*************** Sundry Notes ***********************************************
There is one more variable called max bank, This is found in the shiparch.ini

as per Dev: This determines how far the ship will roll to one side or the other before it will roll no more; I'm sure you've noticed that you can't exactly spin on your front-back axis (although the NPCs can). Setting this number up higher (like to, say, 80 or 90) will cause your ship to turn pretty much on its side when making hard turns at high speed.

[Collision groups these define the parts of the ship which are detachable (e.g. the middle wings on an eagle). These entries hold values for mass. They shouldn't be added to the mass of the ship however, rather they indicate the mass that is lost when they they are blown off the ship.

the table below shows how the drag increases by multiplying the linear drag figures (the coefficient) in the ini's by the current velocity. You can then see the resulting effect on excess force and acceleration
DRAG TABLE
VELOCITY, DRAG, FORCE, EXCESS FORCE, ACCELERATION
1, 600, 48000, 47400, 316.00
2, 1200, 48000, 46800, 312.00
3, 1800, 48000, 46200, 308.00
4, 2400, 48000, 45600, 304.00
5, 3000, 48000, 45000, 300.00
6, 3600, 48000, 44400, 296.00
7, 4200, 48000, 43800, 292.00
8, 4800, 48000, 43200, 288.00
9, 5400, 48000, 42600, 284.00
10, 6000, 48000, 42000, 280.00
11, 6600, 48000, 41400, 276.00
12, 7200, 48000, 40800, 272.00
13, 7800, 48000, 40200, 268.00
14, 8400, 48000, 39600, 264.00
15, 9000, 48000, 39000, 260.00
16, 9600, 48000, 38400, 256.00
17, 10200, 48000, 37800, 252.00
18, 10800, 48000, 37200, 248.00
19, 11400, 48000, 36600, 244.00
20, 12000, 48000, 36000, 240.00
21, 12600, 48000, 35400, 236.00
22, 13200, 48000, 34800, 232.00
23, 13800, 48000, 34200, 228.00
24, 14400, 48000, 33600, 224.00
25, 15000, 48000, 33000, 220.00
26, 15600, 48000, 32400, 216.00
27, 16200, 48000, 31800, 212.00
28, 16800, 48000, 31200, 208.00
29, 17400, 48000, 30600, 204.00
30, 18000, 48000, 30000, 200.00
31, 18600, 48000, 29400, 196.00
32, 19200, 48000, 28800, 192.00
33, 19800, 48000, 28200, 188.00
34, 20400, 48000, 27600, 184.00
35, 21000, 48000, 27000, 180.00
36, 21600, 48000, 26400, 176.00
37, 22200, 48000, 25800, 172.00
38, 22800, 48000, 25200, 168.00
39, 23400, 48000, 24600, 164.00
40, 24000, 48000, 24000, 160.00
41, 24600, 48000, 23400, 156.00
42, 25200, 48000, 22800, 152.00
43, 25800, 48000, 22200, 148.00
44, 26400, 48000, 21600, 144.00
45, 27000, 48000, 21000, 140.00
46, 27600, 48000, 20400, 136.00
47, 28200, 48000, 19800, 132.00
48, 28800, 48000, 19200, 128.00
49, 29400, 48000, 18600, 124.00
50, 30000, 48000, 18000, 120.00
51, 30600, 48000, 17400, 116.00
52, 31200, 48000, 16800, 112.00
53, 31800, 48000, 16200, 108.00
54, 32400, 48000, 15600, 104.00
55, 33000, 48000, 15000, 100.00
56, 33600, 48000, 14400, 96.00
57, 34200, 48000, 13800, 92.00
58, 34800, 48000, 13200, 88.00
59, 35400, 48000, 12600, 84.00
60, 36000, 48000, 12000, 80.00
61, 36600, 48000, 11400, 76.00
62, 37200, 48000, 10800, 72.00
63, 37800, 48000, 10200, 68.00
64, 38400, 48000, 9600, 64.00
65, 39000, 48000, 9000, 60.00
66, 39600, 48000, 8400, 56.00
67, 40200, 48000, 7800, 52.00
68, 40800, 48000, 7200, 48.00
69, 41400, 48000, 6600, 44.00
70, 42000, 48000, 6000, 40.00
71, 42600, 48000, 5400, 36.00
72, 43200, 48000, 4800, 32.00
73, 43800, 48000, 4200, 28.00
74, 44400, 48000, 3600, 24.00
75, 45000, 48000, 3000, 20.00
76, 45600, 48000, 2400, 16.00
77, 46200, 48000, 1800, 12.00
78, 46800, 48000, 1200, 8.00
79, 47400, 48000, 600, 4.00
80, 48000, 48000, 0, 0.00

