Missile Prototype 5

PlayerOLish

I still find mines underwhelming, they aren't the anti light ship destroyers I thought they would be

ucyimDa_Ruler

Why not leave the PD in its current state of power in this experimental version but provide us an option with the "Auto-fire" that reads "Manned". It takes up a crew slot but that crew member provides better intelligence on firing and energy usage. Otherwise it stays automated to provide some basic level defense. This way it is up to the builder to set the state during the ship build. It would solve both problems while introducing balance and cost.

CrazyBirdie

In playing with this in creative, I came across a couple bugs. The first one was that the missiles (I used mainly EMP and HE), especially the HE ones, would occasionally bypass shields. EMPs take out the shield anyways, but the HE missiles would cause damage to the ship.
The other bug I had was with a kiter prototype ship. I would shell an enemy ship with EMPs while keeping a constant distance. Suddenly, the other ship would turn off its thrusters and the two would come very close and my ship would get annihilated. I had multiple ships do this to me (Terror, The Beast, one of my own ships). I am not sure, but it seems as though this happens when I am firing only EMPs.

Walt

CrazyBirdie Saved games that show those problems would be really helpful if you have them.

CrazyBirdie

I have a saved game (my prototype versus Terror); how do I upload it?

Walt

CrazyBirdie Just upload it to any file sharing site and post a link.

CrazyBirdie

Walt I don't know if you saw my link before, so I am posting it again for you.
https://www.mediafire.com/file/cq528ya3zg0unmf/MP5Errors.savedgame/file
The bug occurs shortly after the game loads. If you have my prototype attack a different ship (I would try Cerebus) from a safe distance (out of range of the enemy's weapons), you should see some missiles bypassing shields.

Atarlost

If the fundamental problems are battery powered compartments and the ability to fire too many missiles from side mounted launchers, why not address these directly?

For compartments, implement a cockpit pathing check. Intercoms could be assumed to use wires in walkable tiles and a crewman without a walkable path to a command center can't get orders. Comparments must then include individual command centers making them uneconomical.

For missile spam on large ships implement a telemetry link limit. Every missile fired guided uses up a telemetry link until it stops existing and no missiles can be fired in guided mode if there are no open telemetry links. Telemetry links come from command centers and sensor arrays, not from missile launchers so that missile launcher spam requires large numbers of expensive rooms, but telemetry adequate for one or two launchers for small ships comes free with the first command center that you need anyways to have a ship. Telemetry links could justifiably also limit PD emplacements since they are unmanned and also need to track missiles.

samepage

Atarlost For missile spam on large ships implement a telemetry link limit.

At first I kinda dislike the idea of having a specific part to allow missiles to function. But after such long struggle on the missiles balance, this idea now seems much sensible and probably easily to implant as well.

However, I prefer the telemetry links connecting directly to the launcher instead of each missile. The explanation is that firing a missile is a big deal (especially the nuclear one) and should not be done without consulting to senior officers (those in CR and SA). I am thinking that 2 launchers for 1 CR and 4 launchers for 1 SA is about the right numbers. If one loss its CR or SA, either they get destoried or depowered, the launcher they connect cease to function until the problem is fixed.

Walt

Atarlost For compartments, implement a cockpit pathing check. Intercoms could be assumed to use wires in walkable tiles and a crewman without a walkable path to a command center can't get orders. Comparments must then include individual command centers making them uneconomical.

Well why do people want to use compartments in the first place? Because crew A.I. is bad, right? So I think the ideal way to solve the compartmentalization problem is to make the crew A.I. better, rather than adding an arbitrary mechanic like requiring walkable paths to control rooms.

0.14.5 should improve things, though I'm not sure how much, I haven't gotten much feedback on that. I suspect the final solution will involve giving the player more control over what crew do (i.e., giving them priorities and assigning them to specific parts).

Atarlost For missile spam on large ships implement a telemetry link limit. Every missile fired guided uses up a telemetry link until it stops existing and no missiles can be fired in guided mode if there are no open telemetry links. Telemetry links come from command centers and sensor arrays, not from missile launchers so that missile launcher spam requires large numbers of expensive rooms, but telemetry adequate for one or two launchers for small ships comes free with the first command center that you need anyways to have a ship. Telemetry links could justifiably also limit PD emplacements since they are unmanned and also need to track missiles.

I've thought about doing something like this, i.e. requiring sensors to give missiles guidance or to fire without line of sight. But I haven't really liked these solutions because it makes things more complicated and arbitrary.

I think a better solution is to simply add more weapons that are as space-efficient as missiles. Railguns were a good start, ion beam prisms should help, and maybe roof-mounted weapons would help as well.

