Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

Lord Royal wrote:
Yea, I meant stranded...
It's the only reasonable way to get to Aegurio.

And yapp, I already thought about a "vault" and a team of highly specialized experts... maybe 4... one of them should have some kind of psychic power and they should have a crappy robot as a side-kick... jokes aside. Borderlands became some kind of freakish pillar of the wasteland subgenre in the last years. One could say: The trope is strong with this one. So it HAS to be there in any way. I just don't know how (and if) it works out in a skirmisher game with free model choice. And if it's wise to have a similar faction type twice (Arkers & Vaulthunters)... but on the other hand... VAULT HUNTERS!!!

Vault hunters is just a reason to get the faction into the game - it wouldn't have to be specialist elite. It would be as likely for space pirates to have this as the reason they are here. the Arkers would be sufficiently different to he vaulthunters I feel - I imagine arkers would be quite strict on the modifications to their equipment, as they would still have their idea of society from when they went into their stasis. The vaulthunters, on the other hand, would likely have wildly customised equipment, scavenged or robbed from other races and previous finds. They could even be the ones cracking open arks for the loot, thinking that they are the vault. if the scanner says there's a locked-up ancient man-made cavern under them, what are they going to do?

From that point of view, that's (again) absolutely right... and again some very cool ideas. I like the cracking-up vaults part.

But besides that... NEWS!!!

The team is growing and I'm not alone anymore. A coder and a writer joined the ranks... and proparbly another coder will follow anytime soon.

That's why I started a wiki that will be used as living rulebook:
https://afterthefall.fandom.com/wiki/AfterTheFall_Wiki

Anyone with a fandom-account can edit it or leave comments. In the future there will be an official website of course, but at this early stage I think an open wiki is a good first step.

It's still very incomplete yet but will be filled in the next days and weeks. With lots of fluff following.

I thought about a suggestion by another poster... don't know who was it exactly and too lazy to look it up.

@some bloke was it you?

About dynamic pointcosts:

Imagine a skirmisher where your listbuilding consists of a quick grab into the shelf and pick the models you like. You don't need to track pointcosts or force organization plans. Just pick some models and have a at least semi balanced game.

How should that work?

Simple speaking:
Every model has a base power level that is calculated through its sustain and damage potential right after it's been equipped.
This base value is multiplied by a variable that is calculated through the warband's overall model count and the number of enemy attacks.
The power levels of all warband members are added to determine your warband's power level. This is the only value that'll be visable for players. The warband with lower level gets an underdog bonus which grants better chances to loot cool stuff and to benefit from random events, higher chance for critical hits etc. Additionally the underdog gets more ressources in the beginning of a game for playing cards and/or to get free extra actions.

For now it's just an idea for the most parts. I might have a working power level formula though... but it's not quite finished yet.

Certainly it's not infinitely stretchable that's why we proparbly need a simple levelling system to balance the forces even further.

Every single underdog effect won't have much impact, but alltogther could add up to a fun experience.
Could... I don't know.
I just don't know any one game that ever tried to accomplish a task like this... may be it's a great idea... may be (and that's proparbly more likely) I'm just a crazy person.

Lord Royal wrote:
Imagine a skirmisher where your listbuilding consists of a quick grab into the shelf and pick the models you like. You don't need to track pointcosts or force organization plans. Just pick some models and have a at least semi balanced game.

That was how exactly Warhammer Age of Sigmar was at first, pure narrative gaming, before they brought points back.

Yea, except with AoS it didn't work quite well until they brought points back.

As far as I know AoS didn't have any balancing compensation for uneven matchups aside from fixed scenarios. And the last time I checked it had no dynamic values to determine the combat efficiency more exactly.

Lord Royal wrote:

@some bloke was it you?

I don't think so...

My comment about dynamic points was that the system in GorkaMorka was quite effective, where they had it built into the campaign system that you gained more XP from fighting a powerful enemy than a weak one - so if the top-ranking and the bottom-ranking players squared off, the lower one gets flattened, but the top ranker gains very little from it, whereas the lower ranker gains loads of XP, which allows them to upgrade their mob more.

