Slitherine

stockwellpete wrote: ↑Tue May 19, 2020 2:20 pm

stockwellpete wrote: ↑Mon May 18, 2020 4:27 pm

Schweetness101 wrote: ↑Mon May 18, 2020 4:13 pm and an interesting related question from the flank angle mod thread is:

should pikes be exempt from the changes to flank attacks, ie should they still, like in vanilla, suffer 200 poa disparity and autodrop when 90 degree flanked?

I think we need to measure the difference between vanilla and the alternatives of no automatic cohesion drop with +100POA or no automatic cohesion drop with +200POA. I can start to do that tomorrow if you like. 50 sets will be enough. I will need a +200POA variation of the mod. If you can tell me what line to change then that is all I will need.

I have done the testing for this today . . .

Vanilla
automatic cohesion drop and then +200POA
28 out of 50 no further cohesion drop
15 out of 50 further drop to fragmented
7 out of 50 double drop to routed

Alternative 1
no automatic cohesion drop and just +100POA
41 out of 50 no cohesion drop
7 out of 50 cohesion drop to disrupted
2 out of 50 double drop to fragmented

Alternative 2
no automatic cohesion drop and +200POA
29 out of 50 no cohesion drop
15 out of 50 cohesion drop to disrupted
6 out of 50 double drop to fragmented

I think Alternative 2 might be hitting the sweet spot here. In vanilla, nearly half the sample are either fragged or routed after just one impact phase from an attack from the side, while in Alternative 2 only about 1 in 8 are. 40% are disrupted though.

Alt 2 sounds good, and that's just for pikes? Are you thinking that flanks against infantry should be:

+100 and no autodrop on non-pikes
+200 and no autodrop on pikes

?

Schweetness101 wrote: ↑Tue May 19, 2020 6:46 pm
Alt 2 sounds good, and that's just for pikes? Are you thinking that flanks against infantry should be:

+100 and no autodrop on non-pikes
+200 and no autodrop on pikes

?

Yes, that's correct. Because you were saying that pikes should be more vulnerable on their flanks in line with your pike mod.

stockwellpete wrote: ↑Tue May 19, 2020 6:53 pm

Schweetness101 wrote: ↑Tue May 19, 2020 6:46 pm
Alt 2 sounds good, and that's just for pikes? Are you thinking that flanks against infantry should be:

+100 and no autodrop on non-pikes
+200 and no autodrop on pikes

?

Yes, that's correct. Because you were saying that pikes should be more vulnerable on their flanks in line with your pike mod.

right, looks good to me

ok forum, please review the below algorithm for determining if a unit ought to anarchy charge (work in progress)

Precheck (ie if liable to charge at all):

1) If unit routing, fragmented, in close combat, pursuing, or off playable area, it cannot anarchy charge
2) check if target exists, is valid, is within charge range, in line of sight, and otherwise a valid charge based on pre existing algorithms for that sort of thing from the vanilla AI (i.e. cannot charge across diagonal that is occupied by two units already in combat, that sort of thing)

Chance to Anarchy Charge function:

*all percents are like points out of 100 going up and down, NOT multiplying the chance to charge by 1.x where x is the percent. It's just adding up.
*charge chance and anarchy charge chance both mean chance to anarchy charge here, i used them interchangeably

