Aurora 4x
VB6 Aurora => Aurora Suggestions => Topic started by: dooots on May 09, 2011, 12:09:01 AM
-
This was started in Proposal for TechSystem table update and in the thread it was asked to drop the off topic discussions. I would like to continue the discussion because I think its a better option then some kind of counter measure.
The idea is to add an agility bonus to fighter and gun boat engines. It would act as a speed multiplier in the chance to hit formula so that fighters and gun boats would be harder to hit then ships using normal engines.
The goal of the change is to reduce the effectiveness of a standard ASM and PD beam weapons vs fighters and gun boats. As it is now a standard ASM of equal tech can easily have about a 40-50% hit rate and PD beam weapon will have no penalty. I would like to shoot for around a 20-30% hit rate for standard ASM and a bit of penalty on the PD beam weapons. This would give fighters about a 1.75x speed multiplier and gun boats given their ability to add extra armor would probably only need about a 1.5x speed multiplier (PD beams would still have no penalty on most gun boats).
What does it add to game? It makes large groups of beam armed fighters and gun boats more dangerous to a pure missile fleet (if the fleet is not ready for such an opponent), and makes it more desirable to use laser or meson based PD. The down side is it make missile armed fighter and gun boats harder to hit but I see that as a small cost as they can still sit outside of most sensor range and fire unless the fleet has small res sensors. Plus it makes some sort of interceptor worth more then they are now. In the end it makes it harder to make a well balanced fleet which I think is a good thing.
So why would missiles not gain this agility rating? First I assume they are guided by an on baord computer which means they are dodging by following some algorithm and do it flawlessly. This makes it easier for the computer trying to target them to predict the next move. In the case of a fighter, a human is running through maneuvers they know to work but a human can almost never repeat a complicated action exactly the same. This adds randomness and makes it harder for the targeting computer to predict the next move.
Second the engines on missiles need to be as simple as possible to keep cost and size down. Adding modifications to the engine to increase agility would likely make them more complicated thus more expensive and possibly larger.
Why do normal engines not have an agility rating? Well actually they do but it is 1x so it changes nothing. They would not have the modifications because they need to be reliable, easy to maintain, durable, and efficient. Most of this is already in game as fighter engines are more likely to explode from damage, use more fuel, and have a lower HTK. Gun Boat engines are lacking the lower HTK but they receive less modifications.
So what are the modifications? Well given the limited details of the game and that ships have infinite acceleration the only option I have been able to come up with is adding the ability to increase max speed for short bursts. I know that is on the weak side but with almost no information on how the engines work its hard to come up with any ideas. For an example of how further information could add more ideas lets say the engine creates a bubble around the ship, and for a weapon to hit it had to make it into said bubble. So one modification to increase the appearance of agility would be that through modifications said bubble can be made harder to pass through for a very short time thus changing and glancing hit from a hit into a miss.
So in short add an agility rating to engines that acts as a speed multiplier for the chance to hit formula. Normal engines = 1x, gun boat engines = 1.5x, fighter engines = 1.75x. The only thing keeping larger ships from gaining the agility rating is the limit of one fighter/gun boat engine per ship. If the limit was removed this suggestion may need further modifications to balance it out.
I think I have hit all the points that where brought up in the other thread if I missed something feel free to point it out.
-
So why would missiles not gain this agility rating? First I assume they are guided by an on baord computer which means they are dodging by following some algorithm and do it flawlessly. This makes it easier for the computer trying to target them to predict the next move. In the case of a fighter, a human is running through maneuvers they know to work but a human can almost never repeat a complicated action exactly the same. This adds randomness and makes it harder for the targeting computer to predict the next move.
While I find the idea interesting and in no way wish to sink it with this criticism, I've seen this justification pop up a few times here. It doesn't hold up in two ways. Firstly the "pseudorandomness means predictability" only holds up for very large sample sizes, even with the ability to eventually predict a pseudorandomness-driven behaviour there's no guarantee that you can do so in the span of time in which a missile is in terminal guidance. As the ability to decrypt such things increases so will the ability to encrypt them, so you can't necessarily make a "it will be easier in the future" argument to defeat that point.
