Phases ie ranges aren't symmetrical, if a ship is killed during long range phase it won't be in next phases. If a ship is killed during mid range phase it won't be in close range phase. If a ship is heavily damages during long range phase it will die sooner during mid range phase. Now if you don't see any influence of this in your rules, ie they include this element, then ok.

That's covered in points 7-9.

For the current system it's quite possible that a simplification would be better, but before I think that the current problem is many in game missing information and too many undocumented features and they should be documented in game.

Documenting many of these features in game would be difficult because the effects of many of the troublesome mechanics are only displayed in a combination of weapons, defenses, hulls, and combat. A general documentation of the mechanics would be useful, but implementing them in game descriptions is premature at this point.

Thuvian i was curious how you calculated the damages for so many attacks. Do you use a simulator? I found the mathematical way to be pretty infeasible even after just 4 attacks but if you have better mathfu i would love to know

thuvian wrote:


A general documentation of the mechanics would be useful, but implementing them in game descriptions is premature at this point.

No the point is to have the game compute for you some resulting numbers so at least this should by shown in game and once it is shown there's a reason and a tooltips should explain it and the sources and the effects.



If it's just series of formula it will generate abstract Math debate, well. :-)

Stalker0 wrote:
Thuvian i was curious how you calculated the damages for so many attacks. Do you use a simulator? I found the mathematical way to be pretty infeasible even after just 4 attacks but if you have better mathfu i would love to know

Sorry I didn't see that earlier. I do a couple of things

1. Run simulations to see what would happen after sufficient trials.

2. Just run the numbers for best/worst case scenarios.

3. Calculate the odds analytically and derive from there.

bilun wrote:
but it seems to me your first priority is seeking simplicity for it's own sake even at the expense of strategic depth.

Finally, someone came at me with the counterargument i was expecting Bilun, i salute you.





My point was not to remove the complex mechanics, just because. My point was that the mechanics are both complex, AND that they do not provide that strategic depth you were referring to.



Accuracy at the top level looks like it provides a big difference in how weapons operate. What i was trying to point out in my example is it actually doesn't. Because of the number of shots fired in a typical battle (even early game battles), accuracy doesn't provide a lot of variance. So all accuracy is really doing is slapping on a percentage to the damage the weapon will do. And if that's the case, lets just drop the middle man and remove the extra modifier all together.

Simplify Salvos per phase:

Right now, the number of salvos per phase a weapon system can fire does impact how easily it can switch targets in the heat of battle, so i think there is some worth to keeping it in (although i would be perfectly fine if we wanted to just roll this into the damage number as well).



That said, the salvo delay and the reload time are just complexities that don't serve great purpose. The more important number is how many salvos per phase we get, as that more directly tells us how much damage we do per phase, and how often we could theoretically switch targets in a battle.



Recommendation: Combine these attributes into a new value called "Salvos per Phase". Example would be 1 for long range missiles.

Remove accuracy and evasion from game mechanics:



Now that I have finished with a few softballs, now we get to the meat of my argument. Accuracy and evasion mechanics right now serve these purposes:



1) Provide a way for a ship to nullify all damage

2) Provide variation in combat

3) Provide some differences between small and larger ships.

4) Provide differences at ranges.



For the first two, let's consider a scenario. A ship with 50% evasion vs 4 long range tier 1 missile modules. I chose this scenario because it should represent high variance. Since missiles only get 1 salvos per phase, this is only 4 salvos, so the chance for variance is high. Generally as we increase the number of salvos per phase, variance decreases, so this would represent a "high variance" corner case.

First of all, let's go to the formula every loves:



%chance for a salvo (not shot) to hit = Accuracy - [Evasion-(EvasionDepletion*Numberofmissesintheround)]. Currently, Evasion Depletion is a hard value of .15.





A good example to highlight our scenario: A long range missile has accuracy 1. So against that ship, the first missile has a chance to hit = 1 - [.5-(.15*0)] = .5 = 50%. If the first missile misses, the second shot has chance = 1 - [.5-(.15*1)] = .65 = 65%. If the first missile had hit, the second missile would have a 50% to hit.

The math is very complicated, because this is conditional probability (the timing of the hits affects probability). I have the scenario hard coded in excel if people need to see it, but i will spare the math for now. Please be aware I am rounding slightly.





