I've been doing some brainstorming on the problem of point defense, and I think I've worked out most of the issues that need to be discussed and worked out. I don't have solutions to everything, but I think I can categorise the issues well enough to start a thread off.
There are a number of questions that need to be answered. First off is "What role do we want point defense to play?". Is it a substitute for a shield (AP has such a unit in TBR already..should this be universal?) Is it a hard counter to certain weapons? Or does it simply work to reduce the number of hits you take? I am curious as to what others feel the role of PD should be (not necessarily what it is now).
Personally I see it as a system that can be extremely flexible...but not with the current implementation. With a few changes though, we can make PD fit almost any conceivable role.
Secondly, there are a number of known issues that need to have solutions worked out before point defense can be overhauled properly. These are, in order:
1. Antiquated target selection criteria on existing point defense units.
2. Lack of target selection criteria appropriate to differentiating PD units
3. Lack of missile hitpoint balance
4. Lack of diversity in point defense options
5. The role, diversity and setup of PD-equipped autons
I'll go through these in order.
Antiquated target selection criteria on existing point defense units.
Put simply, the target selection on existing PD units is out of date. The system was updated years and years ago to be able to take far more exhaustive criteria. However, the units ingame were never updated to use this system. This has to be done to deal with the "phantom missile" bug that currently cripples vanilla point defense. However, fixing this introduces issues 2 and 3:
Lack of target selection criteria appropriate to differentiating PD units
Even after updating to the more modern criteria, there are certain things a PD unit cannot do. For example, it cannot prioritise the incoming missile with the highest potential damage, or the lowest hitpoint count. It cannot ignore low-damage fragments or missiles that have such high HP that it cannot possibly deal with them. I do not know (as I'm no coder) how one might set up variables of this type, but they are important for dealing with the lack of diverse PD solutions.
More importantly, we need to think of these nice and early so they can get into the engine to allow for work further down this list.
Lack of missile hitpoint balance
This is the big one. We cannot balance PD units until missiles themselves have balanced hitpoint counts. Automatic HP totals should reflect value, mass, damage, stealth, and rate of fire. Some missiles should use ability to slip past PD as a balancing point. Others should be fragile. Either way, we currently have no balance in this area: it quite simply has not been done. It is an extremely complicated matter ideally suited for the insanely complicated and opaque algorithms that George so loves using for balancing a lot of things at once.
Once missile hitpoints have been balanced across levels, it becomes possible to make PD work because it actually matters what PD unit you use....once the last problem has been looked at.
Lack of diversity in point defense options
Once the system itself has been made workable, there is one massive issue remaining to make it work ingame: There are two point defense units in the game. They are almost entirely interchangable, except for the use of the Longreach as fitted to the Aegis auton as a swarm-busting superweapon thanks to the "phantom missile" bug. There needs to be multiple chains of PD units, with each line specialised in something specific. The units can then be specc'd appropriately. We have different types of weapons we can use, ammunition, intercept range, multi-targeting, autoacquiretarget functionality......we can go wild with PD, as long as we plan it properly and find a niche so as to spread them across the game (while obviously not saturating every store with dozens of different PD units).
As part of this, we need a way to balance point-defense effectively as a function of what it's attacking, how long it has to do damage, and how much damage it actually does. I do not for a minute pretend to have the maths needed to do this (and we would first need missile HP to be balanced before working on these numbers). However it would be negligent not to at least acknowledge that this is going to be a challenging part of the process.
Example of possible PD lines:
ICX family: Short-to-medium range and detection range, rapid fire red-laser weapons using the relativistic beam effect. No fancy targeting logic: will shoot the nearest projectile until it dies or hits, then moves to the next one. General purpose PD unit. Upgraded models have increased rate-of-fire and/or detection range. Could go up to level 5-6. Speciality versions (Eg. Tinker recipes) could use multiple cannons in close-packed custom config with multitargeting.
Longreach family: Long detection range, medium ROF but reduced-velocity kinetic rounds. Prioritises the incoming projectile with the highest HP and damage (IE: Missiles). Does not engage smaller projectiles at all. Smaller line than ICX, with one or two models (straight damage upgrade).
Pest-control family: Ok, I'm not good at naming things. move along. Later-game line that comes in in military space. Uses low-damage relativistic particle beams to provide short range (10-15ls) protection against missiles and small craft. Prioritises missiles over a certain size (or with special effect, eg: pteravores), then enemy small craft (focusing on the ship with the lowest HP), then other missiles, then other enemy craft.
Auton-PD family: Point defense units designed with a long range, prioritising missiles aimed at the "parent" ship over the unit itself. Fitted to autons like the 310A Aegis.
Armor-family PD: Armor segments that employ a short-range PD with limited arc (currently impossible) when something's about to hit that specific segment. Like a sort of pre-emptive reactive armor. Might use power, deplete charges or reduce armor HP by a little (rather than a lot if the missile hit it).
These are purely examples of what might be possible to increase PD utility and diversity, not concrete suggestions (this is a discussion thread, after all). There is, of course, also the matter of mining devices and specialised anti-ship point-defense systems. I haven't really considered them, but they are there as options and should be given attention. They also illustrate the sheer number of types and purposes we will need to have balanced.
The role, diversity and setup of PD-equipped autons
Finally as a lesser but still important issue, I'd like to note that the 310A Aegis auton is currently in need of a complete overhaul. The Longreach I is ludicrously valuable for the early-game, does not improve its anti-missile performance over an ICX, and with the current targeting behaviour and bugs results in what is effectively a glorified fragmentation grenade following the player around: a 310A sent into a swarm of kinetic-armed ships or stations will destroy almost everything around it with stray gunfire.
Beyond that, overhauling PD allows for the creation of more nuanced and specialised autons. Using the above criteria (which are, again, examples) one could have two lines of autons with PD fitted quite easily: one with general purpose PD like the current 310A, and another using the "auton-family" or support-style point-defense to protect other vessels. These could, to prevent exploits, even be set up so they cannot be removed with the auton bay (hardwired in or structurally integral). They might not even have regular weapons. Imagine an upgraded mule varient with an automatic cannon to keep things away from it...but not for "combat" use as such? Many things are possible here as part of a properly updated and integrated point-defense implementation.
(This also incidentally assists with the fleshing-out of the auton system, which is in dire need of more models to fill in gaps in the current implementation.)
-----------------------------------------------------------------------
These issues and possibilities are what I've currently given thought to. I'm certain there are other things that need to be considered. Point defense is too complicated for any one part to be updated to "fix" it. We need to work out everything that needs changing, come up with solutions that are workable enough to at least spark something in George's head...and basically come up with the whole picture before anyone starts putting it into the engine or XML. Any issues, suggestions or existing solutions to issues that haven't been implemented (eg: the target criteria issue mentioned in this post) are welcome. I will, however, be moderating this thread rather strictly so do keep on-topic.