And secondly true randomness is actually really easy - you just sample background cosmic radiation and use that as a seed number, or even carry a bit of radioactive material on board and use the interval between decays. Any kind of sensor can give you true randomness - I've used the interval between a user's keypresses in milliseconds as a seed a few time. So the whole idea that the behaviour of computer-guided objects is predictable doesn't hold up.
So, a possible alternate justification: Missiles have an agility rating that is separate from their speed, missile engines are supposed to be just like small ships, therefore ships should have an agility rating that is separate from their speed. I think that's justification enough. The only extra thing you need to add there is that agility is used solely offensively, and that your proposal is defensive in nature. That needs a little thinking about.
As I said, I'm not saying this is a bad idea, I'm just saying that people need to stop equating computer-guided with predictable. Even modern anti-ship missiles jink randomly before impact to evade today's CIWS systems, and you can bet that if those movements could be predicted the navies of today would want to know about it.
-
I think the fundamental problem here is that anything that can track and hit a missile that is both many times smaller and faster than a fighter will have absolutely no problem in hitting the fighter.
It is hard to think of any justification that would make a fighter harder to hit.
The tailing idea from the other thread is one that could work, since it applies to both missiles and fighters but missiles getting in that close means a hit is scored.
-
I think the fundamental problem here is that anything that can track and hit a missile that is both many times smaller and faster than a fighter will have absolutely no problem in hitting the fighter.
It is hard to think of any justification that would make a fighter harder to hit.
The tailing idea from the other thread is one that could work, since it applies to both missiles and fighters but missiles getting in that close means a hit is scored.
Anything that can hit a missile easily would still easily hit a fighter with the change as it is still slower even after the multiplier. The things that get hit the hardest are ASM that lack agility to increase speed and that is my goal. As for the size nothing in the game has its hit chance scale with size.
The problem with the tailing idea is you still have to make it past the tens of millions of kilometers with missiles being fired at you, and as I pointed out a standard ASM can have a 50% hit rate on a fighter of the same the tech.
While I find the idea interesting and in no way wish to sink it with this criticism, I've seen this justification pop up a few times here. It doesn't hold up in two ways. Firstly the "pseudorandomness means predictability" only holds up for very large sample sizes, even with the ability to eventually predict a pseudorandomness-driven behaviour there's no guarantee that you can do so in the span of time in which a missile is in terminal guidance. As the ability to decrypt such things increases so will the ability to encrypt them, so you can't necessarily make a "it will be easier in the future" argument to defeat that point.
And secondly true randomness is actually really easy - you just sample background cosmic radiation and use that as a seed number, or even carry a bit of radioactive material on board and use the interval between decays. Any kind of sensor can give you true randomness - I've used the interval between a user's keypresses in milliseconds as a seed a few time. So the whole idea that the behaviour of computer-guided objects is predictable doesn't hold up.
So, a possible alternate justification: Missiles have an agility rating that is separate from their speed, missile engines are supposed to be just like small ships, therefore ships should have an agility rating that is separate from their speed. I think that's justification enough. The only extra thing you need to add there is that agility is used solely offensively, and that your proposal is defensive in nature. That needs a little thinking about.
As I said, I'm not saying this is a bad idea, I'm just saying that people need to stop equating computer-guided with predictable. Even modern anti-ship missiles jink randomly before impact to evade today's CIWS systems, and you can bet that if those movements could be predicted the navies of today would want to know about it.
Valid point the only counter I have to it is that given the rules that missiles play by is that they are dodging the entire way to the target which would increase sample size. Not a good counter but given how many times you would have to dodge given the rules of the game it might be plausible. Although as you pointed out they could use a sensor or something to make a random generator which I don't have counter for.
-
I don't think ASMs hit fighters all that well? They're kind of a waste of ordnance since the missiles cost alot and having them miss would suck. 'sides, ASMs tend to have long reload times and you have to survive at most two waves.
Well, not if the enemy has a fleet scout with a 5000ton res 1 sensor like mine... but then in that case fighters ARE kind of screwed regardless. An 80mkm warning time pretty much means you're dead unless you're a missile combatant or have incredible PD. (fleet mounts reduced size launchers)
In my Central Stars game, I'm a pure missile combatant and I think I could get a beam fighter to run at 15kkm/s if I had plasma carronades. My ASMs would miss roughly half the time and an easy 2 layer armour would mean fighters would take ~2.5 hits to kill meaning 5 missiles.
