Additional orders? Nope. not part of the plan.Daveybaby wrote:Give them orders, then watch the mess unfold. Then try and give meaningful additional orders to that mess.
ship technology
Moderator: Oberlus
-
- Large Juggernaut
- Posts: 938
- Joined: Fri Jan 20, 2006 9:34 pm
- Location: GA
So, the complexity of your preferred model for ship movement tightly constrains what's possible for the space combat engine. Great. No thanks.marhawkman wrote:Additional orders? Nope. not part of the plan.
Q: I've decided that all footballs will be painted green from now on.
A: But that will it very hard for players to find the ball against the grass.
Q: Players? Ive decided to do without them, since they conflict with my plans for green footballs.
Completely and utterly missing the point. Space war was a great game, but it amply demonstrates how messy newtonian movement can get, even for 1v1 combat.Zpock wrote:Ever played moo3? That game is a clear example of why making a space 4x is a bad idea.
The COW Project : You have a spy in your midst.
Please, my point is that pointing at a bad implementation of newtonian movement to prove that the concept of newtonian movement itself is flawed is as silly as saying the concept of a 4x is bad since moo3s implementation of that is bad. Was that really not clear? I guess I have to start spelling everything out no matter how obvious.Completely and utterly missing the point.
But yes i agree, it does prove that newtonian movement can get messy. Probably, bad implementation can make anything messy (see moo3). Dosn't mean it has to be messy tough.
-
- Creative Contributor
- Posts: 441
- Joined: Tue Oct 07, 2003 1:00 am
Another Blashphemor! Ye shall be sacrificed to the almighty gods DOGAR and KAZON for that remark!Impaler wrote:Star Conrol only alows you to control 1 ship at a time which resulted in the rather rediculus "lets have 2 fleets destroy each other by a long drawn out series of 1 on 1 duels!". Admitedly their implemtation was very good, the battles were tacticaly interesting and exciting, the simply payed a huge price in bottlenecking the whole game through thouse duels. A 4X game cant stand that kind of bottleneck and a human cant control a more complex battle under newtonian physics (atleast with anything more then trivial acceleration/deacceleration).
There are three kinds of people in this world - those who can count, and those who can't.
Newtonian movement is what it is - i.e. pretty straightforward mathematically (although with complex emergent behaviour - just look at 3 body gravity simulations) - so there isnt really much elbow room when it comes to implementation, i.e. ships have to spend just as long decelerating as they do accelerating, and there is no effective top speed (unless you want to drag relativity into this as well ).Zpock wrote:But yes i agree, it does prove that newtonian movement can get messy. Probably, bad implementation can make anything messy (see moo3). Dosn't mean it has to be messy tough.
So you end up with a bunch of ships accelerating toward each other, trading shots while theyre in range, then shooting past each other and frantically decelerating again. The end result, i guarantee, will be a screenful of ships yoyo-ing back and forth around the screen. A mess.
You can try to mitigate this by using lots and lots of AI, but this is adding lots of extra work for very little gain - work that would be much better spent making the AI play a challenging strategic and tactical game. For what its worth, this is exactly why Moo3 ended up being so bad - an overly complex engine which required a layer of AI between it and the player in order to keep the game managable, resulting in the fact that the player felt they had no control over the game.
Or you can use some other method, such as a some maximum speed limit for some arbitrary reason, or some arcane technology that allows rapid deceleration compared to acceleration, or make weapon range great enough that ships dont have to manouever towards each other to engage the enemy - but if youre going to do that then why bother having a newtonian model at all?
Trust me, i know what i'm talking about. Years and years ago (like, way back in the atari ST days) i started to knock up a little game based on tactical combat between small fleets of ships using a newtonian model. I rapidly come to the conclusion that all i was doing was loads of work to make ship control managable, when I should have been spending my effort making the game strategically or tactically interesting.
The COW Project : You have a spy in your midst.
