Statement regarding Plans for Singleplayer and Engine III

Users who are viewing this thread

While it is ok that armor gets some improvements, I actually prefer how armor works in Bannerlord compared to Warband. Looking an armored guy surrounded by 10 Looters while receiving 0-1 damage is really silly in my view.

On the other hand, armor is not as useless as people say, just pick 100 legionaries and fight 100 looters and check the result, legionaries are going to win that battle without getting any single loss and we can probably get the same result if testing legionaries against some T2/T3 infantry units.

For me, the biggest issue with armor currently is about it is unrealistically weak against arrows, armor should protect more against arrows in general. Thinking about melee combat, I am pretty sure that even if you wear the best possible armor, you are going to get killed most of the time if fighting 5 dudes in melee at the same time or if you get a good hit from someone with a big weapon (maybe not killed but falling unconscious).
 
Was Warband actually mentioned in the particular discussion you're quoting?
There's a middle ground between "get Swadian Knights in a week" and "take a year to get T5 troops unless you cheese a Stewardship perk".
I brought it up because everyone here agrees that Warband was fun. You need common understanding as the basis for discussion.

Anyway, it doesn't even take that long to get T5 troops. You can do it in about twelve battles relatively safely and probably in like six if you just don't give a damn if half die in the process. I'm not the only person who has brought that up; people choose to train up troops in the grindiest manner possible then complain it is grindy.

But taking looters and making them an actual threat wouldn't make the game fun for those people. It would just make an already not-fun thing they do as work while getting to the fun parts into something even more not-fun.
Because sometimes there's a pack of 20+ looters and 1/2 of them bombard you with stones at max distance and it can kill your horse while you're trying to slowly shoot them to death with the starter bow.
...
SO it's wrong to say "people ***** about 9 points of damage against an HP pool of 100+ on this forum" because that is too broad of a complaint and seems unreasonable out of the context of the primary acute complaint that is very reasonable and concerning.
20+ looters.
A sumpter horse.
The starter bow.

This is an example of what I'm talking about: a 100% avoidable situation that is a bit risky but not impossible to win.
I'm sure if you try you can dig up me saying how I killed a garrison alone and the peasants with rocks were to hardest part because of the chip damage building up over time, but even that is just to illustrate the disparity between just having any ranged weapons versus normal infantry being fodder and also the absurdity of the blunt damage formula and the armor formulas. It's never as simple as "It's too much damage".
If I was going to use an example of something you said, it would be this:
@Apocal I went in there to fire squad them and a little stinky militia kill one of my Khan's Guards! Reeeeeeeeeeeeeeeeeeeeee-set
You are the exact example I had in mind when I said wrote my post about people being attached to their troops in a way the AI never will be.

Your counterargument is that Bannerlord in its current state, with genuine challenge added to the current mix of luck and grind, wasn't fun for you.

Of course not, but I said "similar resources to the player" and "intuitive game mechanics".
I don't have a problem with raising or training troops quickly in BL (it doesn't take long if you don't only use looters to do it), the AI's ten dudes free on respawn or with them uprgading mounted troops without horses. Those bits are totally trivial to overcome in comparison the AI factions (plural) showing up with 8-12 armies ready to wreck shop. It is a genuine challenge because the AI isn't given free influence and something like 94% of its troops are cheat-free in the late-game, using strategies that are fairly intuitive (go for the weak settlements) but the AI isn't a player: losses don't sting the way they do for a player, where losing stuff means spending time rebuilding. Potentially a lot of time if the AI speed-sieges your kingdom in half.
 
You are the exact example I had in mind when I said wrote my post about people being attached to their troops in a way the AI never will be.
And if the AI actually does what I want I can out range them and massacre them without loosing a good unit, if it doesn't I re-set the game because it lazy garbage programing that needs to be fixed and I don't respect it. Militia shouldn't even be able to damage t6 units at that range, that's BS too and I don't respect it. Every tier and type of unit needs a distinct identity and role and relative power level to be used correctly and fallow the players instruction 100% of the time, anything less is just a cluster**** garbage pile. "eh they fight and some die and some don't..." isn't what I want in a war game!