Besides, fighters would have to weather a literal storm of AMMs set to offensive fire and fighters don't really have the weight to put much armour and still retain a decent speed (due to the 1 engine limitation)
AMMs, while easier to shoot down, would be launched in the hundreds and could very easily plaster any fighter due to the small surface
area. 10 AMMs per fighter in a single salvo of AMMs would be basically impossible to survive. And cost less than the 5 ASMs required to kill the fighter.
-
The hit rate really is not that bad. For example at tech level 3 with a size 6 missile with a range of 50 mkm and a 6 damage war head you would have a 44% hit rate on a 800 ton fac or slightly higher on a 250 ton fighter. Drop the speed of the missile by 600 km's and it jumps to 47%. Now I have not taken the time to see how this all scales up as techs go up but given that fighters only get 2 tech ups (speed/armor) and missiles get 4 (fuel/speed/damage/agility) I would guess it favors the missile.
As for only two waves of missiles, at tech level 3 its easy to have a PD sensor that can see missiles at 1 mkm, that sensor would see fighters just short of 10 mkm which would take 22 minutes to cross if the target was standing still. If the enemy is also at ion engines the closing speed would be reduced from 7500 km/s to 4500 km/s if they only have 25% engines and now it takes 37 minutes.
Yes AMMs are a problem and they were before the the sensor change and I'm fine with that. Armor is much more effective against them then ASMs, they are much shorter range then ASMs, and using them on fighters means you have less to use on missiles if they show up. I do not want invincible fighters I just want them to stand a chance of making it into beam range against a fleet not prepared to deal with them.
-
What is a fleet not prepared to deal with fighters?
Every fleet has missile defence of some kind.
Be it beam PD or AMMs. AMMs kills fighters. Beam PD will always outgun fighters.
A fleet not prepared to deal with fighters is a fleet not prepared to deal with missiles.
I think the problem here is that fighters die to small caliber anti-missile fire a bit too easily.
It's not the hit rate that's the problem.
EDIT:
Actually, removing the cap on engine limitations should solve some problems. Let me think about it.
-
Actually yes, removing engine cap will do the trick.
That's all that's needed really.
Two engines lets you put on 3 more layers of armour onto the original one-engine beam fighter design, doubling total HTK, and still netting ~25% more speed (and thus be correspondingly harder to hit)
Please ignore the name and class:
Medusa - Copy class Escort Frigate 470 tons 15 Crew 127.5 BP TCS 9.4 TH 75 EM 0
7978 km/s Armour 5-5 Shields 0-0 Sensors 1/1/0/0 Damage Control Rating 0 PPV 4
Annual Failure Rate: 17% IFR: 0.2% Maint Capacity 17 MSP Max Repair 31 MSP Est Time: 2.03 Years
Magazine 1
FTR Internal Confinement Fusion Drive E750 (1) Power 75 Fuel Use 7500% Signature 75 Armour 0 Exp 100%
Fuel Capacity 10,000 Litres Range 0.5 billion km (17 hours at full power)
10cm C3 Soft X-ray Laser (1) Range 80,000km TS: 7978 km/s Power 3-3 RM 6 ROF 5 3 3 3 3 3 3 2 2 0 0
Fire Control S01 40-5000 (FTR) (1) Max Range: 80,000 km TS: 20000 km/s 88 75 62 50 38 25 12 0 0 0
Magnetic Confinement Fusion Reactor Technology PB-1.25 AR-0 (3) Total Power Output 3.75 Armour 0 Exp 20%
Size 1 Missile Launcher (1) Missile Size 1 Rate of Fire 5The missile launcher is there to take up the same space as the second engine. Coverting the launcher into an engine costs only 3BP more, insignificant.
Simply double the speed to get the new speed. And halve range to get new fuel consumption.
Note that it has 5 layers of armour 5 wide. Final speed is just under 16kkm/s too.
This makes AMM & ASM usage against them incredibly difficult. ASMs have 30-40% hit rate and still sandpaper at the armour, AMMs sandpaper away and need ridiculous numbers of missiles to kill.