Result: (50% evasion vs 4 long range tier 1 missiles)

0 hits (0 dmg): .18% - (note that is .18, not 18)

1 hit (150 dmg): 10%

2 hit (300 dmg): 44.8%

3 hit (450 dmg): 38.7%

4 hit (600 dmg): 6.3%



So two key takeaways here:

1) There isn't that much variance. With 5 options, 84% of damage falls squarely around hits 2 and 3.

2) Evasion isn't really "nullifying damage". The chance of a ship escaping damage purely do to evasion is extremely small.



Just another quick data point to further highlight the point. Change the weapon to 4 melee tier 1 kinetic weapons. So now we only have a 50% accuracy, but we get 4 salvos per phase. In this case, the chance of escaping damage entirely is only 1.1%.



And keep in mind, that as the number of weapon modules increases (and it does quickly, even just a fleet of 5 ships with 5 LRM modules each is throwing 25 missiles a phase), that variance tends to drop rapidly, and evasion becomes negligible due to evasion depletion.



So to reiterate my original list:

1) Provide a way for a ship to nullify all damage: The evasion mechanic is not doing this currently.

2) Provide variation in combat: The variance is not that high, even at small amount of weapons fire, and quickly becomes much less varied with higher weapon counts.

3) Provide some differences between small and larger ships: Can easily be done through other adjustment.

4) Provide differences at ranges: We can return to damage adjustments for these, accuracy is not required.

So after that long discussion, here is my recommendation:



Recommendation: Remove accuracy and evasion mechanics and roll there modifiers into the damage roll. Adjust the critical percentage if additional variance is required. Change range modifiers to damage adjustments.

Weapons: The Final Product



So with these changes in mind, what would weapons look like? Here are some examples (the range adjustments might be off, just doing it for highlighting purposes):

LR Missile: 150 dmg/salvo, x2 crit, 1 salvo/phase (100% dmg LR, 50% dmg MR, 25% dmg Melee R)

Melee R Missile: 36 dmg/salvo, x2 crit, 4 salvos/phase (25% dmg LR, 50% dmg MR, 100% dmg Melee R)

It's important to note that even with several variables removed, we still have many ways to customize weapons. Range modifiers, tonnage, cost, some can have special materials, crits can be modified etc.





Defenses

Here is my take on defenses. No matter what fancy extras you throw on, the bottom line is they exist to absorb damage from a specific weapon. If I don't need defenses, i will add weapons or other modules. If I do, I'm going to adapt my defenses to my opponents weapons. I'm not going to use shields if my opponent is all missiles, even if the shields have a nice little side benefit.

So...let's just get defenses back to what they do. Give them a defense value and be done with it. Weapons can be the source of tailoring and special effects, while defense just exists to make your ship live longer.



Recommendation: Return defenses to simple defense numbers, removing adjustments to other attributes.

I agree with most of your points and its nice to see other people interested in this sort of systematic and mathematical approach. It parallels the work I've done in another thread here too.



My 2 cents.

I still like accuracy, but with a stable evasion score. None of this weighted evasion.

The firing rates need to change, currently short range weapons are the counter to long range weapons. This is so very, very wrong.

Alternatively, we can do away with the weapon vs defense pairings, and just make the defenses have different characteristics that apply to all weapon damage. Something more like the systems you see in SotS, MOO2, SoaSE, etc.

I can see where you're coming from, but let me try to approach this from what I think is the dev's perspective.



Flashback to classic ES, the game had functioning but relatively simple combat system. Lots people complain that its simplicity doesn't match up with the rest of the complex dynamics in the rest of the game. Namely the general default to use is always a fleet of beam destroyers.



With Disharmony, devs once to put in more complex decision making for combat, but a complex system is hard to balance and they either didn't have the time or couldn't afford an extensive beta test. I think many of the VIP will agree Disharmony was released well ahead of expected schedule. From a programmer's perspective, you can't just go changing the number of variables in the combat system from iteration to iteration (remember that the devs go through way more iterations of the game then we do). Lots of changes like that is a recipe for game-breaking bugs. So the programmer did the smart thing by simply implementing all the variables they could think of which the game designer (and maybe modder) will ever want to tinker with. Hence is born this complex system, where by most of the changes can take place at the numbers level instead of the formula level.



But then comes the second problem, because Disharmony was released early, there was not nearly enough time to tweak all those numbers. So instead of removing them from the game code (which would change the game code at the formula level), they disabled the ones that aren't working or otherwise making the game too hard to balance. With the intention of turning them back on later once the other major variables are more balanced. You can see a prime example of this in how they used damage reduction instead of accuracy reduction at different distances back in v1.1.9, where damage was a simple multiplication of several variables.