*******************Round Up**********************************************
The figures shown below are all for acceleration at a velocity of one except for Mass and Bank.

Ship Class----- Engine Thruster Strafe Roll Yaw Pitch Mass Bank
Light Fighters
Starflier 474.01 714.01 194.01 23 2.63 2.63 100 35
Startracker 474.01 714.01 194.01 23 0.29 0.29 100 35
Patriot 474.01 714.01 194.01 23 3.21 3.21 100 35
Piranha 474.01 714.01 194.01 23 1.07 1.07 100 35
Dagger 474.01 714.01 194.01 23 1.07 1.07 100 35
Bloodhound 474.01 714.01 194.01 23 1.07 1.07 100 25
Cavalier 474.01 714.01 194.01 23 1.07 1.07 100 35
Drake 474.01 714.01 194.01 23 2.63 2.63 100 35
Hawk 471.01 714.01 194.01 23 1.07 1.07 100 35
Banshee 474.01 714.01 194.01 23 1.07 1.07 100 35

Heavy Fighters
Defender 316.01 476.01 129.34 10.6 1.19 1.19 150 30
Crusader 316.01 476.01 129.34 10.6 0.24 0.24 150 30
Wolfhound 316.01 476.01 129.34 10.6 0.24 0.24 150 20
Barracuda 316.01 476.01 129.34 10.6 1.67 1.67 150 20
Dragon 316.01 476.01 129.34 10.6 0.24 0.24 150 30
Stiletto 316.01 476.01 129.34 10.6 0.24 0.24 150 30
Valkyrie 316.01 476.01 129.34 10.6 0.24 0.24 150 30
Falcon 316.01 476.01 129.34 13.0 1.67 1.67 150 30
Centurion 316.01 476.01 129.34 10.6 0.24 0.24 150 30

V Heavy Fighters
Anubis (std) 316.01 476.01 129.34 10.6 0.24 0.24 150 30
Hammerhead 316.01 476.01 129.34 0.24 1.67 1.67 150 30
Eagle 316.01 476.01 129.34 13.0 1.67 1.67 150 30
Sabre 632.01 952.01 258.68 9.17 0.24 0.24 75 30
Titan 316.01 476.01 129.34 0.24 0.24 0.24 150 30

Freighters
Rhino 158 238 64.67 2.62 -0.36 -0.36 300 15
Mule 158 238 64.67 2.62 -0.36 -0.36 300 15
Clydesdale 158 238 64.67 2.62 -0.36 -0.36 300 15
Drone 158 238 64.67 2.62 -0.36 -0.36 300 15
Dromedary 158 238 64.67 -0.36 -0.36 -0.36 300 15
Humpback 158 238 64.67 2.62 -0.60 -0.60 300 15

This is based on the ship mass alone, excluding engines, thrusters, shields, weapons etc.
The more weapons and equipment you load on a ship the more it will slow down: (because every item has a mass)
And equally, getting rid of things like the middle wings on an eagle will improve the above:
Your roll, pitch and yaw accelerations should be left unchanged.


