[A16] Melee Skill Rebalance [Updated 3/04/2017]

[Canute]

Some suggestions:

- a parry should 100% interrupt the current attack of the parry one entity.

- maybe introduce riposte too. When the entity who parry got enough skill (10+) he got a chance for a extra mellee attack if he wield a mellee weapon.

- Range weapons should be damaged too an a parry.

- Mellee attack should general have a chance to interrupt the others attack. Do you can still shoot at someone who poke you with a knife ?

[oreganor]

Thanks CHjess and Canute for the detailed feedback:

Initial feedback: Melee weapons break pretty darn fast.

Had a colonist fight off a wild hog with a club. The club broke pretty fast from all the parrying. Some way to mitigate weapon from taking damage could be nice i guess.

ATM balance is focused on High Tech melee warriors, wielding Steel Weapons and using personal shields, so they don't become Godly Killing machines. So it's a bit of a small scenario for such a rich game as Rimworld.

Was the colonist skilled in melee? ATM a parrying weapon receives (1 - Melee Skill) * Original Damage / 2. With a minimum of 1 point of damage. So the better a pawn is in melee the less damage his/her weapon receives on a parry. Notice that material of an item has a HIGH impact on its resulting HP.

As with everything related to fine tuning, the scenario & mods you play with have a saying in all of this. Can you ellaborate a bit more on the kind of game you are playing? I'm interested in adjusting the mod so ppl playing in different ways find it useful.

- a parry should 100% interrupt the current attack of the parry one entity.

That would backfire, because high skilled fighters parry a lot... If they wield a slow weapon, they will be effectively denied of any chance to attack by a less skilled warrior wielding a faster weapon.

- maybe introduce riposte too. When the entity who parry got enough skill (10+) he got a chance for a extra mellee attack if he wield a mellee weapon.

Ripostes are typical of most RPGs with some focus on melee mechanics, so I thought on them, the problem is the ammount of blows that is currently exchanged between fighters on the typical melee duel in Rimworld (Even without this mod on). If I had to make Ripostes common enough to matter, a skilled pawn subject to attack from multiple sources will become a killing machine because his applied DPS will be amplified by the number of attackers. I prefer to keep giving players a way to deal with a powerfull melee enemy by sending multiple fighters against it.

There is also the technical problem of triggering actions ON ANOTHER PAWN in the "tick" of the pawn that can potentially receive them. Sequences like that can be sabotaged easily as "first" and "last" can't be clear nor assumed. In time, when I get more familiar with modding, I will be able to do "stunts" like that.

- Range weapons should be damaged too an a parry.

The initial tests of this mod damaged any weapon hold... Until I saw what did happen when an AI shooter lost its weapon. It was very easy to exploit for a Human Player.

- Mellee attack should general have a chance to interrupt the others attack. Do you can still shoot at someone who poke you with a knife ?

I see that you are concerned about that painful scenario on vanilla when your melee pawn has managed to dodge all ranged fire, able to reach it target... Just to be shot miserably to death at point blank . It's one of the original issues that drove me to make this mod for "personal use" in the 1st place.

I plan on including "interrupting" mechanics as a result of good attacks in the form of Disarms & Stuns. It's the "Attack Mode" button that's in the TODO list.

[oreganor]

= NEW VERSION 1.1.0 =

Changes:

- Melee Weapons that would be destroyed on a parry are droped to the ground instead. If there is not enough room nearby for the weapon, it will be destroyed instead.

Warning!!!: Your melee pawns will be droping items on the ground with low HP... They will not last much exposed, so if you value a melee weapon and want to repair it... Do not let it linger around in the open too much .


For as long as I don't move the project to a GITHUB repository, I will keep the current version and the previous as attachments on the Original Post. Just in case you find something you don't like on the current version and prefer to keep using the old one.


The mod is STILL savegame friendly. When/If I do something that leaves a trace on the savegame I will clearly announce it.

[oreganor]

Sorry for delaying the answer to your very usefull comment, PetWolverine:

I think ideally manipulation would affect armed but not unarmed combat, so losing fingers would reduce a pawn's effectiveness at wielding a sword, but scyther blades, as a hand replacement with 20% manipulation, would retain their tradeoff of producing highly specialized melee super-soldiers who can't really do anything else.

I documented the different efects on manipulation as requested by user TAA6 on Vanilla and on some very popular mods, to find 3 issues:

- Most hediffs affecting manipulation ALSO affect conciousness. Meaning that to simmulate Melee Skill degradation due to temporal hediffs is enough with the current weight of modifers.

