Author Topic: I think I create too slow missiles  (Read 6688 times)

0 Members and 1 Guest are viewing this topic.

Offline TallTroll

  • Lieutenant
  • *******
  • T
  • Posts: 154
  • Thanked: 19 times
Re: I think I create too slow missiles
« Reply #15 on: October 27, 2016, 06:10:57 AM »
>> ( who made the claim about a certain agility being the best bang for MSP ).

If you have no info on potential targets, I would always prioritise speed. An ASM that that would have 100% to-hit, but can't catch its' target is worse than a faster ASM that *could* catch its' target, but only has an 80% to-hit. Of course, once you do have some info on your potential opponents, you can design missiles optimised to fight them, since you will have some idea of their size and speed, at least.

I would argue that until then, designing ASMs for max speed and 11 AG is likely to be the superior design constraint, because it gives the highest flexibility. A player may choose different design constraints for other reasons, such as RP value, having a specific playstyle in mind from the start, or just because that's not how they like to design missiles. Aurora is a very catholic game in that way. Max speed / 11 AG is probably a more optimal build when you just don't know what's out there though. But, this is Aurora... when has "optimal" ever been everyones yardstick  ;D?
 

Iranon

  • Guest
Re: I think I create too slow missiles
« Reply #16 on: October 27, 2016, 06:47:43 AM »
The fixation on "11 AG" makes no sense.

You shouldn't care how much you get "per point of agility spent".
If you'd like some apples and I give you a free sample, by your argument you'd never buy more than one - after all, the more you buy, the higher the price per apple.
This makes no sense, unless you can mooch freebies off multiple vendors, which you can't in the case of missile agility.
 

Offline alex_brunius

  • Vice Admiral
  • **********
  • Posts: 1240
  • Thanked: 153 times
Re: I think I create too slow missiles
« Reply #17 on: October 27, 2016, 06:54:33 AM »
I would argue that until then, designing ASMs for max speed and 11 AG is likely to be the superior design constraint, because it gives the highest flexibility.

I'd argue that you never design missiles without knowing anything at all about the missiles intended targets. Same way you never design a single ship to do everything.

If I make a massive Warhead 50 missile I know I'm not going to use it against fighters for example... And opposite I know that an anti FAC/Fighter missile with a small warhead probably won't need to get through massive amount of PD and can put more priority on Agility instead to maximize accuracy/to hit.


If you design a single generic all purpose missile to "kill all stuff" your probably not playing Aurora at an advanced enough level to care a whole lot about math, formulas or targets anyway.
 

Offline TallTroll

  • Lieutenant
  • *******
  • T
  • Posts: 154
  • Thanked: 19 times
Re: I think I create too slow missiles
« Reply #18 on: October 27, 2016, 08:04:33 AM »
I think you are missing the point here, although I may not have been explaining in enough depth, so here goes.

Consider the total possible design spaces for all possible ASMs, and all possible enemy vessels. There are certain design choices for you as the ASM designer that will produce better average outcomes in combat. In the specific case of having no info on enemy designs, one of those possible choices is prioritising speed, because that would allow you to better engage a larger proportion of all possible enemy designs, because the conditions for using ASMs successfully is that they can 1) catch the target, and 2) hit it. Clearly, the 2 conditions are fairly tightly bound, because both speed and AG contribute to final to-hit %age, but only speed contributes to condition 1). Having a high to-hit value is worthless if your ASM cannot in fact catch an intended target, thus one of the design considerations is how to allocate MSPs to maximise your ability to catch and hit a target of unknown performance and specs.

Therefore, one approach to ASM design is to maximise your speed. As we know, both speed and AG factor into the final to-hit calc though, so it seems likely that allocating some MSPs to AG will lead to better overall ASM performance. All missiles start with a base AG of 10. If we choose to allocate enough MSPs to increase that to 20, for instance, we use an amount of MSPs to double the AG stat and thus improve the to-hit value with a method independent of speed, but is there a better choice?