So you see, they were faced with two choices, evolving the combat system by constantly changing the game code and the numbers, or evolving by only changing the numbers and temporarily disabling some variables. If they had more time to work on Disharmony, the later is definitely the right way to go, and perhaps they did think they would spend more time on Disharmony prior to release. But the fact is Disharmony was released early, and everyone got confused about all the new variables in the combat system.



It's not an enviable situation to be in. But if they still intend ultimately to have the more complex combat system, then removing variables now will only add a lot more debugging later.

It's not that we are campaigning against a more complex combat system. It is that the current implementation doesn't actually accomplish what it sets out to do.



Accuracy & Evasion - End result: Between 0 and (1 - Accuracy + Evasion) *.15 salvos are blocked per round. The average depends on how many weapons are shooting, but as you get more weapons it will shift more towards the maximum number of blocked shots.

DeflectionPerTurn - X shots are blocked per round

Interception Accuracy - X missiles are blocked per round

Absorption - X damage is blocked per round

Hull Weakness - Makes defenses more or less valuable. Currently small ships are punished for wanting defense, and large ships are rewarded.

Hull Sizes - Since defenses are fixed in size and hit points increase with ships size, large ships benefit more from defenses (more EHP per % of hull spent on defense)

Weapons - Weapons do not get more potent, except with power modules, which are a questionable value to begin with.

Crit Chance, Crit Multiplier, Min & Max - All of these are hard to manipulate, but if you are able to affect them, you can cause dramatic damage changes. So if you get lucky you do extra damage, that your opponent can't even plan for.



You can stack 14 weapons in a baseline destroyer, you can stack 171 weapons in a late game Harmony dreadnaught. If those are short ranged weapons, they are shooting 168 times and 2052 times per combat! That means all of these miss the first X shot combat mechanics are going to have a tiny effect. By switching to (even if temporarily) to a simpler combat version, you can then establish an interesting and intriguing system, and then you can make it more complex afterwards.



As a metaphor, image the devs standing in front of a table with a huge pile of parts. They need to build something interesting, but are told they have to use all the parts, regardless if they want to. That's hard, that's very hard. It's easier to build a simple structure and then add decorations and improvements.

Yea, I think it great to simplify the current mess too. I was just pointing out how we end it up in the mess in the first place. I mention this in another thread but I feel the devs have been over-reacting to player's suggestion. This large number of attributes, to me, seemed like one such over-reaction.

Considering the total storm of complaints they received about just about anything, can you blame them?

Igncom1 wrote:
Considering the total storm of complaints they received about just about anything, can you blame them?

No. I'm for transparency, not simplification per se.

Igncom1 wrote:
Considering the total storm of complaints they received about just about anything, can you blame them?

At this point, i care very little about blame just correction.



I don't mind a complex system off hand, though i will admit i prefer emergent complexity from simple starting systems. But my main premise is that once we peer into the complexity that this system contains, we find that the variables don't actually have that much impact. We have lots of decisions to make in theory, but those decisions don't amount to any real tangible results.

Combining DMG per shot/shots fired is all good and fine and reload IMO should have just been presented in the tooltip as "shots per battle phase" from the start. But remOve acuraccy/evasion, are you crazy? Among all the stats currently appearing on weapons, accuracy is the only one that is better in some situations then others- and thus is the only stat that contributes much Of a meaningful decision.



Wanting to remove false choice by removing complexity that adds no depth is a worthy cause- but it seems to me your first priority is seeking simplicity for it's own sake even at the expense of strategic depth.

We all desire meaningful choices in Endless Space, and in no area is this more desired than in the combat mechanics. In general, we like choices in our weapons and defenses.



To accommodate this, the designers have given us a large number of variables to showcase differences between the various systems:

1) Accuracy

2) Critical Damage

3) Minimum Damage per shot

4) Maximum Damage per shot

5) Shot per Salvo

6) Salvos delay

7) Salvo Reload time

8) Evasion

9) Hull Weakness

10) Cost

11) Tonnage

12) Accuracy adjustments for range.



The purpose of this article is to highlight that several of the variables represent only superficial differences. That while they provide choices, they do not provide meaningful choices, and as a consequence provide complexity and obscurity in exchange for little benefit. I want to stress again that I am not suggesting these mechanics have 0 benefit, just not enough to justify their complexity. I will be taking the next several posts to go over all the ideas.