Edited by - bane42 on 9/12/2004 7:17:55 PM

Edited by - bane42 on 9/12/2004 7:22:56 PM

Very impressive!
Does cargo have mass as well?
Does ammo have mass?
What are the approximate masses for a weapons loadout?

yep, check out weapon-equip.ini for example and have a scroll down through it.
you will see guns/missile lauchers have a mass of 10 mostly and missiles have a mass of 1

commodities themselves in select_equip.ini dont have mass but loot crates do (10)

any of the equip files have mass entries so it all adds up to slowing u down

good work bane42....

---

a message from firebase...

---

Thks a lot. I am using this to completely rebuild the ship stats (mass, rotation, drag, strafe etc.)

Edited by - bane42 on 9/10/2004 4:32:28 AM

I am really glad this thread got stickied... This is very essential information to understanding the flight physics of the Freelancer ships.

And to answer one of the questions above... Yes, almost everything in Freelancer has some mass, though it is mostly minimal. Look in the various ini files and somewhere in the individual object's description you will find a mass entry.

Forgive me for the friggin huge-ass post, but I felt I must say something on this topic since I researched this stuff a long while back.

"For ships we need to define the excess force (the force needed to go fwd over and above drag)
-----> Law 2 Excess Force = Force – Linear Drag"

The concept of the 'excess' force moving the ship is of course correct... however, there is a critical error of assuming that the linear drag provided in the .ini files is a constant value. If it were a true constant, then giving a normal starflier an engine with only 600 thrust should cause it to remain motionless forever (all ships in normal FL have 600 linear drag, counting the 1 drag contributed by the hull and the 599 contributed by the engine). This is not true; the ship does indeed move, albeit very slowly.

What then? Instead, the value of linear drag should be viewed as a coefficient; presuming Freelancer physics is fairly conventional (which it generally is as far as motion is concerned), multiplying this coefficient by the ship's current velocity will thus give the drag force acting on the ship. Presuming this is indeed the case, as it appears to be, you would get the following:

(just a warning to those who haven't taken calculus yet, this stuff involves some integration, so if you don't know what the hell is going on, don't sweat it and jump the ugly equations)

--> m == mass; a == acceleration; F == thrust; v == velocity; t == time; ln = natural log, or log base e
1: m*a = F - DRAG
--> However, a can be written as the rate of change of velocity, or dv/dt
--> Additionally, DRAG is a coefficient times velocity, or D*v (not the same as dv!)
2: m*(dv/dt) = F - D*v
3: dv/(F - D*v) = dt/m
4: integral of (dv/(F - D*v)) = integral of (dt/m)
--> The left side was integrated with respect to v, and the right side with respect to t.
5: -(1/D)ln(F - D*v) = t/m + C
--> C is some as-yet unknown constant
--> To solve for C, note that at time 0, your velocity must also be 0, hence:
6: -(1/D)ln(F) = C
--> Substituting this back into 5 gives you:
7: -(1/D)ln(F - D*v) = t/m - (1/D)ln(F)
8: ln(F - D*v) = (-D/m)*t + ln(F)
9: F - D*v = e^((-D/m)*t + ln(f))
--> Since e^(ln(x)) = x, raising both sides to their power of e eliminates the ln on the left.
10: F - D*v = F*e^((-D/m)*t)
--> e^(a + b) = (e^a)*(e^b). Use this with the strategy for 9 to simplify the right side.
11: F(1 - e^((-D/m)*t)) = D*v
12: (F/D)(1 - e^((-D/m)*t)) = v
--> Finally, the expression for v... granted, it's rather ugly, but still.

Ok, some things to note about v and how it depends on t. First off, notice that at best (1- e^((-D/m)*t)) will equal 1; this is because e to any power will be greater than or equal to 0. Thus, the maximum velocity for the ship will be F/D. Note, however, that this speed can only be fully achieved after accelerating for an INFINITE amount of time.

