Forums » Bugs
On Engines, Thrust, and Magic
I have come to the conclusion that our ships work on magic. It is simply impossible to have ships that perform as ours do with stats like ours have. Below is the formula for force:
F = m*a (where F is in Newtons, m in kilograms and a in m/s^2)
Now, let's assume that a Prom as of Sunday (I'm just using rough numbers) could accelerate from 0 to 50m/s in 5 seconds, you end up with a linear acceleration of 10m/s^2. So let's plug in some numbers:
680 = 10000 * 10
680 = 100000
Hmm. Interesting equation we have there, eh? Let's see what the numbers come out using a variable for a:
680 = 10000a
a = 680/10000
a = 0.068
The formula for determinining change of velocity based on acceleration looks something like this:
V = U + AT (where V is the new velocity, U is the original velocity, A is the acceleration and T is the time)
So, let's plug in some numbers:
50 = 0 + 0.068t
50 = 0.068t
t = 50/0.068
t ~ 735
Hold it! When was the last time -you- saw a Prom that, under full thrust, took 12 minutes to get from 0 m/s to 50 m/s? Right. So we obviously have a bug. But let's try something, shall we. Let's multiply all thrust values by 100:
68000 = 10000a
a = 68000/10000
a = 6.8 m/s ^2
50 = 0 + 6.8t
50 = 6.8t
t ~ 7.4
Hey, that's a much more reasonably value (although it's still somewhat off). So what's the issue here devs? Are all the thrust factors 1/100th of their true values? Do they represent a minimum thrust? Because under turbo, the engines certainly put out more thrust than that. And, tell you the truth, that value's a little off to me. I'd put it as more like 3-4 seconds to get from 0 to 50 m/s, which means about double the force. Which means that the force on that engine is closer to 136000N. Which is still a long way from 680N.
F = m*a (where F is in Newtons, m in kilograms and a in m/s^2)
Now, let's assume that a Prom as of Sunday (I'm just using rough numbers) could accelerate from 0 to 50m/s in 5 seconds, you end up with a linear acceleration of 10m/s^2. So let's plug in some numbers:
680 = 10000 * 10
680 = 100000
Hmm. Interesting equation we have there, eh? Let's see what the numbers come out using a variable for a:
680 = 10000a
a = 680/10000
a = 0.068
The formula for determinining change of velocity based on acceleration looks something like this:
V = U + AT (where V is the new velocity, U is the original velocity, A is the acceleration and T is the time)
So, let's plug in some numbers:
50 = 0 + 0.068t
50 = 0.068t
t = 50/0.068
t ~ 735
Hold it! When was the last time -you- saw a Prom that, under full thrust, took 12 minutes to get from 0 m/s to 50 m/s? Right. So we obviously have a bug. But let's try something, shall we. Let's multiply all thrust values by 100:
68000 = 10000a
a = 68000/10000
a = 6.8 m/s ^2
50 = 0 + 6.8t
50 = 6.8t
t ~ 7.4
Hey, that's a much more reasonably value (although it's still somewhat off). So what's the issue here devs? Are all the thrust factors 1/100th of their true values? Do they represent a minimum thrust? Because under turbo, the engines certainly put out more thrust than that. And, tell you the truth, that value's a little off to me. I'd put it as more like 3-4 seconds to get from 0 to 50 m/s, which means about double the force. Which means that the force on that engine is closer to 136000N. Which is still a long way from 680N.
No, the mass values are closer to 1000 times their real values, and velocity doesn't increase linearly.
You musta been bored...
Wait, mass is 1000 times? That doesn't work at all! Even an exponential increase in velocity ( V = Ue^kt ) doesn't yield anything anywhere near correct using masses like 10000000kg and thrusts like 680N. More explaination is needed, a1k0n!
You read me backwards. The displayed mass, i.e. 10000kg, is 1000x what the real mass used in the simulation (10kg) is. We probably should change the units to kN on the acceleration to give a more reasonable picture even though we use smaller numbers internally.
Or we could just give the output power in kW or horsepower, taking the velocity damping into account to get a virtual "dyno graph" and finding the peak thrust and power points (although in our model those points would probably all be at 0 "RPM"). Then comparing ships would be as confusing as comparing cars.
Velocity gain is approximately an exponentially damped curve. I'd have to perform an inverse z transform to give you an accurate equation for velocity as a function of time, but it's of the form v(t) = vmax (1 - a-t) where a depends on mass and thrust.
Or we could just give the output power in kW or horsepower, taking the velocity damping into account to get a virtual "dyno graph" and finding the peak thrust and power points (although in our model those points would probably all be at 0 "RPM"). Then comparing ships would be as confusing as comparing cars.