I'm attached to them because I invested time into them! The AI does absolutely nothing deliberate at all.

20+ looters.
A sumpter horse.
The starter bow.

This is an example of what I'm talking about: a 100% avoidable situation that is a bit risky but not impossible to win.
Looters only exist to skill up as a new character and raise up recruits. That's all. Anything that makes thier 1 and only function suck more is a bad game design choice. You better bet I don't have sumpter horse and I've got 3 stacks of arrows and when I see that 30 stack of looter I'm gonna kill em and I'm gonna hate every moment of it! They rout and I'm gonna just hit em again and again! I resent every rock I have to dodge! I resent every arrow that nearly hits thier head! I resent every one that makes it to red wall before I kill it! I resent the garbage loot they drop that's not worth the effort at all! I resent everything about looters their design is bad and it was much better in warband!
 
The entire pretense of more fancy or interesting mechanics == harder bots that nobody wants to deal with is utterly false. They are simply two very different things. For instance if i created an AI for looters that did more varied and interesting type attacks -i would then playtest it 10,000 to make sure

A. Its fun and rewarding
B Its properly calibrated as to not make them overpowered in an unbelievable way

people like fighting against unpredictable and versatile AI -nobody wants aimbot AI that are merely tweaked too high in accuracy - thats not fun gameplay. Neither is simply lowering their damage or making player bullet/stone sponges with armor buffs. These are a "poor mans" game design -totally uninspired.
 
I brought it up because everyone here agrees that Warband was fun. You need common understanding as the basis for discussion.
Warband was indeed a fun game, but again I'm not sure who you're quoting who claims that they found it fun only because you could get gamebreaking stuff relatively easily if you already knew what to look for.

Me personally, I liked it because some elements of it were actually better balanced than Bannerlord is now, parts of it that were meant to be challenging were more challenging (like higher tier enemies vs. a levelled player) and parts of it that weren't meant to be challenging were less challenging (like low tier enemies vs. a levelled player), and because it has a bunch of features that Bannerlord is still missing.
Taking looters and making them an actual threat wouldn't make the game fun for those people. It would just make an already not-fun thing they do as work while getting to the fun parts into something even more not-fun.
I'm a bit confused here because I already agreed that looters shouldn't be challenging for a mid/lategame player.

My stance is that in the mid/lategame, T5/T6 troops and nobles should provide the challenge, while T0/T1 is something you murder your way through, when you are well equipped and leveled.

Maybe you think when we say "challenge" we want every single thing in the game to be challenging at all times? Because that's not what I'm saying. But I am saying that grind, luck, and exploits should be reduced in the game, and replaced by a good scale of challenges for each level of player progression.
It is a genuine challenge because the AI isn't given free influence and something like 94% of its troops are cheat-free in the late-game, using strategies that are fairly intuitive (go for the weak settlements) but the AI isn't a player: losses don't sting the way they do for a player, where losing stuff means spending time rebuilding. Potentially a lot of time if the AI speed-sieges your kingdom in half.
Yeah, so give the players more sensible and skill-based ways to avert their armies being destroyed/fiefs being taken, or to rebuild in the event of failure. Rather than dumbing down the AI or allowing exploits to continue to exist.

Your scenario of the AI intelligently targeting the player kingdom when they're the biggest threat would be so, so much more fun and much less of a slog if:
* You had more control over your vassals and kingdom. It sounds like your war declaration changes made the AI intelligent on a "macro" level at targeting the biggest kingdom (no matter how they choose to do that), but on the "micro" level they're still stupid, choosing targets without good reasoning. So all AI factions are directed the right overall way when attacking, while your faction is undirected every which way when defending, and the net result is a big disadvantage for the player kingdom. It would be much better if you could just tell your vassals "go to that particular place that needs defending". Or if they were smarter.
* Potential defectors and mercs approached you, as it was in Warband for defectors, instead of having to chase them to every corner of the earth to talk to while also being the only one intelligently defending the faction's borders.
* You could delegate the process of rebuilding your own armies and getting reinforcements better.
* You could help your raided fiefs recover.
* Being a ruler and Influence were more useful in changing the outcome of a vote so the player ruler was less at the mercy of stupid AI decisions.
* Lords went for longer periods of peace so the player has grace periods to recover and doesn't get dogpiled so much. The frequent war merry-go-round isn't realistic anyway.
* Getting peace was easier (and this is on its way).
* Battlefield tactics and morale were more impactful, so that if you're facing a doomstack on realistic difficulties, rather than needing exploits, you can make use of high-tier troops whose tier actually matters, flank attacks, shock charges, braced spears, etc to achieve low casualties and high kills.