1) set base chance to charge value from 'Anarchic' column from squads file (something like 60,40,20, and 0 for different unit types, or whatever)
2) various bits of boilerplate code to acquire yours and the enemy's facing, location etc...for determining combat margins if the combat were to occur, disparities in relative shooting capacity, terrain disorder, etc...the raw material values for making your comparisons. This is not really a step so much as a preparation.
[FOG1 anarchy type rules checks follow in number 3]
3) a) Foot will not anarchy charge out of their own protective terrain, b) no unit will anarchy charge into protective terrain or against elephants (dunno about the into river banks bit...), c) no unit will anarchy charge if doing so would cause them to be disordered by the terrain (accounts for diff amounts of disorder based on unit type), d) medium foot (including warriors, bowmen, mob, medium foot), if in rough or difficult terrain, will not anarchy charge an enemy in the open, and e) no unit will anarchy charge lights (definitely up for debate, wasn't sure if this was exactly part of the original rules, but would help make them less annoying).
4) determine relative combat and shooting advantages (close combat margin and shooting margin), to be used for later comparisons (also more like a preparation than a step)
5) if charger has less than 50% chance of winning, decrease charge chance by 10%. If a much less than 50% chance of winning, decrease charge chance by an additional 20% (30% down total by this point)
6) if enemy is ranged and you are not, increase charge chance by 10%
7) if you are ranged, regardless of if enemy is, decrease charge chance by 20% (would be irrelevant if charge chance for shooters is 0 to begin with in all cases, but might end up only being relevant for shooty lancers)
if at major shooting disadvantage, whether ranged or not, and not at too great of a combat disadvantage, increase charge chance 20% (ie increase charge chance, even if at a little bit of disadvantage (or better), if you are getting shot at)
9) if marginal combat advantage, increase charge chance 10%, and if large combat advantage, increase charge chance 20% (not additional 20%, either or)
10) if you are foot and the enemy is mounted shock, then decrease anarchy charge chance by 60%
11) if you are mounted and steady and the enemy is fragmented, then increase charge chance 10%
12) if you are better off being charged than initiating the charge, and you have shooting advantage, then greatly decrease charge chance (by dividing it by 5). I think this is basically to prevent shooty lancers from anarchy charging when they ought to keep shooting and waiting to receive a charge.
13) if outside of general's command range then increase chance to anarchy charge by 20%
14) if mounted and the target are non fragmented pikes or spears then decrease charge chance to zero
15) if charging will involve opening yourself up to flank threat then decrease chance to anarchy charge down to zero.
16) do not allow anarchy charge chance to exceed 90% (ie always retain a random element)

thoughts on the numbers? The numbers are just made up by me and not tested at all. What should the different points values be for different conditions? new factors I am missing?

I do not think Disrupted units should charge, since they are seemingly unsure and apt to be less hot headed.

I think items 3 and 5 should be looked at again. My understanding of Anarchy would be troops acting impulsive manner, not considering the odds. In many of the items listed if the unit did not Anarchy charge then the player would do it themselves. As far as leaving a defensive position, any General that would put Gallic mercenaries in the breastworks is asking for trouble. I think the presence of a General would be the way to hold a Anarchy prone unit in place, not them seeing reason and understanding the need to defend instead of attack.

I also do not think an Anarchy charge should be all bad, how about a POA plus up for troops who think victory is assured (hence the charge)?

I also think they should be unpredictable. Do you want to revisit a range of 3? This would cause both sides a lot of consternation and require extra thinking on both sides.

desicat wrote: ↑Wed May 20, 2020 12:07 am I do not think Disrupted units should charge, since they are seemingly unsure and apt to be less hot headed.

I think items 3 and 5 should be looked at again. My understanding of Anarchy would be troops acting impulsive manner, not considering the odds. In many of the items listed if the unit did not Anarchy charge then the player would do it themselves. As far as leaving a defensive position, any General that would put Gallic mercenaries in the breastworks is asking for trouble. I think the presence of a General would be the way to hold a Anarchy prone unit in place, not them seeing reason and understanding the need to defend instead of attack.

I also do not think an Anarchy charge should be all bad, how about a POA plus up for troops who think victory is assured (hence the charge)?

I also think they should be unpredictable. Do you want to revisit a range of 3? This would cause both sides a lot of consternation and require extra thinking on both sides.

I kind of agree with the reasoning you have behind relooking at 3 and 5, but those (3 anyway) are the FOG1 anarchy rules just carried over basically. I think 3 and 5 are compromises made to ensure that the gameplay does not suffer too much from the most annoying or ill considered anarchy charges. We'll have to just discuss that I guess as we go along.

