|
Post by caiaphas on Feb 17, 2017 18:25:15 GMT
I think any gun intended for use at very long range will necessarily fire guided projectiles. Even if velocity dispersion out of the barrel is zero, there's some uncertainty in the position of the target and in orbital mechanics predictions. But that doesn't necessarily mean gimbals and turbopumps and fuel tanks. A li-ion battery and electric solid propellants could do it. Of course the now guided projectile either receives command guidance or has its own sensor suite. I think that at the ranges we're discussing a rudimentary sensor suite would be somewhat necessary; I mean, my missiles are already designed to engage at 1 Mm and close at 14 km/s from a standing start. At those velocities and at that range it would take a shipboard command suite about 7 milliseconds to receive information from the enemy ship (e.g. its thermal signature, its location relative to the ship, yadda yadda) by which time my missile has covered a full hundred meters of space and might've missed a high-gee jinking enemy ship already. And frankly our guided munitions are probably going to be smaller (since hopefully we're going to get smaller propulsion units fingers crossed), are engaging at longer ranges, and are going to be traveling faster, making that issue worse.
|
|
|
Post by Easy on Feb 17, 2017 18:43:57 GMT
Of course the now guided projectile either receives command guidance or has its own sensor suite. I think that at the ranges we're discussing a rudimentary sensor suite would be somewhat necessary; I mean, my missiles are already designed to engage at 1 Mm and close at 14 km/s from a standing start. At those velocities and at that range it would take a shipboard command suite about 7 milliseconds to receive information from the enemy ship (e.g. its thermal signature, its location relative to the ship, yadda yadda) by which time my missile has covered a full hundred meters of space and might've missed a high-gee jinking enemy ship already. And frankly our guided munitions are probably going to be smaller (since hopefully we're going to get smaller propulsion units fingers crossed), are engaging at longer ranges, and are going to be traveling faster, making that issue worse. It can work at longer ranges and still be relatively dumb with command guidance. Lets say you spit out a thousand little buggers with an initial command to set up into a pattern and which part of the pattern it is in. The Kdrone doesn't need to know where it is, it doesn't need to know where the other kdrones are, the kdrone doesn't even need to know where the target is. It just needs to know its own name and what direction it is pointing so that it can recognize what commands it needs to follow and be able to complete them. Think of it like a shotgun, but you can steer the pattern and make it whatever shape you want.
|
|
|
Post by underwhelmed on Feb 17, 2017 18:51:22 GMT
Some thoughts:
1. Sensor functionality, resolution, and accuracy are non-trivial... But really accurate modeling based on physics is impossible. Parametric equations for performance would be a much more realistic option. Additionally, type of sensor matters too. Passive IR is pretty bad at figuring out things like velocity and heading... Active doppler radar would be good for that, but it's vulnerable to electronic warfare and signal strength scales with 1/r^4. Then you get into wavelength, signals processing, etc.
2. Current system of engagement range is silly because it artificially limits the maximum range of weapons, especially kinetics.
3. Naively aiming at a single point makes dodging a lot more effective than it would be vs blanketing the possible maneuver envelope. In this respect, precision actually is a disadvantage.
