sort of, except scateboards rely on gravity and down hills, and using kickoff force and then swerving is used to maintain movement, not really provide more movement.

[GlyphGryph]

This is how you build speed on flats if you're not pushing off - jerking left provides a tiny bit of forward momentum since the rotation for the effort is offset, as does jerking right. You can literally wiggle your way up to high speeds. It's called tic-tacing.

It's actually a bit more complicated than jerking or wiggling if you want it to work - there's a lifting and smooth motion to building up speed this way, but it works.

But the key is to imperfectly "turn" the skateboard in such a way that you provide a bit of forward momentum each time.

http://www.youtube.com/watch?v=m_6NGjXujxQ

[Shooer]

Thing is that is based off of friction.  Each push maybe focused mostly to the side, but the overall effort is forward (left motions counter the rightness of the right motions leaving forward momentum and vise versa).  The only way to get even close to that in a frictionless environment would be to have thrusters at angles we can not do in the right angled Corneroids.  And even then if you are wiggling them for forward thrust you are wasting time and effort when you could just be firing them off at the same time to go forward.

[andrea]

Hoping that I didn't miss anything, answering to Girlinhat about engine distance from rotation axis being always good.


so, what we want to achieve is angular acceleration A.
lets say our engines output a force F.
the length of an arm is r. if we consider the body of the ship of negligible diameter compared to the length of arms, then the distance between the 2 engines we are firing is 2r=d
2 forces F equal in magnitude but opposite in direction, at a distance of d, generates a torque T=F*d

I is moment of inertia. it represents resistance to rotation on an axis. We can write

A=T/I.

so far, all good. T increases with distance. but what does I do?
since we are dealing with granular mass, we can define I as sum of all m*r^2 , where m is the mass of the elemental particle ( in this case the cubes of which our starship is made)  and r is distance of the elemental particle from the axis.

what does this mean?
lets assume that the ship is rotating on the axis going from its rear to its front.
lets assume the arms are massless.
lets assume that the whole engine block has no volume. ( or, in this granular geometry, it is 1 cube big. so we don't have to deal with distributed mass)

previously, we assumed that the ship's diameter was negligible. if we keep that assumption ( which we can, since we aren't doing any number crunching yet), then moment of inertia of the ship is 0, since all mass is on the axis.
arms are massless, therefore they too contribute nothing to the moment of inertia of the ship.
all that is left is the engines, conveniently compressed.
I=2*m*r^2
where m is the mass of an engine block. it is multiplied by 2 because we have 2 such blocks, one on each side.
r is the length of an arm-

lets look back at A now

A=(2*F*r)/(2*m*r^2)= F/(m*r)

in this case, we see that as we place engines farther, our acceleration decreases!
this alone shows that it is not always best to have turning engines far away from the axis of rotation.
but is there a situation in which such outsourcing of thrust is desired?

lets rethink our assumptions, and move closer to reality.

lets assume our ship has a moment of inertia Is
lets assume the arms have a moment of inertia Ia
lets assume the engines have a moment of inertia Ie

A=(2*F*r)/(Ia+Ie+Is)

where r is length of an arm

both Ie and Ia are proportional to r^2
we can write a combined moment of inertia Ic=K*r^2, where K is a constant, depending on geometry and distribution of mass in the arms and engines.
Ms, instead, is independant from the length of the arms.

A=(2*F*r)/(K*r^2 + Is)

the only difference from the previous case seems to be Is.
what is the effect on Is on our results?

-lets assume Is is very big compared to Ic

ships can be massive, so, for low values of r this is a reasonable assumption.
We can ignore (K*r^2) since adding something small to something very big doesn't change much.

A=(2*F*r)/(Is)

now the only thing increasing by distance from the axis is torque. Which means that moving engines away is a good thing.

now lets change parameters.

-lets instead assume that Is is very small compared to Ic.

this is true for very massive engines, very light ships or, most importantly , high values of r.
we can ignore Is since it is very small compared to Ic.

A= (2*F*r)/(K*r^2)

which is the same result we got in the first case examined. now placing engines farther decreases our acceleration.
which is, needless to say, bad.

how do we apply this to a real case?

We see that, everything else being constant, there are low values of r for which A increases with distance, and high values of r for which A decreases with distance.
this mean that there is a length rc for which A is maximised. before rc, A increases, after rc it decreases.

what really matters when designing our spaceship is determining rc.
rc depends on several factors including geometry of the ship, distribution of mass ( both for Is and K), thrust/weight ratio of engines.
Which means that each ship has its own rc that has to be calculated before building it.

or you can just eyeball it. it is fine, it's just a game.

[andrea]

If that is all wrong, I think I'll need to buy a new face, because mine will be forever lost.
forever being the next hour, but that is a long time.