Well, that's not too useful in and of itself, although it's pretty nice for doing ship-to-ship comparisons. Still, what would be better is an expression that would provide the time to reach a significant fraction of your top speed, say, 90% (call it analagous to giving a car's 0-100 time in addition to its top speed). This is easily achieved:

1: Let v = .9*max_velocity = .9*(F/D)
2: (F/D)(1 - e^((-D/m)*t)) = .9*(F/D)
3: e^((-D/m)*t) = .1
4: (-D/m)*t = ln(.1)
--> The backwards version of what was done in step 9 earlier; take the ln of e^x and you get x.
5: t = -(m/D)ln(.1) = (m/D)ln(10)

If you would like to get the time to reach, say, 99% of top speed, just replace .9 in step 1 with .99 (you will end up with ln(100) at the end rather than ln(10)). However, I happen to like 90%, so I'll stick with that.

So, we have two key equations that are fairly tolerable to look at:

--> Max Velocity = F/D
--> Time to 90% of Max Velocity = (m/D)ln(10)

Consider these for a moment. The total mass of the ship, while affecting how long it takes the ship to accelerate, has no effect on its final velocity. The thrust provided by the engines increases the maximum velocity, and so even though the time to reach 90% of the max is not altered, since the max is higher the actual acceleration must necessarily be higher as well. Hopefully this makes intuitive sense to you, regardless of the fact that the equations indicate this.

Well, that's all very good for linear motion, but what about angular motion? Since the guys that made Freelancer were kind enough to make linear and angular motion of ships function in the same manner, you can completely recycle the above equations. In place of thrust (F above), you have torque (I'll use T); in place of linear drag, you have angular drag (I'll keep D); in place of mass (m above), you have rotational inertia (I'll use i). Thus, you get:

--> Max Angular Velocity = T/D
--> Time to 90% Max Angular Velocity = (i/D)ln(10)

So... enough with equations, let's look at some hard numbers. I'm just going to do one pair of ships here, but if you really want me to I can give stats for every ship (or someone else can, doesn't really matter to me). Note that when I give the ship's mass, that value is for the ship when it is "stripped", meaning without anything visibly mounted. Internal components like the engine still add mass to the ship (for a total mass of 40), and the breakable wings will do so as well (20 for both ships mentioned here).

Considering the Eagle and the Sabre (again, I know), we can see these stats:

Eagle:
linear:: F = 48000, D = 600, m = 210
pitch/yaw:: T = 55000, D = 41000, i = 8400
roll:: T = 48000, D = 35000, i = 1000

Sabre:
linear:: F = 48000, D = 600, m = 135
pitch/yaw:: T = 43000, D = 41000, i = 8400
roll:: T = 63000, D = 41000, i = 2400

Thus, we get the following:

Eagle:
linear:: max v = 80, time to 90% max v = .806
pitch/yaw:: max v = 1.341, time to 90% max v = .472
roll:: max v = 1.371, time to 90% max v = .066

Sabre:
linear:: max v = 80, time to 90% max v = .518
pitch/yaw:: max v = 1.049, time to 90% max v = .472
roll:: max v = 1.537, time to 90% max v = .135

Poor Sabre... turning isn't everything, but it's a lot. Well, anyway, hope at least one or two people followed me through my explanation. For those that don't, well, like I said, don't feel too bad.

---

The concept of the 'excess' force moving the ship is of course correct... however, there is a critical error of assuming that the linear drag provided in the .ini files is a constant value.

---

Dev, this is not a critical error, when you reduce the engine max force to 599/600 the top speed you get is 1 starting from 0 and not using thrusters, so thats not what I would call a critical error when trying to practically judge ship's dynamics in the game. Incidentally you will get to 1 in the same time even if you change the mass of the ship to 10,000.

Yes the linear drag is a balancing force which indicates it will act as a coefficient depending on your current speed. At no point did I state that this was a fixed constant. I simply did not get into dealing with current velocities.

---

Ok, some things to note about v and how it depends on t. First off, notice that at best (1- e^((-D/m)*t)) will equal 1; this is because e to any power will be greater than or equal to 0. Thus, the maximum velocity for the ship will be F/D. Note, however, that this speed can only be fully achieved after accelerating for an INFINITE amount of time.