|
|
|
Post by omnipotentvoid on Feb 17, 2017 21:10:50 GMT
I think that at the ranges we're discussing a rudimentary sensor suite would be somewhat necessary; I mean, my missiles are already designed to engage at 1 Mm and close at 14 km/s from a standing start. At those velocities and at that range it would take a shipboard command suite about 7 milliseconds to receive information from the enemy ship (e.g. its thermal signature, its location relative to the ship, yadda yadda) by which time my missile has covered a full hundred meters of space and might've missed a high-gee jinking enemy ship already. And frankly our guided munitions are probably going to be smaller (since hopefully we're going to get smaller propulsion units fingers crossed), are engaging at longer ranges, and are going to be traveling faster, making that issue worse. It can work at longer ranges and still be relatively dumb with command guidance. Lets say you spit out a thousand little buggers with an initial command to set up into a pattern and which part of the pattern it is in. The Kdrone doesn't need to know where it is, it doesn't need to know where the other kdrones are, the kdrone doesn't even need to know where the target is. It just needs to know its own name and what direction it is pointing so that it can recognize what commands it needs to follow and be able to complete them. Think of it like a shotgun, but you can steer the pattern and make it whatever shape you want. How dumb the projectile can/needs to be would depend on the purpose of the round, the sensory equipment deployed by the fleet already and the tactical situation. For instance, having several drones deployed to form a positioning system would allow for direct guidance to the target with minimal hardware on the projectile itself. Offboard sensory packages close enough to tell in which way the enemy is thrusting would further improve the accuracy. Allowing drones to guide the projectile also reduces signal travel time, making targeting even more precise. This would allow projectiles to be small and dense, improving damage inflicted. The amount of munitions that can be guided in such a way may be limited by the processing power of the packages. Also such dispersed sensory systems may be easily disrupted by attacking its constituent parts. In the absence of offboard sensory packages or the like, the projectile must be guided by the firing ship or its own guidance/sensor systems. The latter has the advantage of being fire and forget, as well as being capable of achieving high hit rates. The downsides are large more expensive munitions that may not do as much damage, as well as possibly being limited in acceleration and thus velocity. The former requires area saturation, thus using more ammunition, and requires the firing ships processing power. It does allow for smaller, cheaper projectiles though. Beyond that, some active targeting systems may be of use in the terminal phase of approach. This would only be true of slower moving projectiles/missiles, but switching to active targeting when on terminal approach would invalidate flares completely and may allow more accurate targeting of subsystems. Over all I'm most interested in seeing what diffuse sensor nets could do in conjunction with munitions with limited guidance. On a different note, I've continued testing of the rail gun vs the coil gun against heavier armor. In doing so I've found a way to get more consistent encounters (1.44Mm every time with TtCA of between 20h and 5d), so I'm redoing all tests including the weaker armor one. Against the heavy armor, the coil gun is getting significantly lower TTKs than the sandblaster, despite firing at about 4 times its in game accurate range. I'll post the final results as soon as I've completed the tests.
|
|
|
Post by vegemeister on Feb 17, 2017 21:11:04 GMT
I think any gun intended for use at very long range will necessarily fire guided projectiles. Even if velocity dispersion out of the barrel is zero, there's some uncertainty in the position of the target and in orbital mechanics predictions. But that doesn't necessarily mean gimbals and turbopumps and fuel tanks. A li-ion battery and electric solid propellants could do it. Of course the now guided projectile either receives command guidance or has its own sensor suite. Command guidance almost certainly, since on-board seekers would be heavy, single use, and could easily be blinded by (comparatively) low-intensity lasers. I imagine you'd launch a number of sensor probes just prior to opening fire. These would spread out under cold gas thrust, and would act as a synthetic aperture telescope so you could get enough resolution to do command guidance at all. They would transmit their data to your capital ship by low-observability spread-spectrum radio or whisker laser. A fleet of 3 or more ships could create a synthetic aperture without launching probes. You could track the projectiles by putting a corner reflector on the back end and painting them with a laser. Or maybe create a temporary GPS constellation with probes, but that'd be vulnerable to lasers at shorter ranges, and would be taken out by the first volley if the enemy chose to do something about it. You could have redundant probes, though, that would only start transmitting after some were destroyed. Of course, since armor's pretty much worthless, if they can get kinetic weapons to your probes, they can get them to your capital ship. Once you know the positions, velocities, and maybe accelerations of the the enemy cap ship and your projectiles, you just transmit them to the projectiles by encrypted whisker laser.