I also suggested tying points costs to action points - thereby balancing AA somewhat. A unit with 3 AP can do 3 actions, one with 1AP can do 1 action, so all players have the same number of actions at the start. it would also directly relate to turn length, so you know a 30AP game will have 60 actions per turn, a 10AP game would only have 20.

another option for pick-up-and-play is to put a "moral victory" chart for fun games. Essentially, if one player has 10 points more than the other, 20 points more, 30 points more etc, on the X axis and the victory points you scored on the Y axis, to show you your amended victory points.

EG if a 50pt army took on a 70pt army, and the 70pt army won 12-10 on VP, the "moral victory" chart might state that the 70pt army only got 10pts, whilst the 50pt army got 11, meaning a moral victory for the 50pt army. This would reinforce the divide between "casual" games (pick up what you want and play) and fixed point games. You could even have a basic chart which states that the higher pointed army gives X VP to the lower pointed army, meaning if these 2 had to give 2VP over, it would go from 12-10 to 10-12. it would be fairly easily balanced through lots of playtests!

you would essentially put a handicap on victory conditions, not on the actual gameplay. a smaller force has to do less to win, essentially.

Lord Royal wrote:


About dynamic pointcosts:

Imagine a skirmisher where your listbuilding consists of a quick grab into the shelf and pick the models you like. You don't need to track pointcosts or force organization plans. Just pick some models and have a at least semi balanced game.

How should that work?

Simple speaking:
Every model has a base power level that is calculated through its sustain and damage potential right after it's been equipped.
This base value is multiplied by a variable that is calculated through the warband's overall model count and the number of enemy attacks.
The power levels of all warband members are added to determine your warband's power level. This is the only value that'll be visable for players. The warband with lower level gets an underdog bonus which grants better chances to loot cool stuff and to benefit from random events, higher chance for critical hits etc. Additionally the underdog gets more ressources in the beginning of a game for playing cards and/or to get free extra actions.

For now it's just an idea for the most parts. I might have a working power level formula though... but it's not quite finished yet.

Certainly it's not infinitely stretchable that's why we proparbly need a simple levelling system to balance the forces even further.

Every single underdog effect won't have much impact, but alltogther could add up to a fun experience.
Could... I don't know.
I just don't know any one game that ever tried to accomplish a task like this... may be it's a great idea... may be (and that's proparbly more likely) I'm just a crazy person.

For a game I created using Plastic Army Men the list building phase was to grab as many guys as you wanted for your team from the pile randomly. Both sides needed the same number of guys.

This technique only worked because:
1. All models pretty much had the same stats- They all moved, shot, and died the same
2. Weapons were pretty much balanced. They had slightly different mechanics on how to use the, but the results were roughly the same, you kill other dudes
3. It was a dexterity based game using WYSIWYG resolution mechanics so there were few if any probabilities, it was based on a players "skill" at flicking stuff.
4. Ultimately, balance didn't matter too much.

So, I guess the key learnings for me were, how to dress up various "units" as unique when ultimately they were all the same. It didn't matter if someone had all bazookas while another guy only had bayonet dudes. They could still kill each other easily enough.

How to do this with random number generators is a bit trickier, but the core ideas might be useful to you. I don't know.

 Easy E wrote:

Lord Royal wrote:


About dynamic pointcosts:

Imagine a skirmisher where your listbuilding consists of a quick grab into the shelf and pick the models you like. You don't need to track pointcosts or force organization plans. Just pick some models and have a at least semi balanced game.

How should that work?

Simple speaking:
Every model has a base power level that is calculated through its sustain and damage potential right after it's been equipped.
This base value is multiplied by a variable that is calculated through the warband's overall model count and the number of enemy attacks.
The power levels of all warband members are added to determine your warband's power level. This is the only value that'll be visable for players. The warband with lower level gets an underdog bonus which grants better chances to loot cool stuff and to benefit from random events, higher chance for critical hits etc. Additionally the underdog gets more ressources in the beginning of a game for playing cards and/or to get free extra actions.