I guess it's kind of fine if the player would charge anyway? dunno.

I think a POA bonus for anarchy charges would be hard to balance or justify, but we could reconsider that after getting a basic prototype up.

You mean charge distance of 3 tiles by 'range of 3'? That is still up in the air, whether to use the unit's AP, or always 2 tiles or what. It would be controversial to enable anarchy charges to go further than commanded charges though, or to permit otherwise impermissible charges. It's an interesting idea though

Schweetness101 wrote: ↑Wed May 20, 2020 12:20 am

desicat wrote: ↑Wed May 20, 2020 12:07 am I do not think Disrupted units should charge, since they are seemingly unsure and apt to be less hot headed.

I think items 3 and 5 should be looked at again. My understanding of Anarchy would be troops acting impulsive manner, not considering the odds. In many of the items listed if the unit did not Anarchy charge then the player would do it themselves. As far as leaving a defensive position, any General that would put Gallic mercenaries in the breastworks is asking for trouble. I think the presence of a General would be the way to hold a Anarchy prone unit in place, not them seeing reason and understanding the need to defend instead of attack.

I also do not think an Anarchy charge should be all bad, how about a POA plus up for troops who think victory is assured (hence the charge)?

I also think they should be unpredictable. Do you want to revisit a range of 3? This would cause both sides a lot of consternation and require extra thinking on both sides.

I kind of agree with the reasoning you have behind relooking at 3 and 5, but those (3 anyway) are the FOG1 anarchy rules just carried over basically. I think 3 and 5 are compromises made to ensure that the gameplay does not suffer too much from the most annoying or ill considered anarchy charges. We'll have to just discuss that I guess as we go along.

I guess it's kind of fine if the player would charge anyway? dunno.

I think a POA bonus for anarchy charges would be hard to balance or justify, but we could reconsider that after getting a basic prototype up.

You mean charge distance of 3 tiles by 'range of 3'? That is still up in the air, whether to use the unit's AP, or always 2 tiles or what. It would be controversial to enable anarchy charges to go further than commanded charges though, or to permit otherwise impermissible charges. It's an interesting idea though

Units already get POA bonuses for high experience or morale, I see a hot headed Anarchy charge as the same thing. The possible charge range of 3 would be to keep players honest and prevent goofy preventive measure - like the player stopping short to prevent charges or the enemy player trying to bait certain units. If the player stops short Anarchy units may leave the rest of the line behind. An enemy trying to lure a charge will have to defend their flanks from an approaching line with room to maneuver. It would only need to be possible, a low percentage but possible.

I also see problems with item 15, this too supposes the unit is considering the odds - and depending upon when the charge occurs and from what distance it would eliminate almost all occurrences.

The bottom line should be to add flavor, not make Anarchy units miserable to command - there has to be some upside or the game just would not be fun. To control Anarchy prone units the player should always have the option of using their Generals.

desicat wrote: ↑Wed May 20, 2020 12:07 am I think items 3 and 5 should be looked at again. My understanding of Anarchy would be troops acting impulsive manner, not considering the odds. In many of the items listed if the unit did not Anarchy charge then the player would do it themselves.

But acting in an impulsive manner does not mean suicidal; they charge because they think they are going to, or at least might, win. Take this out and you'll see all kind of silly charges, such as lancers, irregular foot, etc. right into the front of a phalanx. That's exactly the kind of thing I don't want to see.

In my view the point of anarchy charges should not be to cause senseless, suicidal charges, but to remove some degree of perfect control from the omniscient player. Sure, for many of the resulting anarchy charges the player might "do it themselves", but many times they would not, or would do so in a different sequence or manner, or the next turn, etc.

76mm wrote: ↑Wed May 20, 2020 1:43 am But acting in an impulsive manner does not mean suicidal; they charge because they think they are going to, or at least might, win. Take this out and you'll see all kind of silly charges, such as lancers, irregular foot, etc. right into the front of a phalanx. That's exactly the kind of thing I don't want to see.