|
|
|
Post by bigbombr on Feb 17, 2017 21:58:38 GMT
Of course the now guided projectile either receives command guidance or has its own sensor suite. Command guidance almost certainly, since on-board seekers would be heavy, single use, and could easily be blinded by (comparatively) low-intensity lasers. I imagine you'd launch a number of sensor probes just prior to opening fire. These would spread out under cold gas thrust, and would act as a synthetic aperture telescope so you could get enough resolution to do command guidance at all. They would transmit their data to your capital ship by low-observability spread-spectrum radio or whisker laser. A fleet of 3 or more ships could create a synthetic aperture without launching probes. You could track the projectiles by putting a corner reflector on the back end and painting them with a laser. Or maybe create a temporary GPS constellation with probes, but that'd be vulnerable to lasers at shorter ranges, and would be taken out by the first volley if the enemy chose to do something about it. You could have redundant probes, though, that would only start transmitting after some were destroyed. Of course, since armor's pretty much worthless, if they can get kinetic weapons to your probes, they can get them to your capital ship. Once you know the positions, velocities, and maybe accelerations of the the enemy cap ship and your projectiles, you just transmit them to the projectiles by encrypted whisker laser. Missiles could also be equipped with a weak laser to 'feel up' the target to verify it isn't a flare. Command guidance of some extent would probably be standard, as well as information sharing within the swarm (this way different missiles with different sensors can cover each others weaknesses).
|
|
|
Post by newageofpower on Feb 17, 2017 22:56:30 GMT
How would a tiny laser pointer differentiate a flare from a heat source?
Thermal image recognition using feeds provided by recon missiles and recon drones, command guidance, and swarm attack logic (i.e. don't commit all attackers to one target at the same time if possible) should make flares and simple countermeasures obsolete.
|
|
|
Post by deltav on Feb 17, 2017 23:15:02 GMT
I think a lot of what we have ingame is vastly oversimplified to make the game easier to play and work with.
So in my mind, flares are like a blanket coverall for any kind of countermeasure that could possibly be ever invented. At least ingame that's how they work.
Of course we could have warheads with different kinds of guidance systems (radar, infrared, laser dot seeker, etc.) and countermeasures for all those classes, like chaff for radar, some kind of smoke for laser ranged guided, etc. But then the game would just get more and more complex. Perhaps just too complex.
|
|
|
Post by Easy on Feb 17, 2017 23:23:02 GMT
I think a lot of what we have ingame is vastly oversimplified to make the game easier to play and work with. So in my mind, flares are like a blanket coverall for any kind of countermeasure that could possibly be ever invented. At least ingame that's how they work. Of course we could have warheads with different kinds of guidance systems (radar, infrared, laser dot seeker, etc.) and countermeasures for all those classes, like chaff for radar, some kind of smoke for laser ranged guided, etc. But then the game would just get more and more complex. Perhaps just tood complex. the best flares I have seen are cancer refrigerators. It is a small unmanned space station consisting of a nuclear reactor with no shielding and a radiator running at the highest temperature possible. The reactor produces enough electricity to run the turbopumps, the other hundred gigawatts is waste heat.
|
|
|
Post by deltav on Feb 17, 2017 23:25:00 GMT
How would you use those in game?
|
|
|
Post by Enderminion on Feb 17, 2017 23:46:23 GMT
How would you use those in game? Like flares
|
|
|
Post by underwhelmed on Feb 18, 2017 1:31:06 GMT
I think the idea of simple flare decoys being effective is pretty unconvincing. Any simple discrimination logic would rule them out in most scenarios... Doesn't take a genius to figure out new sources of heat that suddenly appear traveling perpendicular to the target are decoys.