For now it's just an idea for the most parts. I might have a working power level formula though... but it's not quite finished yet.

Certainly it's not infinitely stretchable that's why we proparbly need a simple levelling system to balance the forces even further.

Every single underdog effect won't have much impact, but alltogther could add up to a fun experience.
Could... I don't know.
I just don't know any one game that ever tried to accomplish a task like this... may be it's a great idea... may be (and that's proparbly more likely) I'm just a crazy person.

For a game I created using Plastic Army Men the list building phase was to grab as many guys as you wanted for your team from the pile randomly. Both sides needed the same number of guys.

This technique only worked because:
1. All models pretty much had the same stats- They all moved, shot, and died the same
2. Weapons were pretty much balanced. They had slightly different mechanics on how to use the, but the results were roughly the same, you kill other dudes
3. It was a dexterity based game using WYSIWYG resolution mechanics so there were few if any probabilities, it was based on a players "skill" at flicking stuff.
4. Ultimately, balance didn't matter too much.

So, I guess the key learnings for me were, how to dress up various "units" as unique when ultimately they were all the same. It didn't matter if someone had all bazookas while another guy only had bayonet dudes. They could still kill each other easily enough.

How to do this with random number generators is a bit trickier, but the core ideas might be useful to you. I don't know.

Yapp, that are all thing I found out aswell. That's why I try to avoid too many combat factors that depend on what you shoot at. The relation between damage output and individual sustain is still a head cracker though. But it's solvable and even more or less exactly calculatable.
Even special abilities without conditional components that are directly tied to combat efficiency are no problem. Abilities that are tied to non-combat factors are the ones that will cause problems, that only can be solved through playtestings... I think of effects like invisibility or immunities for certain effects (ie bleed or poison).

But let's go through your list:
1) Check. Have that.
2) Check. Only some special weapons like harpoons (pull) and flamethrowers (those can set the table on fire) in exception.
3) WYSIWYG is pretty much a UI standard to keep the game easily readable. Proparbilities are not the problem. You can calculate them very precisely, IF they're not related to too many variable values that depend on your actual target.
For example: If you don't have a value that influences incoming damage, damage and hp can be compared almost equally. Or to use actual numbers: A sample gun deals 8 damage and can be fired twice per action and has a 25% chance to hit, so your avarage damage per hit is 4... not practically but theoretically. So if you attack 10 times you're likely to deal around 40 damage. It doesn't matter against what you fight, you're damage output is static and therefore easy to calculate. The only time you have to readjust is, if the average enemy health drops under your weapon's damage value in which case it is treated as being equal to the lower value pointwise... having proparbly another multiplier to adjust the relation. The downside: It's kinda boring. As watching chipping little chunks from a health bar isn't really exciting. But it's a problem I'm working on and try to solve this weekend.
4) Check(ish)... competitive play won't be the focus. The system shall provide a game full of suspense, that's why balancing isn't completely unimportant. But aside that, I don't have the ambition to do balancing for tournaments... I don't have the illusion that it's even possible to manage that in the given time frame (~1 year for the free beta version, with 3 factions).

While playing Dark Age I learned that it feels kinda right if your avarage success chance of several rolls is around 50%. It's more or less a coin flip and therefore very transparent for the player. So it serves Timmys and Spikes aswell (http://www.youtube.com/watch?v=QHHg99hwQGY&t=43m38s).

The most of the models in my game are standard infantry. Models that are bigger or heavy armoured cost a huge ammount of "points" so you're highly limited in the number of actions you can perform per round in comparison to your opponent's force. The big question is to what degree uneven point levels can be compensated through different effects. That's a thing only the upcoming next test phase can answer in the next months.


Automatically Appended Next Post:

 some bloke wrote:

Lord Royal wrote:

@some bloke was it you?

I don't think so...

My comment about dynamic points was that the system in GorkaMorka was quite effective, where they had it built into the campaign system that you gained more XP from fighting a powerful enemy than a weak one - so if the top-ranking and the bottom-ranking players squared off, the lower one gets flattened, but the top ranker gains very little from it, whereas the lower ranker gains loads of XP, which allows them to upgrade their mob more.

