Information about developments at snowballing problem

Users who are viewing this thread

In my experience people like selective realism, usually stuff that feels authentic but is really just there to add an air of authenticity. You try adding soldier stamina, dysentery tearing through your troops, seasonal-only campaigning, realistic frequency of battles and non-magic wealth assets and suddenly the pendulum swings firmly back to gamey artifice, and I don't use that term pejoratively.
Yes but there’s a lot of real estate between dysentery and simply implementing land borders and resource based cassus belli
 
Yes but there’s a lot of real estate between dysentery and simply implementing land borders and resource based cassus belli
Would be awesome, but considering their lack of desire to make significant changes to the map leaves me to believe we will have to wait for a mod if anything.

We getting a little off topic here tho.
 
Yes but there’s a lot of real estate between dysentery and simply implementing land borders and resource based cassus belli

But not between that and a 'Mount & Blade real Medieval simulator' which is what I was commenting on. It's selective realism, lots of real estate or not (and there really isn't).
 
Information about problem found yesterday :
Today I made several changes at code and today they are catching 733 men army with these changes. Yesterday this 733 men army was escaping from this bad situation because of fast and small parties were not engaging it, they were coming closer and running away back. However I need to test if there is any side effects. These codes are a bit dangerous to change because they effect many things. Also we have to understand what to do in miliseconds for each party in world because these calculations are done in all map (1000s of parties) so these calculations should be done so effectively means we cannot check all parties in a large area for each party, also it become like AI cheating. This development also will have positive effects on snowball imo, maybe minor, also walking in enemy lands will be more dangerous now (good for gameplay). However more important than snowball this was probably so annoying for player to experience these cases.

y1NcJ.png
Hey Mexxico, since you are tweaking this in parties (and looks like an amazing tweak), any chances we will see also a tweak to ai army behavior like in Party AI overhaul and Commands that makes armies focus on bordering fiefs to siege instead of bypassing them most of the time creating border-gore that's also more difficult to defend and hold like in vanilla?

Description of that feature from the mod's page:

"Kingdom AI Army Leaders will now prioritize besieging enemy towns/castles closest to their own hometown. Their hometown is usually the most prosperous castle/town that their Clan owns (this is a vanilla feature). The idea is that they somewhat selfishly prioritize taking lands close to their home so that their fiefs becomes safer. Armies won't besiege towns/castles deep into enemy territory anymore. An added bonus is that armies with different leaders will have distinct behaviour."
 
But not between that and a 'Mount & Blade real Medieval simulator' which is what I was commenting on. It's selective realism, lots of real estate or not (and there really isn't).

Cmon man that silly and doesnt bring anything to the discussion. Selective realism -equating a status quo feature of 95% of strategy games - cassus belli for land/resources and a probably 0.001% for dysentery is a false equivalency.

The overall point being taleworlds seem to go out of their way to create these artificial determinations and mechanisms for conflict when they are just skirting the most obvious solution. Create some map borders, resource system, populace needs and lords with distinct personalities and boom you got a dynamic strategy game.
 
Last edited:
It is more like a balance problem, as I give Empire and Sturgian equipment that matches their tire, and other factions remain the original. They can actually push back to Vlandia and Khuzait. Without me, the player interference. So the gears that affect the field battle also affect the automatic battle result for AI lords.

Maybe try to make a careful and decent troop balance first before messing around other mechanism like reduce the cavalry speed on the world map. So far, the reducing cavalry speed did nothing on solving snowballing but against the experience of players. The Empire will lose most automatic battles against Khuzait regardless of how fast Khuzait move on the map.
 
It is more like a balance problem, as I give Empire and Sturgian equipment that matches their tire, and other factions remain the original. They can actually push back to Vlandia and Khuzait. Without me, the player interference. So the gears that affect the field battle also affect the automatic battle result for AI lords.

Maybe try to make a careful and decent troop balance first before messing around other mechanism like reduce the cavalry speed on the world map. So far, the reducing cavalry speed did nothing on solving snowballing but against the experience of players. The Empire will lose most automatic battles against Khuzait regardless of how fast Khuzait move on the map.
Gear does not impact simulations. It looks at the tier and then if the unit is mounted, that's it. You simply got lucky in your playthrough :cry:
 
Cmon man that silly and doesnt bring anything to the discussion. Selective realism -equating a status quo feature of 95% of strategy games - cassus belli for land/resources and a probably 0.001% for dysentery is a false equivalency.

The overall point being taleworlds seem to go out of their way to create these artificial determinations and mechanisms for conflict when they are just skirting the most obvious solution. Create some map borders, resource system, populace needs and lords with distinct personalities and boom you got a dynamic strategy game.