Enter the 11 AG break point. Instead of allocating MSPs sufficient to double the AG stat, we use enough to raise it to 11, getting a 10% increase for the commitment. Every subsequent commitment of the same amount of MSPs suffers from the law of diminishing returns, decreasing the benefit we gain overall. Going from 11 to 12 only nets us a gain of about 9%, rather than the 10% we got the first time, and so on until the increase from 19 to 20 is only gaining us ~5%, about half the benefit of that first commitment. Thus, setting AG to 11 is an "optimal" design choice, because it gains us the highest possible increase in overall to-hit, for the lowest investment of MSPs. Ideally, we would like to invest our MSPs in most efficient manner possible, because missiles are pretty expensive, and tie up valuable and irreplaceable resources.

Now, it may prove to be the case that there were better things we could have done with those MSPs, but to predict that we would need to have some idea of the stats of the vessels we would be shooting at, and one of the key assumptions of this design philosophy is that we don't. As I say, as soon as we have some info on enemy ships, we can start to design ASMs that counter them more efficiently, but since most games start with the player knowing little or nothing about enemy designs, it's worth considering.

Nothing in any of this places any absolute requirement to use this design philosophy, since as noted previously, you may be deliberately RPing a race that has other priorities, have other constraints like lack of certain resources or production capacity, or just decide that it isn't a design philosophy you want to use, but it remains the case that if you just don't know what is out there, using this approach is more likely to produce ASMs that are more useful against a larger range of potential targets, for the smallest investment of resources.

I'm not saying that 11 AG is a perfect solution for all situations by any means, but when you have no other info to go on, it provides a better general purpose solution than most other possible choices. Assuming that that is the kind of solution you are looking for, which it may not be... Missile design is sometimes an art as much as a science. That's why there are so many threads about it. The "perfect" solution is to design a specific hard-counter missile for every enemy ship class, but that involves other costs and problems that aren't reflected in the missile design screen, like magazine and ordnance loadout management, complex logistics, tying up finite resources in missiles that may never even see combat, and so on. Trading down to 11 AG for speed is just one approach which will probably be more beneficial for many players in many situations. Noone is going to be upset if you choose different priorities, for any reason or for none.

>> I'd argue that you never design missiles without knowing anything at all about the missiles intended targets.

Ideally, yes. If the first time you meet a given race is when their war fleet drops into Sol or a system one jump away or something, you might want some ordnance on hand though. So having some missiles would be better than having none. And having missiles that are more likely to be at least somewhat effective is generally better than having missiles specifically designed for the wrong target

>> If you design a single generic all purpose missile to "kill all stuff"

That's not what I'm saying at all. AMMs have different design constraints to ASMs for smaller targets, which have different constraints to ASMs for larger targets, which have different constraints to ASMs for long range combat, which  have different constraints to ASMs for short range combat etc, etc. Having a few fire units of general purpose ordnance on hand when you need them is better than being able to design a perfect hard counter ASM too late, and getting your planets nuked before you can build any "good" missiles
 

Offline alex_brunius

  • Vice Admiral
  • **********
  • Posts: 1240
  • Thanked: 153 times
Re: I think I create too slow missiles
« Reply #19 on: October 27, 2016, 08:46:22 AM »
Ideally, yes. If the first time you meet a given race is when their war fleet drops into Sol or a system one jump away or something, you might want some ordnance on hand though. So having some missiles would be better than having none. And having missiles that are more likely to be at least somewhat effective is generally better than having missiles specifically designed for the wrong target

How does having no missiles have anything to do with it?

The choice here is between having a few different types of missiles that perform well in certain roles against different types of targets, and having a generic one size fits all missile.

Since missile engines and missiles are such small components the more specialized ones cost you almost nothing extra to research or build, and you can allow you to cover both cases of missiles which are fast and good at defeating heavy PD, and missiles which are agile and good at defeating faster FAC/Fighters or targets without PD.