That particular scenario doesn't need to define this whole debate either. I should probably walk back my original statement a bit, the AI doesn't have to be exactly as smart as the player, but the point is we don't want it to be so flat out stupid and exploitable. In general I just want to see Bannerlord have less luck, less grind, smarter AI, and more player control. Do you agree with that?
 
The AI should be as smart as possible, and if its cleverness is too much for the average player, then it could easily be dumbed down on lower/normal difficulty levels. But for those that want a real possibility of getting their asses kicked if they slip up, i.e. the hardcore players, no AI could be smart enough.
 
The AI should be as smart as possible, and if its cleverness is too much for the average player, then it could easily be dumbed down on lower/normal difficulty levels. But for those that want a real possibility of getting their asses kicked if they slip up, i.e. the hardcore players, no AI could be smart enough.

Is input reading AI smart? That's where I draw the line between hard and annoying.
 
Warband was indeed a fun game, but again I'm not sure who you're quoting who claims that they found it fun only because you could get gamebreaking stuff relatively easily if you already knew what to look for.
I said (paraphrasing) things that made Warband good were unrelated to difficulty. I've also said (more than once) that Warband did something smart by defaulting its difficulty to the easiest settings then burying them under menus. Too many people play Bannerlord on max difficulty and make the game less fun for themselves in the process.
Maybe you think when we say "challenge" we want every single thing in the game to be challenging at all times? Because that's not what I'm saying. But I am saying that grind, luck, and exploits should be reduced in the game, and replaced by a good scale of challenges for each level of player progression.
Well, there is one person arguing looters should be dangerous to players but yes, I agree in broad strokes. Personally, there is nothing that can make me care about an individual troop's challenge once I have a decent party but I don't think you'd agree with that.
Yeah, so give the players more sensible and skill-based ways to avert their armies being destroyed/fiefs being taken, or to rebuild in the event of failure. Rather than dumbing down the AI or allowing exploits to continue to exist.

Your scenario of the AI intelligently targeting the player kingdom when they're the biggest threat would be so, so much more fun and much less of a slog if:
* You had more control over your vassals and kingdom. It sounds like your war declaration changes made the AI intelligent on a "macro" level at targeting the biggest kingdom (no matter how they choose to do that), but on the "micro" level they're still stupid, choosing targets without good reasoning. So all AI factions are directed the right overall way when attacking, while your faction is undirected every which way when defending, and the net result is a big disadvantage for the player kingdom. It would be much better if you could just tell your vassals "go to that particular place that needs defending". Or if they were smarter.
* Potential defectors and mercs approached you, as it was in Warband for defectors, instead of having to chase them to every corner of the earth to talk to while also being the only one intelligently defending the faction's borders.
* You could delegate the process of rebuilding your own armies and getting reinforcements better.
* You could help your raided fiefs recover.
* Being a ruler and Influence were more useful in changing the outcome of a vote so the player ruler was less at the mercy of stupid AI decisions.
* Lords went for longer periods of peace so the player has grace periods to recover and doesn't get dogpiled so much. The frequent war merry-go-round isn't realistic anyway.
* Getting peace was easier (and this is on its way).
* Battlefield tactics and morale were more impactful, so that if you're facing a doomstack on realistic difficulties, rather than needing exploits, you can make use of high-tier troops whose tier actually matters, flank attacks, shock charges, braced spears, etc to achieve low casualties and high kills.
Just to be clear, the reason I lost frequently wasn't for lack of the things you listed. It was because it was 8-12 armies all coming after me, while my faction only maintained 2-3 armies to stop them. Obviously, those numbers don't add up, even if everything you listed was implemented into Bannerlord. They could invest and assault fiefs faster than I could physically reach them in most cases, just by intelligently using the advantage of more armies fielded.