(As an aside, something I miss about the prototype 4 PD were that they were significantly better against side-fired missiles than front-fired missiles.)

CrazyBirdie

Here you are: https://www.mediafire.com/file/cq528ya3zg0unmf/MP5Errors.savedgame/file
The sudden (almost) collision happens shortly after where it loads. I don't remember exactly, but it may have missiles bypassing shields also. Otherwise, just have my prototype ship attack another large ship from a safe distance and you should see the missiles bypass shields.

nop

Walt Well why do people want to use compartments in the first place? Because crew A.I. is bad, right?

This is one reason. Another reason is the cost savings of modules using energy storage instead of reactors, which is only possible in MP when the rules are set up to allow it, or in SP by gaming the build interface.

Walt rather than adding an arbitrary mechanic like requiring walkable paths to control rooms.

Maybe this measure won't be necessary, but a mechanic that solves a specific problem is not arbitrary. It's no more arbitrary than requiring a control room on a ship in the first place.

Dalas120

nop Another reason is the cost savings of modules using energy storage instead of reactors

Howso? If I can split 1 reactor across 8 different factories, that's far cheaper than using 16 or 24 storages in separate compartments. AFAIK the only reason missiles need to be compartmentalized is because of the crew AI and ability to save $$$ on crew, so I think if that gets changed there will no longer be any advantage to using battery powered missiles.

Also, it's a bit weird that power is the easiest/best to store resource in cosmoteer, while power is the hardest/worst to store in real life. Which is why ships in real life have reactors and fuel to generate power, but store their munitions instead of bringing factories with them. I don't think anybody would have a problem with ships that bring magazines full of ammo and run their cannons off of those, because that's the sort of thing real life ships do all the time.

Walt

nop This is one reason. Another reason is the cost savings of modules using energy storage instead of reactors, which is only possible in MP when the rules are set up to allow it, or in SP by gaming the build interface.

Yeah but people would still compartmentalize even if power storage didn't exist, fixing the power storage problem wouldn't eliminate compartmentalization.

nop Maybe this measure won't be necessary, but a mechanic that solves a specific problem is not arbitrary. It's no more arbitrary than requiring a control room on a ship in the first place.

It's more arbitrary from a lore/fiction perspective. It makes sense that a ship can't fly without a control room. It doesn't make sense that a weapon can't shoot without a path to the control room.

Dalas120 I don't think anybody would have a problem with ships that bring magazines full of ammo and run their cannons off of those, because that's the sort of thing real life ships do all the time.

And in fact this may be how multiplayer Cosmoteer ends up being played by most people once I revamp the storage mechanics.

CrazyBirdie

I use modules to create individual systems that are easy to copy and protected from damage in another module.

nop

Dalas120 If I can split 1 reactor across 8 different factories, that's far cheaper than using 16 or 24 storages in separate compartments. AFAIK the only reason missiles need to be compartmentalized is because of the crew AI and ability to save $$$ on crew, so I think if that gets changed there will no longer be any advantage to using battery powered missiles.

Increasing factory energy consumption to "nerf" single-storage modules also increases the drain on reactors and the crew requirement to haul energy to the factories. Using a reactor necessarily increases energy hauling distance. I don't think anybody really knows if the two methods are equivalent in this MT yet, if they do they haven't shared the results of their tests.

Dalas120

Walt So regarding missile balance:

Right now, all missiles have the same 1.5 second loading timer. This means that the limiting factor on all missiles is still only crew, since every launcher can fire faster than crew can bring it ammo (even when there is plenty available in nearby factories). I strongly suggest making load timers long enough to be a limit on fire rate, and having different loading timers for different missile types.

Also, it looks like the missile timer is triggered after each missile is fired. I would suggest having the timer before each missile is fired (as preparation time), so that you could (for example) shoot off a volley of slow-firing nukes, then swap to HE missiles without having to wait for a nuke reload timer.

Atarlost

Walt It's more arbitrary from a lore/fiction perspective. It makes sense that a ship can't fly without a control room. It doesn't make sense that a weapon can't shoot without a path to the control room.

Why not? The Honorverse uses this concept to limit missile volleys in the MDM era. Battletech uses very similar mechanics to limit teleoperated capital missiles.

Walt

Dalas120 Right now, all missiles have the same 1.5 second loading timer. This means that the limiting factor on all missiles is still only crew, since every launcher can fire faster than crew can bring it ammo (even when there is plenty available in nearby factories). I strongly suggest making load timers long enough to be a limit on fire rate, and having different loading timers for different missile types.

