Help with Modding Phalanxes

[YourMadDoc]

To make sure no one will rout in that 30 min and I will have comparable results. Idk about other diff but on VH battle it didn't make viable difference.

[moisesjns]

You know what i do to make pikes not so useless??

I just create them with Create_unit and give them 9 experience. They tend to not be useless than. Also all i am getting from this thread is. Avoid pikemen. As historically they sucked after alexanders age and everyone pretty much got lazy and just fought each other with pikes and thus had no way to fight the new evolving threats.

[Krampus]

You know what i do to make pikes not so useless??

I just create them with Create_unit and give them 9 experience. They tend to not be useless than. Also all i am getting from this thread is. Avoid pikemen. As historically they sucked after alexanders age and everyone pretty much got lazy and just fought each other with pikes and thus had no way to fight the new evolving threats.

Naw, phalangites are still really useful. They are still the best at holding a line, even the Hellenic native phalanx. You can stretch 2-3 units of them to 4 deep and they'll still form a massive near impenetrable center capable of tying up many more units of significantly better quality, protecting your missile units and freeing up your other infantry for maneuvers on the flanks.

And they still slaughter cavalry from frontal charges, which the AI just loves to do apparently...

[Pooploop]

So, will 2.3a have definite improvements (any, actually) for the performance of the phalanx?

[Pooploop]

Quintus, anybody?

[QuintusSertorius]

We'll have to see. Some of the changes suggested by Cultist have already been applied, but no one has had a chance to do any extended playtesting with it yet. I'm busy with rewriting the native pools for half the map (Balkans and Anatolia done, Caucasus WIP now, Iran and western Steppe next) at the moment.

[rhavviepoodle]

Lots of interesting stuff in this thread. I'm guessing pooploop feels quite validated, too, after finding out there really was a problem with phalanxes. Who would have thought the problem was not that phalanxes had been changed, but that everything else had been changed? The native pool rework sounds pretty cool, too. Main problem I can see with this thread is it kinda has a bit of drama in it, but that happens sometimes.

[Cultist]

I don't know what to say.
I left with a bitter taste and am reluctant to engage this topic after posting my findings here.
There was a lot of nitpicking and bad blood going on, especially from high rep users and users with "modicum knowledge".
I still stand with my rediscovery of the two important parameters.
They work as they should but while unit radius effect is straight forward the unit height is not and it requires a lot of fidgeting and testing to find the sweet spot.
For example for an already weak unit with an unit radius of 0.3, unit height has next to no effect but it becomes more relevant when coupled with a unit radius of 0.2.
This won't fix the main problem, it can't be fixed with any EDU editing i believe.
But it does make a unit weaker or stronger compared to the top ones without modifying shield values or defense skills.
I stand by these values even if the M2TW designer himself would tell me that they don't do anything (which is a blatant lie btw..)
Cheers

[Gunydii]

I've done a few test runs (10 or 20) with increased mass and radius on pikes, mass, radius and height on agema and i think ive found a significant confound that needs to be controlled in any testing. formation depth.

As an example, Pikes vs elite KH hoplites you get big differences in casualties inflicted (if i was 'really' playing probably battle outcome too) based on the depth of the pike formation. if you have your line 4 deep each unit sort of bows in the middle, the unit gets separated down the middle and your pikes die with the KH hops at like 80 to 90. keep your line 10 deep and whilst you still may lose, its because your right flank eventually gets eaten (and in a lot of tests - with other tiers of hops, not THAT many with elites - their right will break first). casualties On the KH are up to 60 or 70 - excluding the unit that eventually eats your right.

If you protect those flanks (I've only tried one of these tests and it got fiddly due to the ai being less dutiful re head on charging), i think its conceivable the pikes either last (functionally) forever or even straight up win.