I think by now it's apparent there's no way to handle sensors from a physics ground up approach and do it justice. An essential part of game design is choosing what are the meaningful decisions a player makes... For example, one laser frequency doubler is clearly the best choice and no knowledgeable player is going to pick another in any normal circumstance, so why bother giving a bunch of false choices? My approach would be to simplify sensor (and jammer) design to just the critical trade-offs - weight, size, cost, power, etc and leave it at that. Make some assumptions about this sensor fusion and data link stuff (as the game already does with command guidance) or it'll never end.
|
|
|
Post by bigbombr on Feb 18, 2017 7:11:06 GMT
I think the idea of simple flare decoys being effective is pretty unconvincing. Any simple discrimination logic would rule them out in most scenarios... Doesn't take a genius to figure out new sources of heat that suddenly appear traveling perpendicular to the target are decoys. I think by now it's apparent there's no way to handle sensors from a physics ground up approach and do it justice. An essential part of game design is choosing what are the meaningful decisions a player makes... For example, one laser frequency doubler is clearly the best choice and no knowledgeable player is going to pick another in any normal circumstance, so why bother giving a bunch of false choices? My approach would be to simplify sensor (and jammer) design to just the critical trade-offs - weight, size, cost, power, etc and leave it at that. Make some assumptions about this sensor fusion and data link stuff (as the game already does with command guidance) or it'll never end. I strongly disagree. This game is about uncovering what realistic space combat would actually be like. For that, we need as many options and detail as possible, as conventional wisdom gets overturned every update or so. Simplifying something for the sake of simplifying would do this injustice. How would you even determine how effective a sensor would be without going into detail? Just look at how varied coilguns and railguns are. Could you simulate these in the way you propose? Determining what is optimal for a given situation is one of the core principals of the game. Detection and tracking (and even communication) are too important to oversimplify or abstract away. The importance of these becomes clear when observing current military hardware and the emphasis on awareness and communication, from AWACS to Global Hawks to tiny quadcopters. Just look at current day thanks and how loaded they are with sensors and comm gear!
|
|
|
Post by omnipotentvoid on Feb 18, 2017 7:12:33 GMT
I think the idea of simple flare decoys being effective is pretty unconvincing. Any simple discrimination logic would rule them out in most scenarios... Doesn't take a genius to figure out new sources of heat that suddenly appear traveling perpendicular to the target are decoys. I think by now it's apparent there's no way to handle sensors from a physics ground up approach and do it justice. An essential part of game design is choosing what are the meaningful decisions a player makes... For example, one laser frequency doubler is clearly the best choice and no knowledgeable player is going to pick another in any normal circumstance, so why bother giving a bunch of false choices? My approach would be to simplify sensor (and jammer) design to just the critical trade-offs - weight, size, cost, power, etc and leave it at that. Make some assumptions about this sensor fusion and data link stuff (as the game already does with command guidance) or it'll never end. Don't all systems in the game use assumptions and simplifications in order to simulate them in a way that doesn't fry our computers? As far as I can remember from researching rail guns, rail geometry is incredibly important to the calculation of the force applied to the projectile. The game doesn't simulate this, rather it assumes a rail geometry for which the equations are known/simple and uses them. I imagine implementation of sensor systems would be similar: assumptions are made so that the calculations that the game has to make based our input parameters are simple enough that our computers can handle it.
|
|
|
Post by apophys on Feb 18, 2017 8:01:27 GMT
Don't all systems in the game use assumptions and simplifications in order to simulate them in a way that doesn't fry our computers? As far as I can remember from researching rail guns, rail geometry is incredibly important to the calculation of the force applied to the projectile. The game doesn't simulate this, rather it assumes a rail geometry for which the equations are known/simple and uses them. I imagine implementation of sensor systems would be similar: assumptions are made so that the calculations that the game has to make based our input parameters are simple enough that our computers can handle it. The only in-combat difference that a different geometry simulation would cause is a change in the parameters applied to the barrel and projectile. This doesn't add new parameters, hence it has no effect on the processing load outside of the module editor. I'm quite sure the simplification for railguns is either due to unknowns or to reduce game development time, or both. It may be added sometime in the distant future, or when/if the dev team expands. The current railgun geometry as displayed is actually nonviable. It is one solid barrel of metal, which would short circuit at its base.
|
|