I really don't follow. Are we talking medieval simulator or 95% of strategy games? I was answering the former. Predefined borders? I assume you mean like modern nation states have. There were none in medieval times aside from maybe major rivers. They were porous and royal authority rarely extended beyond places of importance like castles and cities, which Bannerlord kinda has (though mostly accidently as it's part of the design). I really think we got our wires crossed.
 
ppl are never happy with what they have.....only thing i kinda dislike in the game, is the fact everytime we go to a ally town/castle our companions dump the troops on it, making impossible to get them back.other then that game is amazing...
 
ppl are never happy with what they have.....only thing i kinda dislike in the game, is the fact everytime we go to a ally town/castle our companions dump the troops on it, making impossible to get them back.other then that game is amazing...

it worked in 1.5.6
i haven't tried 1.5.7 yet.
 
Gear does not impact simulations. It looks at the tier and then if the unit is mounted, that's it. You simply got lucky in your playthrough :cry:
As I gave Battanian wildling troop line with longbows that match their tire. Battanian starts conquering like crazy, as I change back to the javelin, it's slowing down.(changed their weapon skill too) And I also loaded a save where both south and north empire already lost most territory to Khuzait, which they manage to hold the rest of their territory after the gear upgrade and fought back for some time. If those are pure coincidences, then I would be way too lucky.

But yes, all those are my mod free game experience that will be affected by other reasons, so I got no solid proof for the above.

However, The troop tree needs to rebalance anyway, there are countless mistakes for me to spend weeks fixing like Sturgian heavy spearman will randomly spawn "Nordic hauberk" instead of their "Sturgian cavalry armor" that shows in the troop tree. The Khuzait raider has a better bow than its horse archer, low tire soldiers like Sturgian woodsman and Battanian skirmishes randomly spawn their noble bracer and plated gloves. mercenary cavalry lost their shield because the shield is in the same slot with their spear so they can not spawn their shields and more.....

The troop tree looks and functions poorly than many nexus troop mods. TW should do something to improve the game experience, not the other way around.
 
As I gave Battanian wildling troop line with longbows that match their tire. Battanian starts conquering like crazy, as I change back to the javelin, it's slowing down.(changed their weapon skill too) And I also loaded a save where both south and north empire already lost most territory to Khuzait, which they manage to hold the rest of their territory after the gear upgrade and fought back for some time. If those are pure coincidences, then I would be way too lucky.
HBLtdVT.png

This is the entire GetPower used for autocalc to determine troop strength. It only care about three things:
1. Hero unit?
2. Tier?
3. Mounted?

As of a few patches ago, the mounted bonus was disabled in sieges. That changed almost nothing about snowballing because most sieges are 'overdetermined', as they put it in mil-speak. Battania does uncommonly well because it has (or had) an uncommonly high amount of cavalry in its party compositions. That was it. If you adjusted Battania's troop tree to eliminate the excess cavalry, then their performance dropped accordingly.
 
...

However, The troop tree needs to rebalance anyway, there are countless mistakes for me to spend weeks fixing like Sturgian heavy spearman will randomly spawn "Nordic hauberk" instead of their "Sturgian cavalry armor" that shows in the troop tree. The Khuzait raider has a better bow than its horse archer, low tire soldiers like Sturgian woodsman and Battanian skirmishes randomly spawn their noble bracer and plated gloves. mercenary cavalry lost their shield because the shield is in the same slot with their spear so they can not spawn their shields and more.....

The troop tree looks and functions poorly than many nexus troop mods. TW should do something to improve the game experience, not the other way around.

For me those are mostly not mistakes but variety. I find it terrible if all units have uniform equipment, it's a medieval and not 19th c. AD setting. I give my custom troops (from ATC) all different equipment as much as possible. If not all in a tier have the best armor, that's life.

There are several mistakes in the troop tree (like having a onehanded sword as sidearm, but twohanded stats) but it's not that bad.
 
HBLtdVT.png

This is the entire GetPower used for autocalc to determine troop strength. It only care about three things:
1. Hero unit?
2. Tier?
3. Mounted?

As of a few patches ago, the mounted bonus was disabled in sieges. That changed almost nothing about snowballing because most sieges are 'overdetermined', as they put it in mil-speak. Battania does uncommonly well because it has (or had) an uncommonly high amount of cavalry in its party compositions. That was it. If you adjusted Battania's troop tree to eliminate the excess cavalry, then their performance dropped accordingly.

I cannot change troop trees. I can give this as suggestion but I do not think they will change it. So these changes are out of my limits.