(Caveat: this was way back when factions habitually rolled multiple armies; now factions average one army in the field. I don't know when it changed but mexxico mentioned it has been this way for a long time. I'm not sure that 4 armies vs. 1 changes things compared to 8 armies vs. 2 but it might ?)
That particular scenario doesn't need to define this whole debate either. I should probably walk back my original statement a bit, the AI doesn't have to be exactly as smart as the player, but the point is we don't want it to be so flat out stupid and exploitable. In general I just want to see Bannerlord have less luck, less grind, smarter AI, and more player control. Do you agree with that?
I agree: I don't want it to be flat out stupid and exploitable either.
 
Is input reading AI smart? That's where I draw the line between hard and annoying.
I think it's gotta be used sparingly as it will achieve desired results of the bots putting up a fight against the player but it shouldn't be for low tier units and needs to be watched to make sure it doesn't seem fake and obnoxious.

The AI should be as smart as possible, and if its cleverness is too much for the average player, then it could easily be dumbed down on lower/normal difficulty levels. But for those that want a real possibility of getting their asses kicked if they slip up, i.e. the hardcore players, no AI could be smart enough.
This is true, but with lots of caveats. And this is especially true for combat and I get that it's a hard point for me to get across.
I make good troops, I use them the best way I can, if then enemy manages to defeat a few of them it's only because my sides AI had a woopsy. The enemy AI is so ineffective that there's no legit way it can get any of my guys or me, so when it does, if feels like a cheap shot and a undeserved loss of material, it's annoying and I might just re-set the game because I know full well I can take that fight zero losses.
"Oh so you want it easier so you don't lose troops?" NO, I want the AI to actually perform well and put up a fight so that my losses feel deserved and unavoidable! Like I did the best I could each battle and that's that.

And campaign side of war really needs some other utility or value to fiefs that extend a faction's borders. It needs this and then it needs the AI to commit to an objective in war. I do not like or respect the current AI magically know the power value of everything and constantly changing it target like a chicken with it's head cut off. "Oh you want them to be dum and attack things they shouldn't" NO, I want them to have clear reason to take one fief and commit to it, if they can't do it then they back off and make peace and... get this...it's really hard to grasp, The AI parties should go SPEND TIME fighting some bandits and do other things until thier parties are built up enough to have enough power to launch a campaign to take a fief! You know like they were the same type of thing the player is?
 
I do not like or respect the current AI magically know the power value of everything and constantly changing it target like a chicken with it's head cut off. "Oh you want them to be dum and attack things they shouldn't" NO, I want them to have clear reason to take one fief and commit to it, if they can't do it then they back off and make peace and... get this...it's really hard to grasp, The AI parties should go SPEND TIME fighting some bandits and do other things until thier parties are built up enough to have enough power to launch a campaign to take a fief! You know like they were the same type of thing the player is?

This is exactly why they needed to flush out lords personalities and traits. Let this be a guide to their actions -not just math. If a Lord really hates another Lord/Faction for whatever reason -he may act brash and fight against the odds simply because he hates them. This is far more interesting them just constantly making magical mathematical calculations as an On/Off switch to deciding combat
 
I'm not the only person who has brought that up; people choose to train up troops in the grindiest manner possible then complain it is grindy.
Yeah, I still think people in general sleep on autocalc. It's super useful for reducing grind, and it's really not unfair unless you expect to not lose a single troop like you would in live battles or you're dragging around a party full of cavalry (which is pretty much unnecessary anway). A numerical advantage and a good surgeon are basically all you need to come out of fights mostly unscathed. Good tactics skill helps too. I basically autoresolve all looter, bandit, and village militia fights from the start of the campaign these days.