As I wrote earlier I would introduce either energy (regenerates) and/or fuel (dosn't regenerate) as a limit to how much you can deecellerate or accelerate. If you really knew what you were talking about concerning newton models you would know any force is matched by an equal opposite force so you can't keep accelerating a starship forever unless its made of 100% fuel. But since I am not really interested in realism I could certainly go for the energy model with unlimited fuel to solve the problem of what happens when a ship runs out of fuel and can't do anything to change where it is going.
So the result would be that you would get a kind of resource (energy/fuel) that is spent to do different maneuvers on the battlefield. A huge battleship might only get a chance to do one high speed attack run for example, so you have to choose wisely your maneuvers. I would probably also only use the newtonian model as a base and add in stuff to spice it up like teleporting or short range hyperdrive devices and stuff that lets you do things like instantly change velocity vector. This would be exotic and limited technology of course. Just to clear things up, I am also of course thinking in the classical way that combat movement (thrusters) is separate from movement between stars (hyperdrive).
Something that is needed is a good UI so that the player can easily give orders to his ships how they will move. This is what I mean by implementation is important. For example some basic control would be that a single click somewhere would give a ship an order to move to that spot in a slower conservative speed saving fuel/energy. A double click would tell the ship to go there at maximum speed. This would include deacelerating to stay at the spot clicked. Dragging could mean a ship will continue in that direction forever. A right click could bring up a meny with more controls for maneuvers like orbit this point.
You could probaly implement the whole maneuvering thing using some other type of movement system that you think is easier to understand, but I don't think it would be much simpler. If you want simple just skip it all and just have movement as a means of getting your guns in range for the slugfest. Then all needed is a set speed ships move at.
Hopefully Im going to do what you did and run a few tests with this kind of stuff myself.
So the result would be that you would get a kind of resource (energy/fuel) that is spent to do different maneuvers on the battlefield. A huge battleship might only get a chance to do one high speed attack run for example, so you have to choose wisely your maneuvers. I would probably also only use the newtonian model as a base and add in stuff to spice it up like teleporting or short range hyperdrive devices and stuff that lets you do things like instantly change velocity vector. This would be exotic and limited technology of course. Just to clear things up, I am also of course thinking in the classical way that combat movement (thrusters) is separate from movement between stars (hyperdrive).
Something that is needed is a good UI so that the player can easily give orders to his ships how they will move. This is what I mean by implementation is important. For example some basic control would be that a single click somewhere would give a ship an order to move to that spot in a slower conservative speed saving fuel/energy. A double click would tell the ship to go there at maximum speed. This would include deacelerating to stay at the spot clicked. Dragging could mean a ship will continue in that direction forever. A right click could bring up a meny with more controls for maneuvers like orbit this point.
You could probaly implement the whole maneuvering thing using some other type of movement system that you think is easier to understand, but I don't think it would be much simpler. If you want simple just skip it all and just have movement as a means of getting your guns in range for the slugfest. Then all needed is a set speed ships move at.
Hopefully Im going to do what you did and run a few tests with this kind of stuff myself.
-
- Large Juggernaut
- Posts: 938
- Joined: Fri Jan 20, 2006 9:34 pm
- Location: GA
I'm not gonna bother replying to that many posts specifically. I'm just gonna describe it the way I think it would work best.
1: accel/decel is determined by your ship's speed.
2: turning is independent of accel
3: if a ship's engines are disabled it will keep moving in whatever direction it was moving before
4: since the AI is gonna be handling movement, braking can be done to make the ship stop movement completely. (realistically you wouldn't be directly controlling the movement of a ship anyways unless it was something like a fighter)
5: ships turn to shrapnel when blown up and the shrapnel keeps moving the same as the ship was. ships that get hit by the shrapnel will take damage.
6: there are weapons that can be used to affect the speed and course of ships.
note: I feel it should be mentioned that this system wouldn't have decel as something only done to stop after passing an enemy. You could decel so that you stopped on top of them.
1: accel/decel is determined by your ship's speed.