Also when examining your screenshot I realized someone changed power calculation formula. It was not like this before. There was higher strength difference between tier-0 and tier-5 troops like 10x as it should be. Because 5 villagers cannot win aganist 1 tier-5 troop. I checked and see it is changed at 1.4.0 but I could not find who changed it. Asked info about this at office comunication channels but probably I will not get any answer (communication is not so good there). I am placing an image to show you old formula. Thats why players are losing high tier troops at simulations aganist even looters. However if we make this 10x like in old days probably it will be worse for snowballing but for better and equitable gameplay we need to change formula as in old days.

also related :
tier-0 troop wage is 1 / daily
tier-1 troop wage is 2 / daily
tier-2 troop wage is 3 / daily
tier-3 troop wage is 5 / daily
tier-4 troop wage is 8 / daily
tier-5 troop wage is 12 / daily

oldest formula (before EA) :
tier-0 strength : 0.19
tier-1 strength : 0.40
tier-2 strength : 0.74
tier-3 strength : 1.20
tier-4 strength : 1.79
tier-5 strength : 2.50 (6x of tier-1 (looter) 13x of tier-0 (villager))

old formula (1.0-1.3s) :
tier-0 strength : 0.40
tier-1 strength : 0.74
tier-2 strength : 1.20
tier-3 strength : 1.79
tier-4 strength : 2.50
tier-5 strength : 3.30 (4x of tier-1 (looter) 8x of tier-0 (villager))

new formula (1.4s-1.5s) :
tier-0 strength : 0.40
tier-1 strength : 0.66
tier-2 strength : 0.96
tier-3 strength : 1.30
tier-4 strength : 1.68
tier-5 strength : 2.10 (3x of tier-1 (looter) 5x of tier-0 (villager))

nWKlo.png
 
Last edited:
I cannot change troop trees. I can give this as suggestion but I do not think they will change it. So these changes are out of my limits.

Also when examining your screenshot I realized someone changed power calculation formula. It was not like this before. There was higher strength difference between tier-0 and tier-5 troops like 10x as it should be. Because 5 villagers cannot win aganist 1 tier-5 troop. I checked and see it is changed at 1.4.0 but I could not find who changed it. Asked info about this at office comunication channels but probably I will not get any answer (communication is not so good there). I am placing an image to show you old formula. Thats why players are losing high tier troops at simulations aganist even looters. However if we make this 10x like in old days probably it will be worse for snowballing but for better and equitable gameplay we need to change formula as in old days.

also related :
tier-0 troop wage is 1 / daily
tier-1 troop wage is 2 / daily
tier-2 troop wage is 3 / daily
tier-3 troop wage is 5 / daily
tier-4 troop wage is 8 / daily
tier-5 troop wage is 12 / daily

oldest formula (before EA) :
tier-0 strength : 0.19
tier-1 strength : 0.40
tier-2 strength : 0.74
tier-3 strength : 1.20
tier-4 strength : 1.79
tier-5 strength : 2.50 (13x of tier-0)

old formula (1.0-1.3s) :
tier-0 strength : 0.40
tier-1 strength : 0.74
tier-2 strength : 1.20
tier-3 strength : 1.79
tier-4 strength : 2.50
tier-5 strength : 3.30 (8x of tier-0)

new formula (1.4s-1.5s) :
tier-0 strength : 0.40
tier-1 strength : 0.66
tier-2 strength : 0.96
tier-3 strength : 1.30
tier-4 strength : 1.68
tier-5 strength : 2.10 (5x of tier-0)

nWKlo.png

Even if You change it back this simplistic formula is really bad.

Months ago I made excel sheet that simulated battles the same way and allowed me to predict outcome of 8 out of 10 equal battles. It also made me realize how big of a buff is cav buff in this calculation cause armies with bigger cav % in army composition won in simulations battles that were 90% lost for them in mission.
 
Last edited:
Even if You change it back this simplistic formula is really bad.

Months ago I made excel sheet that simulated battles the same way and allowed me to predict outcome of 8 out of 10 equal battles. It also made me realize how big of a buff is cav buff in this calculation cause armies with bigger cav % in army composition won in simulations battles that were 90% lost for them in mission.

Do you have that sheet? Can you share it?
 
So we've already got a ridiculously oversimplified Auto-Calc system that takes in a whopping 3 elements to determine outcomes in a battle and someone came in a just up and changed one of the variables to close the gap between a peasant and a top tier unit? Are AI Leadership and Tactics not even involved in this calculation??

Can we get an open letter to Armagon to remedy this situation? We deserve a real, robust auto-calc system or else all these armies mulling about are utterly uninteresting and makes the world strategy game paper thin
 
Back
Top Bottom