- Some Unarmed Melee Enchancing Bionics reduce the efficiency of the Body Part chain they are attached to. In turn, Body Part chain efficiency degrades any Capacity they are linked to. Sadly, Manipulation is a component on most upper limb definitions.

- Loosing completely a Body Part chain, triggers a loss on efficiency also. Which should affect BOTH Armed & Unarmed melee Capacities.


So I see now why Ludeon left manipulation out of melee modifiers ... It's more important to simmulate the Productivity degradation ATM, while leaving the option to specialize pawns as melee fighters.

I need to think carefully on how to do this WITHOUT forcing other content creators to redefine the way their Parts/Bionics/Hediff behave... Vanilla gives me tools to do it such as creating new Capacities, spliting Melee Chance stat between Armed and Unarmed... This is a hard nut to crack, so it will be very enlightening to solve it gracefully, but I will add it to the pipeline. I need to do a decent Draft even before adding this to the TODO list.

[ChJees]

Thanks CHjess and Canute for the detailed feedback:

Initial feedback: Melee weapons break pretty darn fast.

Had a colonist fight off a wild hog with a club. The club broke pretty fast from all the parrying. Some way to mitigate weapon from taking damage could be nice i guess.

ATM balance is focused on High Tech melee warriors, wielding Steel Weapons and using personal shields, so they don't become Godly Killing machines. So it's a bit of a small scenario for such a rich game as Rimworld.

Was the colonist skilled in melee? ATM a parrying weapon receives (1 - Melee Skill) * Original Damage / 2. With a minimum of 1 point of damage. So the better a pawn is in melee the less damage his/her weapon receives on a parry. Notice that material of an item has a HIGH impact on its resulting HP.

As with everything related to fine tuning, the scenario & mods you play with have a saying in all of this. Can you ellaborate a bit more on the kind of game you are playing? I'm interested in adjusting the mod so ppl playing in different ways find it useful.

The colonist survived, barely. It was very early on and i did the "Start with nothing" challenge . Also i know all about this with materials.
The colonist in question had a melee skill of 12 something with burning passion. Also was a Brawler in question if that matters. Did a lot better against a horde of manhunting hares with a steel gladius .

Feel like this calls me to make a mod that let Smiths mend melee weapons in a sensible way . A Anvil, the Weapon and the appropiate Stuff it is made out of.

[oreganor]

Also was a Brawler in question if that matters. Did a lot better against a horde of manhunting hares with a steel gladius .

So you found that turning to a faster and sturdier weapon felt better?

I chose the "balance" point regarding pawn skill versus unskilled entities (ie ALL animals) at exactly 10, so going higher will make your pawn harder and harder to be hit.

On the current test version I finnaly found a way to implement "Parry Counters" so it will become increasingly difficult to defend against multiple concurrent targets in the near future , although.

On the subject of Brawler Trait, I have pending an investigation on why on the tooltip of melee chances it doesn't seem to have an effect... Because I considered it overpowered combined with this Mod for the advertised x 1.75 bonus to Melee Chance. I need to go back to it... Because Brawler + skill above 10 should make your pawn a deadly foe with this Mod against animals.

[Canute]

- a parry should 100% interrupt the current attack of the parry one entity.

That would backfire, because high skilled fighters parry a lot... If they wield a slow weapon, they will be effectively denied of any chance to attack by a less skilled warrior wielding a faster weapon.

Thats intend,  the one with the slower weapon got draw into defence, thats why the ripose comes in handy. He got chances for counterattacks.
And with the backpack function, you can carry a second weapon. Slow weapon vs. range, fast vs. mellee.

- Range weapons should be damaged too an a parry.

The initial tests of this mod damaged any weapon hold... Until I saw what did happen when an AI shooter lost its weapon. It was very easy to exploit for a Human Player.

Isn't any weapon a mellee weapon at first hand, and at emergency you can shoot with them.
If you don't want that range weapons got damaged, then you should remove it for mellee only weapons too. And just keep the reduced damage on a parry.
And the parry ability of range fighter wouldn't happen that much ,Except a brawler raider got a mellee weapon. So these exploid wouldn't count that much, but that should be tested.

[oreganor]

Also was a Brawler in question if that matters.