2: turning is independent of accel
3: if a ship's engines are disabled it will keep moving in whatever direction it was moving before
4: since the AI is gonna be handling movement, braking can be done to make the ship stop movement completely. (realistically you wouldn't be directly controlling the movement of a ship anyways unless it was something like a fighter)
5: ships turn to shrapnel when blown up and the shrapnel keeps moving the same as the ship was. ships that get hit by the shrapnel will take damage.
6: there are weapons that can be used to affect the speed and course of ships.
note: I feel it should be mentioned that this system wouldn't have decel as something only done to stop after passing an enemy. You could decel so that you stopped on top of them.
Computer programming is fun.
This is what I was saying in one of the movement types I listed. In space there is no friction, so stopping your engines does not stop you. The only way to stop is through applying opposite thrust. It would make it interesting, if your ships drifted in this way, so that if there engines were disabled, then they would continue to drift, maybe even away from the battle, if they are lucky or into the sun.marhawkman wrote: 3: if a ship's engines are disabled it will keep moving in whatever direction it was moving before
You misunderstand how things work. If a ship moves under power in a direction, and the engines stop, well the ship would keep moving, forever. You see, the ship would not decelerate because in space there is no friction.Zpock wrote:If you really knew what you were talking about concerning newton models you would know any force is matched by an equal opposite force so you can't keep accelerating a starship forever unless its made of 100% fuel.
So while it is true the ship cannot accelerate forever, it can move in a direction forever. It does not have to accelerate much at all to move in a direction forever. It could accelerate for 5 seconds and that would be enough, moving forever in a direction, unless it decelerated through reverse thrusters or was affected by an asteroid impact, etc.
I have no idea where I wrote that a ship would stop by friction. What I meant is that they yo-yo effect could be avoided in a newtonian system if the ships generally accelerate in short bursts instead of slow constant acceleration. So for a simple example if a ship is going somewhere 100m away it accelerates the first 10m then goes 80m in constant speed while recharging its engine power to deecelerate the last 10m to a stop.You misunderstand how things work. If a ship moves under power in a direction, and the engines stop, well the ship would keep moving, forever. You see, the ship would not decelerate because in space there is no friction.
Last edited by Zpock on Thu Mar 16, 2006 1:34 am, edited 1 time in total.
I don't think that implementing Newtonian physics would be very fun. We could say that forward motion is limited by artificial gravity, <insert bogus explanation>, so that ship movement is limited and fits into a turn-based system.
There are two cases where I think Newtonian physics might be fun:
1. Disabled engines mean the ship keeps moving at a constant speed. Once that ship leaves the combat area, there should be a chance (helped by onboard damage control) of repairing the engines, otherwise the ship spirals into the sun. If that player wins the battle, the other ships in the fleet may put the disabled ship into a stable orbit or repair it (depending on whether we build dedicated repair ships or not). If that player loses the battle, the enemy can choose to let it drift into the sun, or blow it up, or board and capture it.
2. The exploded ship. It would be kinda fun to have a shrapnel cloud move across the battlefield at the speed the ship was originally going (if it had moved 3 spaces the previous turn, the cloud would go 3 spaces on its turn), gradually dispersing and becoming less deadly.
There are two cases where I think Newtonian physics might be fun:
1. Disabled engines mean the ship keeps moving at a constant speed. Once that ship leaves the combat area, there should be a chance (helped by onboard damage control) of repairing the engines, otherwise the ship spirals into the sun. If that player wins the battle, the other ships in the fleet may put the disabled ship into a stable orbit or repair it (depending on whether we build dedicated repair ships or not). If that player loses the battle, the enemy can choose to let it drift into the sun, or blow it up, or board and capture it.
2. The exploded ship. It would be kinda fun to have a shrapnel cloud move across the battlefield at the speed the ship was originally going (if it had moved 3 spaces the previous turn, the cloud would go 3 spaces on its turn), gradually dispersing and becoming less deadly.
Maybe in the particular model you've dreamed up that's the case, but fuel doesnt have to be an issue in a short term space combat unless you want it to be - its not inherently part of a newtonian combat model. I know what i'm talking about - but that doesnt mean i'm necessarily taking into account stuff you made up.Zpock wrote:If you really knew what you were talking about concerning newton models you would know any force is matched by an equal opposite force so you can't keep accelerating a starship forever unless its made of 100% fuel.