---

we both agree the max velocity of an eagle for example is 79/80. You can reach that speed in the game without accelerating for an infinite amount of time. I do it every day

In your examples for mass you have included the collision groups mass figure in your values (the wings on the eagle that break off etc.) This is not an addition to the ships mass, it instead indicates to the game what mass is lost when the wings break off. Take out the collision groups and lauch the game. You still have the same eagle of hull mass = 150. The only difference is that the wings dont fall off.

also your values for calculating max velocity are flawed if we go back to the concept of excess force:

you have just divided torque/angular drag, more correct imo is
(torque-angular drag) = excess steering torque force, and then divide that by drag again which changes the figures for an eagle for pitch and yaw from 1.341 to .34.

I focussed more on acceleration as in the game ppl do not fly in one angular direction for very long, they jump around in combat. Acceleration is therefore imo a more indicative and useful figure when analysing ship dynamics. But yes when rebalancing a ship, its something I take into account.

Thanks for posting, but I would prefer if ppl didnt start posting large maths lessons in this thread potentially confusing ppl, and diminishing what I am trying to do here, when what is laid out initially is valid, and can be used by anyone to really make their ships fly well, which is all I want to do .

There is no critical error in my post that will affect someones ability to understand and set good values for their ships. Unfortunately now that may appear less clear to ppl.

Bane





Edited by - bane42 on 9/12/2004 4:33:44 AM

Ok... well... I have no intention of turning this into a flame fest, since those are very stupid.

Righy-o... so here's my main beef with your equations and why I said there is a critical error (perhaps a poor choice of wording... it sounds rather caustic). In the presentation of force vs. drag and how that leads to a net force (the excess force) that is responsible for the net acceleration, drag is given as a fixed value. However, this cannot be (and I'm aware that you know this, but it's hazardous to assume that everybody that reads this will); when you are not moving, it stands to reason that the drag on your ship is zero, and when you've hit your maximum velocity, it is equally apparent that the drag on your ship exactly equals the thrust provided by your engines so that there can be no more acceleration.

Basically, the difference all comes down to this (similar argument applies to angular movement):
'Acceleration = Excess Force / Mass "
Yes, this is of course true... for that particular moment in time, which in the numbers you present is (assuming I am correct) when your ship is moving with a velocity of 1. This acceleration then continues to drop until it finally begins levelling out near the maximum velocity; that's how exponential curves look. So, giving a very high acceleration number seems very misleading to me, since this only lasts a short while.

As for this:
"we both agree the max velocity of an eagle for example is 79/80. You can reach that speed in the game without accelerating for an infinite amount of time. I do it every day"
According to the equations I have presented, assuming you could put the throttle at 100% instantly from a standstill, you could reach a speed of 79 in 1.534 seconds and a speed of 79.99 in 3.145 seconds. But, to truly reach 80 would indeed take you forever. Eventually, you probably exceed the precision of the game, so it gives up and displays an 80, but this is only a guess on my part.

Guess I might as well mention this, too:
"also your values for calculating max velocity are flawed if we go back to the concept of excess force:
(torque-angular drag) = excess steering torque force, and then divide that by drag again which changes the figures for an eagle for pitch and yaw from 1.341 to .34."
Ok... in shorter form, you're saying that max velocity = (T - D)/D = (T/D) - 1
Notice that .34 is indeed 1.34 - 1. I can't think of a good reason to universally subtract 1 from all maximum velocities, especially when you consider that many ships can't even reach an angular velocity of 1 (freighters come to mind). Since they obviously do turn, their maximum angular velocities must be positive.

Ok... I think I said all that I wanted to. Again, I am trying to argue my point, but I'm not trying to flame.

Dev
you are correct when you said that 'critical error' was perhaps a poor choice of wording.

That and the comment at the end where you said

---

Well, anyway, hope at least one or two people followed me through my explanation. For those that don't, well, like I said, don't feel too bad.