In terms of the 'look' of the pike formation, 'by eye' its better from a deeper base, but infiltration on the opponent charge animation is still a real problem (my only solution to this was to give opponents a silly low mass as proof of concept, but that just had them stuck in the charge animation but without actually getting into the formation). I think a significant advantage of deep formations is that (if a few enemies make it through) you have men at the backs and sides who can turn and engage without it 'taking spear points away' from the main body of enemies.

All this is based on putting together from this thread what i think the recommendations are (eg 1.9 mass, 2.5 radius on Phalangitai) so if ive not followed this could be of limited use. But on these numbers, in proper use, I think pikes are strong, if they're strong in quite the way they're supposed to be, i really couldnt say.

[rhavviepoodle]

Does mass have much of an effect on being charged by cavalry and that sort of thing? While it seems jacking up the phalangites' mass would give us one desirable outcome with regards to melee animations, I worry that making numbers phalanx-centric could drag other parts of the game down.

[Gunydii]

Does mass have much of an effect on being charged by cavalry and that sort of thing? While it seems jacking up the phalangites' mass would give us one desirable outcome with regards to melee animations, I worry that making numbers phalanx-centric could drag other parts of the game down.

Short answer, no idea, but id imagine so, others will know. But higher phalanx mass seems to have relatively modest impact on infiltration. Tried progressively higher values, to no great effect up to silly values (i think about 10?) without the 'frontal charge -> infiltration' problem going away. the thing that seemed very important was the aggressors mass.

Lower masses (at least at the very low end) made the problem go away utterly- and creating the new problem of endless charge animations, but hey, such is life. Indeed the phalanxes vs very low mass opponents would often get kills without actually appearing to be 'in combat'. I'm not saying its the way to go, but I'd concur with earlier assessments that aggressor mass has a real impact even beyond just adding more to phalanx mass to compensate.

As a very vaguely related aside I was slightly surprised how hard it was for the AI to break the moral of my phalanx line. I 'fairly' regularly saw my phalangites fight to the death, infact they seemed way way less flighty than all flavors of hoplites I sent at them, spartans and elites included. I'd personally had a similar 'feel' from my campaigns, but it was pretty stark a contrast in this kind of controlled environment. I think if the frontal performance gets definitively 'fixed' I'd be vaguely worried about the line being impossible to break moral wise too.

[YourMadDoc]

If you place unit with mass 0.1 against units with mass 1 effect will be much stronger as if you place unit with mass 1 against unit with mass 10. The difference is counted by some formula not just by comparing mass value. That's why giving phalanx 10 mass had much lower effect as giving very low mass to hoplites. If you wish to express ability of phalanx to stop unit with so high mass as hoplitai, you will have to give them extremely high mass.

I tried 50 mass for 4 phalangitai against 6 Hoplitai and no one was able to charge through first row or get stuck in charge animation.

If you have problems with enemy units avoiding a direct charge on your line you might try this file. It is experimental alpha version so make sure you back up original file.

EDIT On what difficuly were you playing when your phalanx fought to the death ? On medium your units get boosted(or enemies gets crippled). I tried versus 800 hoplitai haploi head on charge on medium and I won loosing 167 hoplitai without killing enemy general or moving with my units twice.

[Gunydii]

EDIT On what difficuly were you playing when your phalanx fought to the death ? On medium your units get boosted(or enemies gets crippled). I tried versus 800 hoplitai haploi head on charge on medium and I won loosing 167 hoplitai without killing enemy general or moving with my units twice.

I was playing on medium I think *somewhere* in here it was mentioned as the appropriate difficulty for testing, I know these tests are something of a mess as in campaign armies under fms (especially AI fms) get buffs re moral. your results (and mine by reflection) are probably why theres a greater interest in how the fight works rather than pure numbers.

That said id be interested to know if the outcome was similar if you controlled the phalanx and didnt attack (i believe the 'done thing' had tended to be get phalanx leave phalanx alone') i presume the ai doesnt behave like that.

[YourMadDoc]

I was playing on medium I think *somewhere* in here it was mentioned as the appropriate difficulty for testing, I know these tests are something of a mess as in campaign armies under fms (especially AI fms) get buffs re moral. your results (and mine by reflection) are probably why theres a greater interest in how the fight works rather than pure numbers.