I'm actually doing a strictly autoresolve only ironman run right now to see how it plays out, and it's going just fine; I have no problem keeping my party full of high tier troops. So I agree, spending hours grinding looters is totally unecessary to succeed in this game, and therefore completely optional.
 
it's really not unfair unless you expect to not lose a single troop like you would in live battles
Is it really unreasonable though for me to expect that if I have a party of 50 Banner Knights facing 10 looters that I should not be at significant risk of losing anyone? As opposed to currently where you'll quite often have an elite troop die in that scenario?

I think the purpose of autoresolve should be allowing the player to skip through boring battles which are a foregone conclusion. Punishing the player for using autoresolve by taking a troop who never in a million years would have died otherwise means players don't want to use the feature because let's face it, we do get attached to our troops.

Is it really such a bad thing if it was more accurate to the results of an actual battle? I'm not saying I should be able to autoresolve something like, say, 20 T1 vs. 20 T6 with no risk of casualty. Any sort of remotely close battle, you should risk losing troops if you autoresolve it. But utter curbstomp battles should always play out like they would if you actually played them.
 
Can I ask why you would even do that though? If you have 100 Banner Knights there shouldn't be much reason to be fighting looters in the first place.
I was exaggerating a tad, and I think I edited my post to be less exaggerate-ey at the same time you made your post (50 knights vs 10 looters).

But this same complaint applies for parties of mixed level, where you might have just a few elite troops. It almost feels like they're getting picked out to die by autocalc, despite that it would never happen in real battle.
 
And campaign side of war really needs some other utility or value to fiefs that extend a faction's borders. It needs this and then it needs the AI to commit to an objective in war. I do not like or respect the current AI magically know the power value of everything and constantly changing it target like a chicken with it's head cut off. "Oh you want them to be dum and attack things they shouldn't" NO, I want them to have clear reason to take one fief and commit to it, if they can't do it then they back off and make peace and... get this...it's really hard to grasp, The AI parties should go SPEND TIME fighting some bandits and do other things until thier parties are built up enough to have enough power to launch a campaign to take a fief! You know like they were the same type of thing the player is?
Basically, you want the AI to adopt proven player strategies, which would make it smarter. But how would Taleworlds devs know this is they don't play the game or follow what players are doing? They seem to live in a bubble disconnected from their player base.
But this same complaint applies for parties of mixed level, where you might have just a few elite troops. It almost feels like they're getting picked out to die by autocalc, despite that it would never happen in real battle.
Not that I know any autocalc details, but this kind of complaint looks biased. Players would notice and remember autocalc outcomes that seem unfair to them, but forget the rest of unremarkable and more realistic outcomes, so they will have an exaggerated impression of the percentage of "unfair" outcomes. Particularly if they are already primed to be upset by previous "unfair" losses and forum talk of similar experiences.
They should be doing Autocalc Science instead, like true gentlemen and scholars.
 
It almost feels like they're getting picked out to die by autocalc, despite that it would never happen in real battle.
Okay this isn't meant as an argument at all, I loathe the auto calc too, but I 100% guarantee Bannerknights and other t6 Cavalry will get killed by looters and recruits in live combat sometimes. Cavalry just dies to it's own speed applied to trash unit's attacks, and all units seem over tuned to hit mounted units, meanwhile mounted units seems to have low success rate in landing thier attacks, regardless of skill and tier. It sucks.

I wish auto-calc had the option to send only certain troops or groups, that would make it very useful and ditch this problem all together. I would very much like to smash up recruits and other rif raf into looters or bandits (or Korte) and have some wounded for my medic to skill up on, but I don't want to lose my good units randomly. I often have 1/2 good units and 1/2 fodder units I intend to put in harms way for my medic.

Basically, you want the AI to adopt proven player strategies, which would make it smarter. But how would Taleworlds devs know this is they don't play the game or follow what players are doing? They seem to live in a bubble disconnected from their player base.
Well.... because I keep telling them on this forum! It's all I can do! I tried making cool videos! I tried streaming my gameplay! Nobody wants to watch the video game equivalent of those animal stomping vids that get banned on youtube. And I get it: it's much more amusing to watch a player blunder through battles and get knocked of his horse all the time and listen to them speculate on how perks and mechanics might work.
 