So why bother with the newtonian system at all? For all the massive extra complexity youre handing the player (not to mention the enemy AI) what significant gameplay benefits do you get over a simple system where ships have simple system of motion with a fixed maximum speed?But since I am not really interested in realism I could certainly go for the energy model with unlimited fuel to solve the problem of what happens when a ship runs out of fuel and can't do anything to change where it is going.
Most of the rest of your post lines up with the points i made in my previous post - by adding the fuel system into the mix youre either adding even more complexity (bear in mind we could be talking about controlling dozens of ships at a time here) with the associated UI and AI issues, or you compromise by limiting newtonian motion to a level where you may as well not bother with it at all.
The fundamental issue i have with a newtonian system of motion is that youre basically changing the significant problem of space combat from "where do i want to position my ships in order to gain a tactical advantage?" into "how am i going to convince the game to put my ships to where i want them to be?". This is on a par with the player having decide what direction to fire weapons (taking transit time and enemy speed and distance into account) instead of just saying "fire at that ship over there" - its something that would be *great* if you were only controlling 1 ship at a time (or had hours to play the combat through with multiple ships) - e.g. something like 'star fleet battles', but for a tactical subgame it's focussing at entirely the wrong level.
The COW Project : You have a spy in your midst.
-
- Large Juggernaut
- Posts: 938
- Joined: Fri Jan 20, 2006 9:34 pm
- Location: GA
BUT..... That's only an issue if the player is directly controlling the fleet. If not then we can just let the AI do whatever the player told it to do.Daveybaby wrote:The fundamental issue i have with a newtonian system of motion is that youre basically changing the significant problem of space combat from "where do i want to position my ships in order to gain a tactical advantage?" into "how am i going to convince the game to put my ships to where i want them to be?". This is on a par with the player having decide what direction to fire weapons (taking transit time and enemy speed and distance into account) instead of just saying "fire at that ship over there" - its something that would be *great* if you were only controlling 1 ship at a time (or had hours to play the combat through with multiple ships) - e.g. something like 'star fleet battles', but for a tactical subgame it's focussing at entirely the wrong level.
Computer programming is fun.
Let me try to put this in a simple way. If you are floating in the middle of space somewhere, to move yourself somewhere you have to throw something in the other direction. So a spaceship would do the same thing, throw fuel out the back in order to accelerate forward. I wish it was me who came up with these basic physics but sadly someone in the 15th century dreamed it up before me.Maybe in the particular model you've dreamed up that's the case, but fuel doesnt have to be an issue in a short term space combat unless you want it to be - its not inherently part of a newtonian combat model. I know what i'm talking about - but that doesnt mean i'm necessarily taking into account stuff you made up.
Wrong.Zpock wrote:Let me try to put this in a simple way. If you are floating in the middle of space somewhere, to move yourself somewhere you have to throw something in the other direction. So a spaceship would do the same thing, throw fuel out the back in order to accelerate forward.
f = ma
You dont throw fuel out the back - you fire reaction mass. Fuel (i.e. energy) is used to accelerate this reaction mass. The more acceleration you apply to the mass the greater the acceleration of your ship. So by spending more energy you can get a larger reaction from a small mass to accelerate it to very high speeds. So energy is the key here, not reaction mass - and if you have a highly efficient energy source (such as, i dunno, matter/antimatter annihilation) then energy is practically free in terms of fuel mass.
Besides all of the above - so what?. The model you have invented for yourself out of thin air may very well involve fuel/reaction mass supplies that are insufficient to last more than 1 space combat, but this is just something that exists in your head. If the rest of us decide that maybe we can have supplies which last much longer than that, to the extent that fuel becomes a non-issue for combat, then it doesnt mean that we dont understand the physics, it just means that your particular preference isnt the only possible way that things are ordered.
All you keep saying is: "Youre wrong you dont understand physics, because i've decided that i'm going to paint my ships green".
The COW Project : You have a spy in your midst.