In my view the point of anarchy charges should not be to cause senseless, suicidal charges, but to remove some degree of perfect control from the omniscient player. Sure, for many of the resulting anarchy charges the player might "do it themselves", but many times they would not, or would do so in a different sequence or manner, or the next turn, etc.

that is also how I see it. I also think with something like this that isn't even in the vanilla game we should lean towards not disrupting gameplay too much. Do enough anarchy to add flavor and remove some of the weirdness with micro without forcing stupid decisions.

Schweetness101 wrote: ↑Tue May 19, 2020 11:37 pm 14) if mounted and the target are non fragmented pikes or spears then decrease charge chance to zero

Not sure, but perhaps charges against disrupted pikes and spearmen should be allowed? Not sure of the odds at that point... Also, this should probably be limited to frontal charges against pikes or spearmen, if that is possible?

Actually with #5, not sure if you need this specific rule at all--any really dumb stuff should drop out automatically.

Also, for #5, perhaps drop the percentage to 40%; you want to weed out suicidal charges, but I not sure if we should fine-tune it too much.

Also, maybe I missed it, but what range are you working with now? 3 seems odd to me, don't get that one. I think limiting it to adjacent hexes is an interesting idea, but haven't thought through all of the potential issues.

76mm wrote: ↑Wed May 20, 2020 1:51 am

Schweetness101 wrote: ↑Tue May 19, 2020 11:37 pm 14) if mounted and the target are non fragmented pikes or spears then decrease charge chance to zero

Not sure, but perhaps charges against disrupted pikes and spearmen should be allowed? Not sure of the odds at that point... Also, this should probably be limited to frontal charges against pikes or spearmen, if that is possible?

Actually with #5, not sure if you need this specific rule at all--any really dumb stuff should drop out automatically.

Also, for #5, perhaps drop the percentage to 40%; you want to weed out suicidal charges, but I not sure if we should fine-tune it too much.

maybe allow anarchy charges against disrupted pikes and spears if the combat modifier is good, but maybe just let the player choose if they want those, dunno. These don't prevent you from charging of course if they don't roll.

it should be possible to limit to frontal charges. And/Or, I could do some across the board increase chance to charge if you get a flank, although I think #5 might effectively take care of that like you mentioned. Some rules are kind of a failsafe to prevent really stupid stuff that shouldn't, but still might, get past #5.

by "perhaps drop the percentage to 40%" do you mean by 40? If not, I kind of need to do it 'drop by x%' rather than 'drop to x%' because it's x% changes from a base anarchy charge chance determined by unit type (except for dropping to 0% for some things of course).

76mm wrote: ↑Wed May 20, 2020 1:57 am Also, maybe I missed it, but what range are you working with now? 3 seems odd to me, don't get that one. I think limiting it to adjacent hexes is an interesting idea, but haven't thought through all of the potential issues.

It's limited to the unit's AP currently. You can anarchy roll for any charge you could normally make, but no more than that. I've considered altering that, and there is basically a whole side discussion to be had on that one issue.

If you can anarchy charge further than normal charges, you could end up anarchy charging some impermissible charges, and maybe confusing and angering existing players, but could also add some interesting aspects to the goading of anarchy troops.

Limiting it to 1 tile distance charges would make it easier, but perhaps too easy, to control and game.

Limiting it to 2 tiles for everyone (or normal ap for infantry, but half for cavalry) would make more sense if it's a being within earshot type thing that makes you more likely to charge, and that should be the same for everything, and make it easier to control cavalry anarchy.

I dunno, there is a lot to be said about that specific issue.

Schweetness101 wrote: ↑Tue May 19, 2020 11:37 pm 5) if charger has less than 50% chance of winning, decrease charge chance by 10%. If a much less than 50% chance of winning, decrease charge chance by an additional 20% (30% down total by this point)

I mean instead of "if charger has less than 50% chance of winning" say "40% chance of winning". Although does "winning" mean during initial impact, during an extended melee, or what?