Combine Min/Max Damage, Shot per Salvo, and Damage per Shot



With current disharmony mechanics, the number of shots per salvo has little effect on damage dealt:

1) Accuracy and evasion is gauged on a per salvo basis, not shot.

2) Defenses work on a percentage of damage.

3) In the current beta patch, deflectors do deflect a hard number of shots. However, that number is small compared to the number of shots fired, and ultimately does not justify the inclusion of this layer of complexity.

4) Min and max damage values are always the same in the core game and don't contribute to variance.

So we have several game factors that are ultimately doing the same thing, affecting damage dealt.

Recommendation: Provide a single damage number that calculates all of these values in one.



So for example: A medium range laser does 11 max/11 min damage per shot at 8 shots/salvo. With this adjustment, you simply have 88 dmg/salvo.

Accuracy isn't percentage damage reduction it is fixed damage reduction (with a small variance). It should be Hit chance = max(Accuracy - Evasion + Evades*0.15,100).

Stalker0 wrote:
Finally, someone came at me with the counterargument i was expecting Bilun, i salute you.



My point was not to remove the complex mechanics, just because. My point was that the mechanics are both complex, AND that they do not provide that strategic depth you were referring to.

Good to hear- and my apologies if my first response was a bit abrasive; there was a bit of knee-jerk in it.

Accuracy at the top level looks like it provides a big difference in how weapons operate. What i was trying to point out in my example is it actually doesn't. Because of the number of shots fired in a typical battle (even early game battles), accuracy doesn't provide a lot of variance. So all accuracy is really doing is slapping on a percentage to the damage the weapon will do. And if that's the case, lets just drop the middle man and remove the extra modifier all together.

I'm not sure I agree- what you've showed is that there isn't a whole lot of deviation in a given case. Deviation isn't really necessary for strategic depth. In my mind a more important consideration would be comparing how a given weapon with X accuracy matches up against a ship with 30% evasion vs how it matches up with a ship with say 70-80% evasion(or higher if the evasion boosting battle action is used).



The main depth created by accuracy as a stat so far as I can see is that high accuracy weapons are(or at least should be) drastically more effective against small-high evasion targets then low accuracy weapons. In a rather extreme example, medium range kinetics(having only .3 accuracy) without any accuracy bonuses would be guaranteed to miss the first 3 attacks against a 75% evasion target(and would have a significant chance of missing for a few shots after that). By contrast a long range beam(sporting 1.2 the best accuracy in the game) would have a 45% chance of hitting that 75% evasion target right off the bat(the first example would only reach that value after being evaded 6 times). Of course this is a very extreme example- but it demonstrates how high accuracy weapons gain a huge efficiency advantage over low accuracy weapons against enemies with high evasion.



As a result a player may choose drastically different weapons depending on whether the opponent is running a few heavily loaded battleships/cruisers/dreadnauts specialized in avoidance or is spamming evasion heavy corvettes with frequent evasion boosting battle actions.





Also keep in mind the 15% evasion depletion is in beta(the current value is 5%)- personally I think 15% is a bit on the high side and there is a decent chance it drops to 10-12% by the time the patch his final release. The lack of variance you've described is directly related to how quickly evasion depletes even when the stat does pay off.









All that said, I don't think the accuracy evasion system is perfect- in less extreme cases the gaps is smaller. But the framework is solid. in my opinion rather then removing it, we should be thinking about reinforcing the strategic role it already partially fills( good start would be reducing the upcoming boost to evasion deterioration).







Again though you are absolutely right that a majority of the combat stats are superfluous- shots fired, damage per shot, and crit chance are all effectively multiplicative increases to damage of the same magnitude in all situations. But the accuracy-evasion-deterioration mechanic is unqiue in being involved in an additive relationship- which has the intrinsic effect of making evasion take a substantially larger cut of a low accuracy weapon's damage output then it would have taken from a higher accuracy weapon; and a stat being more useful against certain enemies then others always has the potential to increase strategic depth in ship creation.





Well that's my angle in anycase- in closing I'd liek to say it's always a pleasure to debate with someone who bringers numbers and statistics to the table, even should I interpret them in a different manner

bilun wrote:


The main depth created by accuracy as a stat so far as I can see is that high accuracy weapons are(or at least should be) drastically more effective against small-high evasion targets then low accuracy weapons.

This is a solid counterargument, let me crunch some numbers and see if there is merit.