Plus they can survive gauss cannon PD and dedicated laser PD pretty well. Only meson PD will totally scrap them.
You can also drop the armour by 1 layer + halve range, go up to 500tons (15kkm/s final speed) and mount a 15cm C6 laser + associated power.
-
What will the removal of the engine limit do to the effectiveness of missile-armed fighters? It's good to see effective armoured beam fighters, but I'm a little concerned that anything that increases the effectiveness of beam-fighters will make missile fighters crazy-good.
-
I don't think that removing the engine limit will make much of a difference for missile fighters. What makes missile fighters effective is their long range and heavy payload in comparision to their size. Having a second engine is going to up the fuel required as well as upping the total size of the fighter. While it will increase the speed, that has less effect than something to increase the range without changing the size of the fighter. With a little planning missile fighters can stay out of effective missile range, unless the ships firing on them are very specialized designs with specialized missile designs.
brian
-
Well I'm dropping this idea, as it is now if you turn a fac into a flying ball of armor (engine, small fuel, small crew quarters , and armor) with this idea it becomes way to expensive to kill with ASM which I am sure NPR's will happily shoot at you (you could break even at low techs and I expect it to become more cost effective at higher tech levels). Worse is with some shield tech and luck it becomes possible to fly in and out of sensor contact and drain a fleet of all its ASMs for almost free.
If the engine limit gets dropped it would become even worse as you could still fairly cheaply fly a larger ship with fac engines and have little risk of it dieing before you could pull it out for repairs. If this would be a problem with out the multiplier I'm not sure but someone may want to look into it.
-
I have to disagree with a lot of the posts above, we have missiles that work on interia based system where it is possible for them to miss. If it was inertia less then it would be impossible to miss if the missile travel at a higher speed then the space craft. I am happy to argued on this but you would be wrong <smile>
We have space craft which is interia less. So manuvuering is out as a dreadnought can turn in space as fast as a fighter.
But the basis of treat missile AI the same as a human pilot, I don't think so, the AI missile may dodge a little, or zig its way to the target but, it would not do a loop, roll, jink and fly 180 from the missile, so you cannot treat a missile AI the same as a human pilot because the unpredictability factor increases 100 times. You could create a manuvuer rating with the commander if you liked.
Now if increasing of fighter speeds just make them more armour facs it make no sense either.
What we need is fighter countermeasures against missiles, to allow beam fighters a chance to close in. I am not talking ECM, but things like thermal flares, that can create enough thermal signature to confuse the missile. Mix that with a 6 or 8cm Beam weapon you would have a fighter that runs roughly the same speed as a missile fighter, but has an chance ability to close on a missile based ship.
-
It is still possible to miss with inertia-less drives. Simply put, your target is also moving, unless you can predict the target's movement, you can still miss. And the closer the speed match, the easier it is to "dodge".
Especially at the speeds we're dealing with, in the thousands/ten thousands of km/s, considering the size of the ship, the time that a missile has to "react" to a ship's last minute jinking is almost nil.
As the missile flies it, it has to attempt to follow a ship which is running around on an inertia-less drive, which could change direction basically instantly. Can a missile even receive the information that the ship "dodged" and change it's thrust to match that fast? I presume this is what the agility rating is for.
AI can be easily made to be pseudo-random enough that it's basically impossible to predict what it will do in any reasonable time. It would simple to give the AI an algorithm that generates random output such that working out the algorithm from observing the missile is an NP complete problem (of which you need to start over when the missile inevitably swaps to the next iteration a millisecond later) and determining what will happen next is far far harder than running the algorithm in the first place. Then the missile just needs to run through the random dodging routine faster than you can work it out.
And it's not like you have a few days to sit around and watch the missile do it's thing. You have a few hundred of them showing up and two minutes later, you're either space dust or they got shot down.
Thermal flares are fine for missiles that run on onboard thermal sensors. But unless you have an active sensor spoof (which would be an awesome addition to ECM), most missiles are radar guided from the firing ship.
-
You cannot dodge purely the fact that a missile can turn 180 degrees on the spot, and maintain momentum, so there is no dodge have a think about it. You could not avoid a missile, if you plot it ever millisecond on a piece of a paper where you can turn at the same speed with no loss of momentum. You could not dodge.