Although perhaps this calculation should be tied to the likelihood of losing during initial impact rather than "winning"? In other words, look at the rightmost number in the pre-combat odds calculation. In other words, if the odds are 10/80/10 they would anarchy, but not if the odds were 10/60/30 (just stuck in arbitrary numbers to make the point).

Schweetness101 wrote: ↑Wed May 20, 2020 2:07 am Limiting it to 1 tile distance charges would make it easier, but perhaps too easy, to control and game.
****
I dunno, there is a lot to be said about that specific issue.

Agreed, the range would have a dramatic effect on the impact of the mod, just not sure how it should play out.

76mm wrote: ↑Wed May 20, 2020 2:11 am

Schweetness101 wrote: ↑Tue May 19, 2020 11:37 pm 5) if charger has less than 50% chance of winning, decrease charge chance by 10%. If a much less than 50% chance of winning, decrease charge chance by an additional 20% (30% down total by this point)

I mean instead of "if charger has less than 50% chance of winning" say "40% chance of winning". Although does "winning" mean during initial impact, during an extended melee, or what?

Although perhaps this calculation should be tied to the likelihood of losing during initial impact rather than "winning"? In other words, look at the rightmost number in the pre-combat odds calculation. In other words, if the odds are 10/80/10 they would anarchy, but not if the odds were 10/60/30 (just stuck in arbitrary numbers to make the point).

as in is their combat margin negative, is how it is calculated, but basically if they are at disadvantage at all, and then if at significant disadvantage. And of winning on a mixed consideration of impact and melee, calling an existing function called CalculateModifiedCloseCombatRating() which is described as follows:
// Returns unit's close combat rating against an individual enemy unit, taking into account own morale and disorder. The resulting rating is only used for AI purposes, not to calculate results of combat.
// If melee_only flag set, reports melee rating only, otherwise reports rating based on 1/3 impact, 2/3 melee.
and not setting melee_only flag

the numbers go the other way around though, ie the close combat outcomes are simulated hundreds of times in a row to produce the W/D/L rating ad hoc (that's why it changes a bit if you hover over the tooltip, on and off, repeatedly)

Schweetness101 wrote: ↑Wed May 20, 2020 2:56 am ie the close combat outcomes are simulated hundreds of times in a row to produce the W/D/L rating ad hoc

1000 times.

Great ideas here. I’m really looking forward to your work!

I definitely think that all the units should be prone to anarchy charges (of course with different chances of occurrence) and that winning chance of the impact phase should be taken into consideration. Otherwise the anarchy rule could be abused. Let’s take an example of phalanx in the open next to medium foot standing in rough terrain. Unless there is a high chance of winning the impact phase/there is no commander/one of the units is under heavy missile fire/etc., no unit should be willing to leave its favorable terrain. Player should never be certain that if he moves close to the opponent, that his unit will not charge and the other sooner or later will.

I'll start by saying that I don't really have a dog in this fight but did sometimes enjoy the anarchy-prone armies in FoG1; that frisson of fear and excitement as your army got to the point where it may just self-destruct

Genuine open question: by tying unexpected/unauthorised charges up in a lot of detailed logic, are you not watering 'anarchy' down so much that it becomes an irrelevance or solely an inconvenience? Does this suggested mod reduce it to a bookkeeping exercise where you can play the percentages so not really reducing control in any way?

History is littered with examples of people who thought it was a really good idea to charge when with the benefit of hindsight it really, really wasn't

Schweetness101 wrote: ↑Tue May 19, 2020 11:37 pm 3 e) no unit will anarchy charge lights (definitely up for debate, wasn't sure if this was exactly part of the original rules, but would help make them less annoying).

In FOG1 a deliberate tactic was to use your lights to bait enemy units in defensive positions (e.g. on a hill) so that they would charge off it onto more level terrain. It was annoying, but it was a poor general who tried to fight a defensive battle with a whole load of shock troops in that game, particularly if the other player had skirmisher supremacy.