Velocity gain is approximately an exponentially damped curve. I'd have to perform an inverse z transform to give you an accurate equation for velocity as a function of time, but it's of the form v(t) = vmax (1 - a-t) where a depends on mass and thrust.
Also note that when you're turboing, the -only- thing that changes in the above equation is v_max. Your faster acceleration is due only to the change in the shape of the exponential curve.
Let me understand if I'm reading you correctly... Acceleration is calculated based on current speed? So my acceleration at 50m/s is higher than at 0 m/s? How is this explained in-game? Does increasing your speed increase your thrust, or something?
You've got the graph the wrong way rogue. It's a case of diminishing returns.
velocity
|.............................x
|..................x
...........x
|......x
|....x
|...x
energy used
like so I think.
velocity
|.............................x
|..................x
...........x
|......x
|....x
|...x
energy used
like so I think.
yah, which is why certain ships can infiniboost at slightly lower speeds by tapboosting. For example, the vult mk III can infiniboost at 180 m/s since staying at that velocity requires 50 m/s. (interesting, if you were to load it down with Xith, it can actually tapboost to higher speeds, since the deceleration is slower... go figure)
Boost-tapping was removed, wasn't it? They put in a slight pause between the time you release the boost key and the time the boost disengages... Whatever, anyhow, it's still weird.
Oops, I'm wrong. There is a separate thrust figure for turbo. It was just always the same as the normal thrust figure for ages, and that's no longer the case.
So would you care to give me some actual thrust formulas? I can't update the Numbers thread if I'm unable to come up with real numbers.
It's magic, for crying out loud!
elementry physics, i love it. I tried to calculate exactly where to shoot if a ship circle dodged with my knowledge of dynamics. These numbers tricked me out. curvilinear motion, woo!
just call it all kN, the numbers work out right that way...
Sorta, but not really.
Nah boost tapping still works... The delay makes it so that you can't gain energy back, but the tapping still works
> Oops, I'm wrong. There is a separate thrust figure for turbo. It was just always the same as the normal thrust figure for ages, and that's no longer the case.
So, if turbo does have a different thrust, could we have it displayed on the ship stats screen?
So, if turbo does have a different thrust, could we have it displayed on the ship stats screen?
Wow, nice use of physics roguelazer.
I didn't know the devs payed so much attention to physics in the game! The more I find out about the specifics of Vendetta, the cooler it gets! Between the real physics, mythical faction names and ship names, and complexity of in-game workings such as the dynamics of trading, this game is, in a word, thorough. Most players probably wouldn't care about the real specifics such as these, but they're definetely worth making correct. Almost like 'easter eggs' in console games.
I didn't know the devs payed so much attention to physics in the game! The more I find out about the specifics of Vendetta, the cooler it gets! Between the real physics, mythical faction names and ship names, and complexity of in-game workings such as the dynamics of trading, this game is, in a word, thorough. Most players probably wouldn't care about the real specifics such as these, but they're definetely worth making correct. Almost like 'easter eggs' in console games.
Yeah, I noticed the same order-of-magnitude boo-boo when comparing ship stats and then looking at my real space in-game experience; and back-of-the-envelope estimates using Newton's equations of motion.
Basically my initial conclusion was that all Thrust stats quoted in vo-wiki.com (and also in-game) were 1000 times too small.
The Thrust for a Warthog Mk II (as of this writing) is listed as a force of 210N, the 'hog has an (unladen) mass of 5810kg (1 FC batt). The 'hog 2 behaves almost as if its main engines gave a thrust of 210kN force, if you allow for the non-linear damping. It flies with an acceleration (from rest, U=0) of 36 m/s^2 (not 0.036 m/s^2).
The same scale issue holds true for all ships I examined.
This is easily papered-over: label all thrust values in kilo-Newtons kN, instead of Newtons. After all, a Newton is a very small force for a ship.
-- Foo Fighter :-)
Basically my initial conclusion was that all Thrust stats quoted in vo-wiki.com (and also in-game) were 1000 times too small.
The Thrust for a Warthog Mk II (as of this writing) is listed as a force of 210N, the 'hog has an (unladen) mass of 5810kg (1 FC batt). The 'hog 2 behaves almost as if its main engines gave a thrust of 210kN force, if you allow for the non-linear damping. It flies with an acceleration (from rest, U=0) of 36 m/s^2 (not 0.036 m/s^2).
The same scale issue holds true for all ships I examined.
This is easily papered-over: label all thrust values in kilo-Newtons kN, instead of Newtons. After all, a Newton is a very small force for a ship.
-- Foo Fighter :-)