Also the reaction time on a AI is far faster then a human.
So you combine a sqwaker and a thermal together, think submarines not fighters and you will know what I mean.
-
But the basis of treat missile AI the same as a human pilot, I don't think so, the AI missile may dodge a little, or zig its way to the target but, it would not do a loop, roll, jink and fly 180 from the missile, so you cannot treat a missile AI the same as a human pilot because the unpredictability factor increases 100 times.
Why would an AI not be able to do those things? It's not for mechanical complexity, every one of those things can be coded fairly simply, especially in an environment without obstacles! The state of machine learning now is such that making a drone aircraft do tricks or making a missile unpredictable is pretty easy. Imagine what it would be like with TN advances in computing. This idea that human pilots are going to be more evasive is not a good justification, and it's Hollywood, not computer science, that spawned the idea.
-
You cannot dodge purely the fact that a missile can turn 180 degrees on the spot, and maintain momentum, so there is no dodge have a think about it. You could not avoid a missile, if you plot it ever millisecond on a piece of a paper where you can turn at the same speed with no loss of momentum. You could not dodge.
Also the reaction time on a AI is far faster then a human.
Reaction time of humans never matter. The ship is not "doding" the missile actively. The ship is just changing directions in unpredictable ways as fast as possible.
You just need to change directions faster than the missile to get it to miss. The higher the missile's speed advantage, the faster you need to do it.
The agility rating is how fast the missile processes changes in target speed and adjusting course to match.
So you combine a sqwaker and a thermal together, think submarines not fighters and you will know what I mean.
A squawker is essentially an ECM emission made to look like a ships' return on the grav-radar.
Which is a nice addition and definitely one I would like to see in.
-
Again changing of directions without interia is instant. So it either reaction time that make it miss (which I am assume your getting at with processing information, in that respects we talking about AI response not speed of missile), or it cannot miss based on a interia-less environment.
Unless you are stating ships are interia-less and missile have interia.
-
I don't think missiles have inertia i think hits are not the missiles hitting the hull but the shaped warheads detonating in/out of range.
what happens is the the ship and the missiles all are moving at high speed some missiles attack from above, some below, etc so if the ship moved down at the time of missile detonation it would move in the range for some of the explosions and out of range for others.
-
So it either reaction time that make it miss (which I am assume your getting at with processing information, in that respects we talking about AI response not speed of missile)
Sort of like this.
And only the reaction of the missile matters. The ship just needs to jink blindly faster than the missile can react.
The reaction of the missile is a combination of sensing that the ship isn't where it was, AI making needed course changes and drive responding to AI commands. All this takes time, and is harder than blindly changing to random new direction.
-
Again changing of directions without interia is instant. So it either reaction time that make it miss (which I am assume your getting at with processing information, in that respects we talking about AI response not speed of missile), or it cannot miss based on a interia-less environment.
Unless you are stating ships are interia-less and missile have interia.
I think what is happening is that the missiles are moving at very high speeds, and the ships are the same, the missiles detonate not when they hit the ship but when they get close enough, and as such sometimes the missiles miss because its like trying to hit ant from a kilometre away, if you are off by a fraction of a millisecond the ship is already kilometres away.
-
Again changing of directions without interia is instant. So it either reaction time that make it miss (which I am assume your getting at with processing information, in that respects we talking about AI response not speed of missile), or it cannot miss based on a interia-less environment.
Unless you are stating ships are interia-less and missile have interia.
What's 'interia'? I'll assume you meant inertia. ;)
If intercept was a purely movement mechanic then you'd be correct, but your not. There are aspects that are abstracted and not modelled, one of them being detailed evasion maneuvering.
-
Don't bother correcting my spelling it be a lifetime habit for you, and I still wouldn't care ;D
Whether the missile hits the ship directly or hits by explosion is a mute point it needs to hit an area that causes damage, that area may be dead space but it still classified as an effective target area. All we are saying here is the ship target size is increased for the effectiveness of the missile.