That's of personal preference. I have met too many armies led by a captain or general get himself killed in first few minutes to consider medium as an option. Player's armies are led by a superb generals and units fight to death my whole camapaign. AI has a few medicore generals and they are done after one battle. AI is penalized enough, there is no need to penalize them even in battle.

That said id be interested to know if the outcome was similar if you controlled the phalanx and didnt attack (i believe the 'done thing' had tended to be get phalanx leave phalanx alone') i presume the ai doesnt behave like that.

I'm sorry but I'm afraid I wasnt clear. I played with 800 hoplitai haploi vs 800 hoplitai haploi not against enemy phalanx. I tested same units against each other to compare their performance on medium difficulty.

In battle 11 charging hoplitai haploi vs 4 defending phalangitai, phalangitai won no matter if they are controled by the player or AI.

[Aper]

Sorry for necroing, but i find this thread very interesting and I think I can add something to the discussion.
I ran some test battles and I found that mass difference is indeed the most important factor influencing the behaviour of phalanxes versus other units: the mod may benefit from some tweaks in this area.
Another interesting and unexpected result is that adding the "prec" attribute improves a lot the charging behaviour of javelin infantry, namely they don't penetrate the phalanx all the way to the rear with some man while others watch and do nothing anymore; instead, they throw their weapons and charge all together, stopping some at sarissa lenght and some at shield distance (this is inluenced by mass). If anyone can confirm this I'd appreciate, thanks.
My 2 cents.
EDIT I think I should mention my changes: I gave Phalangitai 1.9 mass and I upped a bit their attack and defense to 6 each (because right now they have the stats of levies). I accordingly upped the Agema to 9 attack and 9 defense and 1.95 mass. You can tell I disagree with the idea that phalanxes were just mobile walls meant to stop enemies passively 'cause this doesn't make a whole lot of sense IMHO. Now they also kill, albeit slowly.
I tested them 3 vs. 3 against pedites extraordinarii or principes mainly, with useless toxotai generals, and IMO they perform fine, winning slowly and with many casualties in frontal clashes (phyrric victories, you may say).

[QuintusSertorius]

I've reduced the mass of all non-phalanx units in the mod to cap at 1.2.

It's a shame about the prec attribute; while it might have some value here, we removed it because it led to a lot of units freezing in mid-preparation. The AI suffered quite badly from this.

I don't know now if the trade-off we made in removing it might have proved worse than putting up with all the freezes, given it does bad things to charging and apparently impacts phalanx infiltration as well.

[Aper]

Too bad about the prec attribute, It seems that after so many years this game is still quote buggy.

[QuintusSertorius]

prec is a vestigial attribute from RTW that I don't think was ever properly accounted for in M2TW. I don't believe it appears in the vanilla game, nor do many other mods use it.

We're not changing the kill power of phalanxes, though. When they're not being infiltrated, they have a huge advantage in terms of WYSIWYG - they can attack without being attacked in response.

[Aper]

Slightly off-topic, but can I ask why you reduced the radius of pretty much all units? I mean the last value in the following line

soldier machimoi_toxotai, 64, 0, 0.8, 0.35

Is it really necessary for missile unites, like in the example above?

[Kull]

Another interesting and unexpected result is that adding the "prec" attribute improves a lot the charging behaviour of javelin infantry, namely they don't penetrate the phalanx all the way to the rear with some man while others watch and do nothing anymore; instead, they throw their weapons and charge all together, stopping some at sarissa lenght and some at shield distance (this is inluenced by mass). If anyone can confirm this I'd appreciate, thanks.

This was a VERY valuable insight, thanks for posting! Quintus and I spent all day testing out variations on the javelin unit stats, and in the process we may very well have solved the "broken prec" problem AND the broken phalanx. As a result, there's a good chance that battlefield activity will be significantly improved in v2.35