That's not what I'm saying at all. AMMs have different design constraints to ASMs for smaller targets, which have different constraints to ASMs for larger targets, which have different constraints to ASMs for long range combat, which  have different constraints to ASMs for short range combat etc, etc. Having a few fire units of general purpose ordnance on hand when you need them is better than being able to design a perfect hard counter ASM too late, and getting your planets nuked before you can build any "good" missiles

Where did I argue you should wait and design the ASMs to counter specific alien designs?

The first thing I wrote was that you can design missiles against generic target types without knowing anything about your enemy ( or even having met them ).

If you have to use them against the "wrong" target they will not be performing worse then a compromise "one size fits all" design will, which also is not tailored against the target.

They will however perform better when they are used against their intended target, so there is no downside to the approach.
« Last Edit: October 27, 2016, 08:49:35 AM by alex_brunius »
 

Offline 83athom

  • Big Ship Commander
  • Vice Admiral
  • **********
  • Posts: 1261
  • Thanked: 86 times
Re: I think I create too slow missiles
« Reply #20 on: October 27, 2016, 10:57:17 AM »
The choice here is between having a few different types of missiles that perform well in certain roles against different types of targets, and having a generic one size fits all missile.
There is also the option of having a few types of generic missiles to fit certain conditions (like range).
The first thing I wrote was that you can design missiles against generic target types without knowing anything about your enemy ( or even having met them ).
Newer players can't really do that as they don't really expect what kin of things they may go against before they do.
If you have to use them against the "wrong" target they will not be performing worse then a compromise "one size fits all" design will, which also is not tailored against the target.
This is highly circumstantial. Sometimes using a specialized missile against the "wrong" target has the same effectiveness as a generic. Other times it is worse off. It depends on the specialized missile, the generic missile, and the enemy that it is hitting.
They will however perform better when they are used against their intended target, so there is no downside to the approach.
A misrepresentation using the "because there is an upside, there is no downside" way of thinking.

1) catch the target, and 2) hit it. Clearly, the 2 conditions are fairly tightly bound, because both speed and AG contribute to final to-hit %age, but only speed contributes to condition 1).
Not necessarily. While that is true against enemies headed in the opposite direction, it does not apply to enemies headed directly at you or sitting still. I've had plenty of cases in my own games where I have been hit by surprise missiles that were slower than my ships, and where I've hit enemies who were faster than my missiles.
Having a high to-hit value is worthless if your ASM cannot in fact catch an intended target, thus one of the design considerations is how to allocate MSPs to maximise your ability to catch and hit a target of unknown performance and specs.
Just because you sacrifise some MSP % from engines and put it into agility, does not mean it will not catch the enemy. Your argument seems to be based on sacrificing most of your engines to put into agility.
Therefore, one approach to ASM design is to maximise your speed. As we know, both speed and AG factor into the final to-hit calc though, so it seems likely that allocating some MSPs to AG will lead to better overall ASM performance. All missiles start with a base AG of 10. If we choose to allocate enough MSPs to increase that to 20, for instance, we use an amount of MSPs to double the AG stat and thus improve the to-hit value with a method independent of speed, but is there a better choice?
Another factor is retargeting when the designated target is taken out, so you want onbord sensors which sacrifice either warhead or engines/fuel. Another factor is making sure the enemy can't hit your missiles, so that is space into missile ECM. If it does get hit, you don't want it to die so you need to armor it. You see where this is going? It isn't about 1 is the best, its about tradeoffs. While having a generic missile is a very good idea, it lacks the specialized performance these other types of missiles get.
Every subsequent commitment of the same amount of MSPs suffers from the law of diminishing returns, decreasing the benefit we gain overall. Going from 11 to 12 only nets us a gain of about 9%, rather than the 10% we got the first time, and so on until the increase from 19 to 20 is only gaining us ~5%, about half the benefit of that first commitment. Thus, setting AG to 11 is an "optimal" design choice, because it gains us the highest possible increase in overall to-hit, for the lowest investment of MSPs..
Engines also suffer from diminishing returns while you don't see it unless you go crazy at mid level techs, as you start advancing into really powerful engines, you notice it quite a lot. It gets to the point where even the "optimal" percentage of engine to missile starts becoming sub-optimal compared to other designs.
Ideally, yes. If the first time you meet a given race is when their war fleet drops into Sol or a system one jump away or something, you might want some ordnance on hand though. So having some missiles would be better than having none. And having missiles that are more likely to be at least somewhat effective is generally better than having missiles specifically designed for the wrong target
However, you can also design differing generic types of ASMs, like Long range that sacrifices speed/warhead for a bit more fuel, or short range that sacrifices its fuel for more speed/warhead. Etc. There is more than one type of "generic missile". My generic ASMs always have some onbord sensors so they can retarget while inflight as to not waste resources.
Give a man a fire and he's warm for a day, but set fire to him and he's warm for the rest of his life.
 