So we come back to my original point, now that we are talking about reaction times, and if we assume 'agility' is to do with AI processing, and that processing is fairly consistent across missile. We now talking about the random variable which is pilot skill, unless we assume that the piloted craft uses AI enhanced ship combat manoeuvres.
I suppose the assumption is present all AI processing is created equally, which it is not you can look at missile here that have different ratings.
In that case there is nothing stopping putting an agility score in either the engine or piloting.
QED.
It now be interesting which way you change the argument direction this time :P
-
I did justify that.
Only the agility of the missile matters in my argument. The ship just needs to jink very very fast, it doesn't need to know where the missile is.
Hence, agility is purely offensive. It's the ability to track and compensate for very rapid changes in the target's speed.
When jinking to avoid shots, that's not needed at all.
-
Sorry but my imagination of thinking a craft is jinxing randomly whenever someone fires something at it, and moves at the same time in the direction of the course at maximum speed, made my head explode.
It was like me thinking about time travel and trying to alter history only to find out history never altered.
-
Sorry but my imagination of thinking a craft is jinxing randomly whenever someone fires something at it, and moves at the same time in the direction of the course at maximum speed, made my head explode.
As evasive manoeuvres are abstracted into the object's speed, this is exactly what's happening in Aurora all the time. It is quite head-explodey, but it's true. Not necessarily "randomly" jinking (as in Brownian motion style movement), but certainly taking evasive action while maintaining a constant course and speed. Don't try to understand it in terms of realism, it's a gameplay abstraction.
(Also, I'm not actually following this currently side-discussion, but I will say that knowing where a missile is makes even "random" jinking more effective - you only jink on the plane perpendicular to the missile, so you move your as far as you can relative to the missile)
-
Sorry but my imagination of thinking a craft is jinxing randomly whenever someone fires something at it, and moves at the same time in the direction of the course at maximum speed, made my head explode.
It was like me thinking about time travel and trying to alter history only to find out history never altered.
Something to remember is that Aurora is not a physics simulator - it's a game. Certain things have been abstracted away in favor of playability. One of these things is the player knowing about and/or deciding "how much of my speed am I going to devote to jinking, and how much to moving where I want to go". Steve considered putting this complexity in, and decided not to do it. It is assumed that ships can be jinking around while they're moving without a movement penalty.
John
-
I understand it is a game ::), my posts are a theoretical discussion on how the game works in reality, based on the physics principles provided.
If I wanted to post about the game design I would say this.
Beam fighters broken, missile are overpowered, fighter are just a way to provided longer range missiles. Balance is wrong. There is no rock paper scissor for missile design.
Above is a personal choice about my views, steve who designed this great game has 'rightfully so' other ideas, that why I was deliberately avoiding the 'remember is was a game not a physics simulator' aspect.
If we looked at the argument based on this being 'a game' there is no reason why not to add agility to fighters, no reason not to add ion cannons that knock out an enemies power temporarily or any other aspect. Because in a game you can make and choice or direction and match up the physics after to create the story why.
If Steve decided tommorrow he wanted to include small craft with agility, I am sure the same faces in this post would come up with a plausible reason how it was included, and do a 180 backflip.
I wanted to wrap my head around based on the current principles of physics 'in the game', why you could or could not have agility in engine included. And I now believe I understand it, based on this random jinking concept.
-
The idea that you can either think about something as a game (in which case anything goes) or as a simulation (in which case realism (or at least verisimilitude) is paramount) is a false dichotomy. It's a continuum, and Aurora is quite far towards the "simulation" end of the scale. There are, however, still points where abstractions made for game design or implementation purposes induce a disconnect in that verisimilitude. This is one of those points.
Claiming that "There are specific points in this game at which gameplay trumps simulation" equates to "you can make any choice or direction and match up the physics after to create the story why" is not sensible. The point people were making to you is not "It's a game, don't talk about reality at all" but "This is one point in the game at which you cannot use a realism-based argument as abstractions built into the system negate it".
-
OK OK I surrender
Attacking my intelligence with the obvious after another obvious stated is now becoming a gang bang ;D
I am a programmer myself, I understand abstraction and compromises. What I wanted to understand was the blueprint which I got.
-
I think this topic has run its course and the horse has been sufficiently beat. To prevent further bloodshed, I'm closing it.