---

offended me.

There are a lot of extremely smart ppl on this forum. I never make assumptions or comments about what ppl can or cant follow. I just post the facts as I see them. I accept that you were not trying to offend anyone, but perhaps some better wording could have been used .

As a general point as I said earlier, I was trying to produce an easy to use set of figures that could produce practical results. For me the difference between 80 and 79 is meaningless. The acceleration to top speeds happen so quickly in the game that I just didnt see the point in talking too much about tail offs in acceleration etc or increasing drag.

I can use my figures to easily generate figures that makes ships perform as I expect them to. Thats all I need for FL. The finer points of maths theory are just overkill and could turn this thread into one long maths debate.People have limited time and I dont want to force them to read through pages and pages of detail.

This thread is just to give practical examples that can be easily and quickly applied to put some ship values in.

With that in mind can the moderators if they are in agreement please consider locking this thread?

Oh, come on, I wouldn't have posted anything if I didn't think there were people that would know what I was talking about. But... whatever. Not the focus of this debate.

"The acceleration to top speeds happen so quickly in the game that I just didnt see the point in talking too much about tail offs in acceleration etc or increasing drag."
Well, this is true enough; you don't really notice the acceleration approaching zero until you get pretty close to your max velocity (since m*a = F - D*v, a = (F - D*v)/m).

However, my main concern is that you use the equation (F - D)/D to obtain the maximum velocity rather than simply F/D. This creates problems with comparisons made between ships, particularly when it comes to angular velocity. Say you have a ship with 10000 torque and 10000 drag for pitch and yaw (inertia won't matter here). Now, sitting in one spot in space, you begin turning until you are going essentially as fast as you're ever going to. By picking a spot in space as your mark, you can count the seconds it takes to turn 360 degrees. Now, halve the torque to 5000 and repeat. What you will find is that it takes exactly twice as long to turn 360 degrees. Using F/D to determine maximum angular velocity, 10000/10000 gives a velocity of 1 and 5000/10000 gives a velocity of .5, so turning in a full circle would be expected to take twice as long with half the force. Now, using (F - D)/D to do the same thing will get you (10000 - 10000)/10000 = 0 and (5000 - 10000)/10000 = -.5; even ignoring the fact that one is 0 and one is negative, the ratio of 0/-.5 = 0, not 2 as it should be given the in-game results.

My point is this: when you say you have a ship with a maximum angular velocity of 1.5 and another with a maximum angular velocity of 1, I think it makes the most sense for the first ship to have a max angular velocity that is in fact 50% higher than that of the second ship. I would find it difficult to balance ship against ship if this was not so.

Totally unrelated:
"There is one more variable called max bank, which I presume is the maximum in degrees a ship can turn."
This determines how far the ship will roll to one side or the other before it will roll no more; I'm sure you've noticed that you can't exactly spin on your front-back axis (although the NPCs can). Setting this number up higher (like to, say, 80 or 90) will cause your ship to turn pretty much on its side when making hard turns at high speed.

Dev
I agree with you that my formula for max velocity is wrong and that force/drag is correct. So I will update the original posts above and also clarify the idea of drag as a coefficient of speed and add in a table which shows the increase in drag and the effect on acceleration and excess force.
I will also add in your comments re the max bank angle and mine on the collision groups.
The post should then be complete and clear to all.
Bane

quick 2 cents...if you deactivate mouse flight and hold W, in about 6 seconds you'll register 80. simple fact, easy to test...and thank you for stopping the arguing over sematics.... your sayin the same thing, everyone knows that,so thank you for agreeing to disagree that your disagreeing to agree...

Just my 2 cents...well maybe just 1....

Edited by - rtg593 on 9/14/2004 8:39:46 PM

PS...thanks, I know now why my ship goes 534 instead of 79/80:-D

Edited by - rtg593 on 9/14/2004 8:40:25 PM

Now tell us where volume(engine related) fits into these calculations.