I was exaggerating a tad, and I think I edited my post to be less exaggerate-ey at the same time you made your post (50 knights vs 10 looters).

But this same complaint applies for parties of mixed level, where you might have just a few elite troops. It almost feels like they're getting picked out to die by autocalc, despite that it would never happen in real battle.
Sure, I understand why it's frustrating to lose an expensive troop to looters. However, autoresolving looters is rather economical overall, in that you will almost always earn more troop XP as a whole from the fight than you lost in dead troops. If you have a bunch of mixed tiers, you will still usually come out positive, despite losing a high tier troop or two, especially if you have a good surgeon. And that holds true up until your party starts to be mostly composed of tier 4 troops and higher (since losing a T4 troop loses all of the XP it took to get from T1 => T4). So at that point I will make the call to transition away from autoresolving looters and go get some real value out of my troops by fighting stuff that matters. A party of T4+ troops is going to blow through most lord parties anyway. You don't need anywhere close to a max tier party to accomplish stuff in this game, and to me, It's more fun when you don't.

Now of course when you lose a mounted troop you will lose a horse or horse + warhorse. So If you have a lot of cavalry it can be more costly, but the 125 medicine perk Veterinarian also works for autocalc, so you will get back the troop's equipped horse 50% of the time if you have it. I still think that in those cases you will usually gain more than you lose, since autoresolve really favors stronger parties.

Here's what happens when you throw 100 Banner Knights up against a typical lord party in autocalc.
5xnod.png
 
Last edited:
@Apocal Sorry for slow response.
I said (paraphrasing) things that made Warband good were unrelated to difficulty.
I can agree with that.
It was because it was 8-12 armies all coming after me, while my faction only maintained 2-3 armies to stop them. Obviously, those numbers don't add up, even if everything you listed was implemented into Bannerlord.
If there were longer periods of peace and lords were dogpiling into war less often, then there would be less attacking armies versus your defending armies, and everything else I listed would make it easier to deal with the armies or clean up after losing to them. For example, easier lord recruiting for the player means you can accelerate the process of weakening enemy factions.

But yeah, that's all moot since I agree that particular scenario wouldn't be very fun and we don't need it in gameplay. The AI needs to be smarter and more challenging, but not player-smart.
I agree: I don't want it to be flat out stupid and exploitable either.
I'm glad we're on the same page, and you always give valuable contributions to discussion. The issue for me is when you're playing devil's advocate that kind of comes across as defending silly things even though I'm fairly sure you're not.

@Ananda_The_Destroyer
I loathe the auto calc too, but I 100% guarantee Bannerknights and other t6 Cavalry will get killed by looters and recruits in live combat sometimes
Okay terrible example on my part since we all know how garbage lance-using cavalry units are at doing their job, which is a separate issue in itself. Just replace "banner knight" with "high tier units in general".
Not that I know any autocalc details, but this kind of complaint looks biased. Players would notice and remember autocalc outcomes that seem unfair to them, but forget the rest of unremarkable and more realistic outcomes, so they will have an exaggerated impression of the percentage of "unfair" outcomes. Particularly if they are already primed to be upset by previous "unfair" losses and forum talk of similar experiences.

They should be doing Autocalc Science instead, like true gentlemen and scholars.
There's potentially confirmation bias at play, but it really does seem like tier isn't weighted- or isn't weighted enough - when taking into account which troops die in autocalc. Feels like the system just decides the player needs to have X amount of troops die and then picks them at random, even if it's your one Elite Cataphract when you have an army of T1-T2 fodder that should be much more likely to die. Perhaps Bannerman knows the actual code on it though.
 
When will banners be added to the roadmap? Right now you can just add them cosmetically. People will be happy.
Don't be silly, just because the game is called BANNERlord doesn't mean you get it in the game. You should be glad that you have contributed with your money to this company, that should be considered the "final product" not the game itself. Now be a good lad and buy 3 more copies.
 
Last edited:
Back
Top Bottom