Mmm... Apparently if ILSpy info is corrent and my understanding of Ludeon code is correct, I have an idea on why Brawler doesn't have an impact on reported melee chances (My debug code is telling the same... Not a surprise, because I use what vanilla games calculates as Melee Chance, which is what Vanilla game uses to determine if a melee attack lands or not).

The code responsible of calculating ALL stats of pawns is ignoring StatFactor tags defined on Traits (Honors StatOffsets, although)...

...I may miss something here because I swear that young animals got their stats changed when changing Lifestages and those are defined precissely as StatFactors... But on another entity.

[XeoNovaDan]

The code responsible of calculating ALL stats of pawns is ignoring StatFactor tags defined on Traits (Honors StatOffsets, although)...

This is at least hopefully going to be fixed in Alpha 17 though... but then again, traits like Careful shooter, Trigger-happy and Fast walker work fine. It's only Brawler AFAIK that is broken

[oreganor]

This is at least hopefully going to be fixed in Alpha 17 though

Ah k, so it's a known issue then? (I was about to fill a bug report)

[XeoNovaDan]

It is indeed a known issue. I think the original bug report for it was filed a few months ago, but I can't remember off the top of my head.

After checking, it has indeed been filed and resolved: https://ludeon.com/mantis/view.php?id=2586

[henk]

Have you considered increasing the changes of hitting vital areas with highly skilled warriors? It would make sense that the level 20 person is aiming for the throat, eyes and head areas compared to the wildly swinging level 0. Also, perhaps you would consider even hitting internal organs as well? Like some sort of martial art.

[oreganor]

= NEW VERSION 1.2.0 =

Changes:

- Full Harmony Library integration. Thanks to pardeike work and advices, not only detours, but access to private data/methods goes through Harmony Library.

- Parry chances of a pawn are reduced the more parries it has to do between his/her melee attacks.

Some remarks about Parry Counters mechanic:

- Parry Counters only accumulate in the case of successful parries or hits that go through one.

- The current penalty for parries beyond the 1st one is a flat -25% chance per extra parry (-50% for 2, etc, etc). I'm not happy with the way it's evaluated AFTER skill normalization happens ingame. I want that pawns with effective skill chance above 100% can also "negate" this penalty as they do with other factors like wounds, fatigue, etc. ATM trying to alter this would require "touching" deeper methods used by Vanilla that would increase the chance of a "mod collision". Once I get more practice with modding, I will revise this to see if I can make Elite fighters to be able to parry more times before seeing their chances reduced.

- Weapon Speed gives a net advantage in 1 v 1 duels. The faster pawn will sometimes manage to connect 2 or more attacks between swings of his target, meaning the 2nd and following hits will have less chances to be parried. Obviously this is amplified if the same pawn is been attacked by multiple enemies. Take this into consideration when having to deal with a Veteran Figher as enemy.

- If a Pawn is retreating from melee, will not have his/her parry counters refreshed, meaning that hitting a fleeing target will become easier over time.

- ATM Parry Counters AREN'T SAVED. This means that if you save in the middle of a melee combat, when you reload, ALL pawns will have their parry chances back to full. It's a small discrepancy that I hope you don't mind for the sake of been able to keep this MOD savegame friendly for now.



This will be the last version I deploy as an attachment. The next one will come as a download from a GITHUB repository and will contatin also the source code of the mod. It may take a bit longer than the previous updates as I have to refactor/clean my chaotic code full of test/debug/trash lines .


The mod is STILL savegame friendly. When/If I do something that leaves a trace on the savegame I will clearly announce it.

[XeoNovaDan]

Keep up the good work!

I genuinely think at some point you should suggest this mod to be integrated into vanilla, when it's at the right stage

[oreganor]

Have you considered increasing the changes of hitting vital areas with highly skilled warriors? It would make sense that the level 20 person is aiming for the throat, eyes and head areas compared to the wildly swinging level 0. Also, perhaps you would consider even hitting internal organs as well? Like some sort of martial art.

Thanks for the feedback. Yes, I have considered it at draft time... Then I checked how incredibly detailed Bodies are modeled into Rimworld's code and...

...Decide to model extra lethality in a simpler way as I described in the TODO list.

My next target, after code cleanup -> GITHUB migration, will be to add "extra lethality" options in the shape of toggle buttons in the UI.

It's not the kind of detailed effects you seek... But it's how far I can go for now regarding increasing lethality of Expert Fighters while keeping a decent chance of providing to you what I promised.