Iranon

  • Guest
Re: I think I create too slow missiles
« Reply #21 on: October 27, 2016, 11:09:07 AM »
@ Tall Troll: You're misapplying the concept of diminishing return. Your optimum can be at a point far beyond the spot where diminishing returns kick in.

If Manoever Rating was tracked to 2 significant figures rather than integers only, you'd presumably argue for enough agility for a MR of 10.01, satisfying your "maximum gain per agility point spent". That reasoning would be more conistently applied by assigning no agility at all.
MR 10 or 11 is terrible for a general purpose missile, whether we know specifics or not. The only thing it's good for is as a dedicated decoy to run the opponent out of AMMs - hard to shoot down, we don't care that accuracy is terrible because they aren't expected to reach them (when the opponent has run dry, we switch to missiles that can actually hit).

*

If we are afraid that we may fail to penetrate the point defence screen at all or that our missiles may not catch a fleeing opponentl, speed > accuracy.
If we expect that few missiles will be engaged, let alone shot down (thousands of simultaneous missiles, or hundreds split into one-missile salvos), accuracy > speed.

With enough details, we could work out the optimal split between engine and agility.
If we fire one salvo of 100 missiles at a target that has 60 CIWS shots tracking at 20000km/s, expected damage is proportional to 100*accuracy - 60*0,5*20000/speed.
Assuming our missiles have a speed of at least 20000km/s, that is.

In most realistic scenarios where the missile strike isn't borderline pointless, a reasonable amount of agility will perform better.
 

Offline GetStqned (OP)

  • Petty Officer
  • **
  • Posts: 25
Re: I think I create too slow missiles
« Reply #22 on: October 30, 2016, 04:45:06 AM »
If you want to be effective , why not specialize missile ? One for heavy enemy ship , one for lighter , for fighter , for fast or low speed , it will be more costly in resources and money but you will theoretically invinsible not?
"An army of lions led by a deer and less formidable than an army of deer led by a lion"
 

Offline Garfunkel

  • Registered
  • Admiral of the Fleet
  • ***********
  • Posts: 2781
  • Thanked: 1048 times
Re: I think I create too slow missiles
« Reply #23 on: October 30, 2016, 07:38:11 PM »
Remember that you cannot carry endless numbers of missiles with you. The more types you have, the bigger the risk that you will run out of effective missiles at a critical moment.
 

Iranon

  • Guest
Re: I think I create too slow missiles
« Reply #24 on: October 30, 2016, 08:07:16 PM »
If you field specialised missiles with varying capability, it may become problematic to concentrate missile fire to overwhelm enemy defences.
If you field missiles of varying sizes, logistics may become an issue, especially if you run dry on a smaller type and your ships can't fire any ordnance that is on hand.

Sometimes I create variants of identical size and speed though, say a long-ranged variant for ships and a short-ranged variant for fighters that devotes more space to warhead or agility.