Also, it looks like the missile timer is triggered after each missile is fired. I would suggest having the timer before each missile is fired (as preparation time), so that you could (for example) shoot off a volley of slow-firing nukes, then swap to HE missiles without having to wait for a nuke reload timer.

It's a shared timer between all missile types. So if you fire an HE missile, wait 1 second, and then switch to nuke, the nuke will be ready to fire in 0.5 seconds (assuming it can get loaded by then).

Something I could do (maybe this is what you're suggesting?) is give each missile type its own separate reload timer that continues counting down even when the launcher is not set to that missile type. The only issue I can think of is if the reload timer is longer than the time it actually takes to load the missiles, theoretically you could fire more missiles by frequently switching types.

Dalas120

Walt Something I could do (maybe this is what you're suggesting?) is give each missile type its own separate reload timer that continues counting down even when the launcher is not set to that missile type.

I was thinking having a single reload timer for the missile launcher, but different missile types would have different requirements on that timer.

So maybe I fire a nuke, then the timer starts counting how many seconds have elapsed since the nuke was fired. If I want to fire another nuke I have to wait until that timer reaches 10 seconds, but if I want to fire an EMP I only have to wait until that timer reaches 5 seconds (for example). Or if I want to fire an HE maybe I only have to wait until it reaches 2 seconds.

That way the timer would still be shared across all missile types (so no wasted time if you switch), and there'd be no way to fire faster by rapidly swapping types.

Walt

Dalas120 I was thinking having a single reload timer for the missile launcher, but different missile types would have different requirements on that timer.

Ah yeah that makes a lot more sense. I think I can do that easily enough.

Dalas120 The opposite is something like 0.14.4 PD, which will only ever demand 1 crew even though they often need 2 or 3 to stay powered. But this at least can be mitigated by moving a reactor near the PD, or by using storage to recruit more crew.

Maybe I should just double the PD power storage. Because the crew demand algorithm isn't predictive, it'll never cause PD (or any part) to demand more power than it actually has capacity for.

Dalas120 I'm not sure what the best solution to this is. Of course it would be great to let the designer specify how many crew each part would demand, but I doubt that's practical to implement this update (and has potential for abuse, like assigning 50 crew to a single shield). A much simpler solution might be to just have the HE launcher demand fewer crew, since it's the biggest culprit.

Well in the long-run I'm planning to let the player assign crew to specific parts and configure their own priorities, but yeah, that's definitely not happening this update.

I'm not too concerned about abuse like assigning 50 crew to shields. I suspect that'd just be incredibly inefficient strategy (super expensive, tons of wasted power), and worst-case scenario I can just limit how many crew can get assigned to a part.

I don't want to artificially limit the number of crew that a missile launcher will demand, because that'd be optimizing for the situation where the factory is very close to the missile launcher at the expense of greatly nerfing missile launchers with farther factories. (Which is something that both less experienced and more aesthetically-oriented players do a lot.)

(As an aside, I wonder how much it would change things if the cost of crew were substantially lowered.)

Atarlost Why not? The Honorverse uses this concept to limit missile volleys in the MDM era. Battletech uses very similar mechanics to limit teleoperated capital missiles.

I'm not very familiar with either universe. How do they justify those mechanics within their lore?

CrazyBirdie I don't know if you saw my link before, so I am posting it again for you.

I did, thanks, just haven't got a chance to look at it yet.

Dalas120

A broad issue related to crew AI:

Parts commonly demand too much (or too little) crew for the situation. HE missiles are the most severe example of this - it only takes ~5 crew to load a HE launcher at the maximum rate, but a launcher will keep 12 crew busy loading almost all the time. This is part of the reason that missile battery compartments are powerful, because you can save a lot of $$$ by only putting in 8 crew (5 for reloads + 2 for manning + 1 for power) instead of the 14-15 demanded.

Another example is supplying 2 small cannons from an adjacent ammo factory. It only takes 2 crew to keep both cannons supplied, which works fine as long as there's plenty of ammo. But when the ammo factory eventually runs out of ammo, the small cannons demand extra crew even though it doesn't benefit them at all. In fact, the increased cannon crew demand sometimes means that crew stop bringing power to the ammo factory, slowing down ammo production.

The opposite is something like 0.14.4 PD, which will only ever demand 1 crew even though they often need 2 or 3 to stay powered. But this at least can be mitigated by moving a reactor near the PD, or by using storage to recruit more crew.

I'm not sure what the best solution to this is. Of course it would be great to let the designer specify how many crew each part would demand, but I doubt that's practical to implement this update (and has potential for abuse, like assigning 50 crew to a single shield). A much simpler solution might be to just have the HE launcher demand fewer crew, since it's the biggest culprit.