Turning speed studies

[Zaflis]

Has anyone found out how turning speed works? I did some testing on 2 of my ships, orienting towards point on station and doing 360 degrees turn sideways.

Ship 1:
- Turning time: ~43s
- Mass 1722
- Measures: 92 x 26 x 104 (meters)

Ship 2:
- Turning time: ~43s
- Mass: 498
- Measures: 34 x 18 x 34 (meters)

That's right, other has over 3 times the mass, but they turn the same speed, and it's kind of disappointing. I also made sure that camera was length-wise on shipcore level, because if camera is on the front, it might turn slower, or not. (I tested 180 turn on the big ship once for ~21s with camera in front) You can see both ships in this screenshot:
https://www.dropbox.com/s/iben5b0n9mtvxvt/size_comp.png

 

[mou]

sir we have a problem

 

[Comr4de]

Once your mass hits around 1000 the turn rate is the same. Ships below that threshold have a little bit more wiggle room.

 

[Zaflis]

I wanted to bust the myth that \"the longer the ship, slower it turns\":

Ship 3:
- Turning time: ~5s
- Mass 26
- Measures: 88 x 90 x 88 (meters)
- It\'s a 3-dimensional cross made of power and shield blocks. 0 thrust.

It\'s really about mass, not size of the ship.

 

[DracoNB]

He is saying that a ship under 500 mass and one over 1700 are the same...

 

[Moph]

Yeah, it\'s about mass only, as Zaflis said.
The reason why it seems as if ship length gives different speeds is if the camera/cockpit is located at the core, or at an extremity. At the core, the camera is at the centre of the turn so it seems normal, at an extremity, the camera/cockpit also has translational movement as well as angular movement.
The angular movement in both cases will be the exact same which is what you\'re measuring.



Also, I imagine the reason why the turning speed doesn\'t decrease further, is because it would be too problematic for standard gameplay, with regards travel. It\'s already slow enough to give a smaller ship a tactical advantage, but any slower and the pilot would likely want to kill himself.

 

[Morhe]

It really should be a bit higher, in my opinion; you can hit 300 mass accidentally by sneezing blocks onto your ship schematic.

 

[Zaflis]

Current way makes 2 problems:
- Is there a reason to limit size of your ship, if its movement and turning speed is not affected by it? Exactly as agile as smaller ship.
- What is \"small\" ship? The UFO type ship i showed in first post, falls well within small category in my opinion. But for its size it feels way too slow for me. I am really serious that i have much more fun playing with the large ship in combat, and travelling. The UFO is 1 of my \"epic fail\" designs that i\'ve left to rot on station

 

[DracoNB]

Yeah, even before, like 100 mass it starts to feel slow. I made a fighter @ around 50 mass, covered it in hull for \"looks\" (since hull does crap to prevent actual damage), and it ended up around 150ish, and felt so much slower, and was so much worse in combat because I lost the speed / manuvering advantage.

 

[DracoNB]

Here is a good explaination of why the outside cockpit seems to move fast compared to the actual core speed:







http://web.mit.edu/2.972/www/reports/differential/differential.html



The outside has to move farther than the inside, so while it looks much faster (and is...) the actual turn speed is still slow overall.

 

[Jcewazhere]

Experiments and results:


60mass, 100m long, controlled from core/center 0 thrust: ~12 seconds to do a 360

60 mass, cube shaped, controlled from core/center, 0 thrust: ~12 seconds to do a 360

60 mass, cube shaped with 237.3 thrust controlled from core/center: ~12 seconds for a 360.

60 Mass, 100m long, controlled from core/center, 237.7 thrust: ~12 seconds

60 mass, 100m long, controlled from core/center, Side facing thrusters at the far points of the \"ship\": (note, the animation for the engines still points to the rear despite making sure thruster is facing sideways) ~12 seconds

60 mass, 100m long, 237.3 thrust Controlled from the cockpit at front of ship: ~12 seconds

60 mass, 100m long, 237.3 thrust, Controlled from rear of ship: (rather disorienting controlling from the rear) ~12 seconds.



Conclusion: the only thing that matters for turning (as of the time of this posting) is the mass of the ship. Not where you control, nor how much thrust you have.

Hopefully the turning physics will be improved, either by adding \"Manuvering thrusters\" or making turning physics be bassed off of mass/thrust so you can design more manueverability or more shields/weapons/whatever into your ship.



A simple way to improve turn mechanics may be to simply change from using total mass to mass/thrust in the turn speed calculation. Obviously some tweaking in the code would need to be done to use this smaller number but that should be simple. Thrusters have diminishing returns as far as I can tell so while using mass/thrust super massive ships shouldn\'t ever be able to be as maneuverable as much smaller ships.

With this system you could add more engines, upping your thrust, to increase turn rate. Or you could choose to be more powerful in other ways.

A more complicated calculation should be used so \"fighters/bombers\" turn faster even with an equivalent mass/thrust ratio to a \"titan\". For example my current fighter is 147.5 mass (not counting turrets, I\'m not sure if they add to the total mass) and it has 204 thrust so that ratio is .723. With the current system it takes ~25 seconds for this ship to do a 360. I\'d like that to be around 10 seconds I think (I\'d need to play with it, but with this system a ship could be easily customized to turn at a rate you like, within reason). However if a ship had 72303 mass and 10,000 thrust it would also have a .723 mass/thrust ratio; but it shouldn\'t be able to turn as fast as the fighter, so some sort of consideration must be taken to prevent titans from turning like fighters.

Maybe a tier system: 0-300 mass = fighter, 301-600 = bomber and so on, each with it\'s own configurable speed multiplier. These are totally off the top of my head so feel free to suggest different tiers, or make them configurable (I love messing with config files). So a \"fighter\" could have a tier rating of 1 giving a calculation of (147.5/204)*1 so a .723 ratio. Whereas the \"titan\" could have a tier rating of .5 giving a calculation of (72303/10000)*.5 equaling a .3615 ratio. Then maybe make the ratio be 36 degrees per second * the ratio, giving my \"fighter\" a turn rate of 26.028 degrees per second. That would make it take ~14 seconds to turn 360 degrees. The \"titan\" would therefore turn at 13 degrees per second or one rotation per 28 seconds (that seems fast, maybe make the tier smaller?). Again, these are off the top of my head number that seem to work so if you have better ideas I\'m open to suggestions.

Edit: I\'m a derp, this only works if thrust>mass, would someone fix my math, while I go study for my trig test because I obviously need to >.>



I also saw a suggestion to make engines configurable like weapons, which I thought was spiffy.

And I saw a post about ships not naturally being able to reach the speed limit of the server, instead blocks are needed which increase the maximum possible speed up to the speed limit depending on mass and how many of the blocks you have.

 

[Zaflis]

DracoNB\'s note was good also, i made a principle for myself to place core not in the middle of ship, but further to back (like 1/3rd of ship length). Ship doesn\'t turn faster, but the front cockpit moves faster in relation to objects right in front of it, compared to if core was middle or front. With addition of reversing and sliding flight manoveurs, i have easier time handling combat and things.