I also suggested tying points costs to action points - thereby balancing AA somewhat. A unit with 3 AP can do 3 actions, one with 1AP can do 1 action, so all players have the same number of actions at the start. it would also directly relate to turn length, so you know a 30AP game will have 60 actions per turn, a 10AP game would only have 20.

another option for pick-up-and-play is to put a "moral victory" chart for fun games. Essentially, if one player has 10 points more than the other, 20 points more, 30 points more etc, on the X axis and the victory points you scored on the Y axis, to show you your amended victory points.

EG if a 50pt army took on a 70pt army, and the 70pt army won 12-10 on VP, the "moral victory" chart might state that the 70pt army only got 10pts, whilst the 50pt army got 11, meaning a moral victory for the 50pt army. This would reinforce the divide between "casual" games (pick up what you want and play) and fixed point games. You could even have a basic chart which states that the higher pointed army gives X VP to the lower pointed army, meaning if these 2 had to give 2VP over, it would go from 12-10 to 10-12. it would be fairly easily balanced through lots of playtests!

you would essentially put a handicap on victory conditions, not on the actual gameplay. a smaller force has to do less to win, essentially.

nvm
I cannot find the original post... right now I'm not even sure if it wasn't in another forum XD
Anyway.

VP having different values for different warbands is a great suggestion and solves even some other design problems. Because side objective cards are a pure analog component and are therefore tracked analogically (each card giving you 1 VP to keep it simple), while scenario objectives are tracked through the software. By having them dynamic would be a good reason to type it into the software to calculate your actual outcome. That could be a great bonus for the underdog. Thank you!

Lord Royal wrote:
Yea, except with AoS it didn't work quite well until they brought points back.

Many would argue that Fantasy has never worked well with points, and that narrative play is superior because it has no such expectation.

 JohnHwangDD wrote:

Lord Royal wrote:
Yea, except with AoS it didn't work quite well until they brought points back.

Many would argue that Fantasy has never worked well with points, and that narrative play is superior because it has no such expectation.

I know that point costs are inherently broken but because of factors that can't be displayed through analog components without doing more calculations than actually playing the game. A classical analog wargame with pointcosts for matchmaking have static values that display the efficiency of a unit. The problem with static values ist however that they can't react to situations where the model cannot interact to its full potential. Take a Lascannon for example (or any other one-shot-high-damage weapon): In a game against a few enemies with high health values, that weapon is almost overpowered because it has the right answer to most combat problems. While in a game against a swarm of goblins, that weapon is entirely useless.
If you have dynamic costs that are calculated on the fly however your system can react to those mentioned issues, by adjust the model's efficiency in relation to the opponents' forces. That way even synergies could be calculated just by adding another layer of calculation.
At least theoretically that might be possible with a software. It's a hell lotta work though. But I have a great team of skilled coders.
So I think it's worth a shot. Maybe in a few months I'm smarter and call it a fail, but maybe (less proparble but maybe) it's something new worth playing.
I'm pretty aware of the fact that I try to find some kind of holy grail and proparbly will fail... so what... I've nothing to lose but a lotta time doing things I like.

Point costs are never right, because they can't capture all factors, and player skill is another one. Trying to get them "perfect" is kind of a fool's errand, and you start getting too complicated as you add extra factors.

Honestly, it's a question of getting "close enough", and having enough randomness in the game that any small costing error gets washed out.

It's the final goal to get close enough. I don't start with it, of course. But I keep that goal in mind and try not to make decisions that would make it impossible. It's a thin line between "balancing friendly" simplicity and engaging outcomes.
For now the combat works without a defence value to start simple. The formula seem to work in those boundaries. The defence will be added in the next days. Step by step I hope it will grow into a game.

I don't have the illusion that I can calculate a perfectly balanced game. Instead I want to design a fun balanced(ish) game that is fun no matter what models you choose but without getting lost in simplicity or playing games that are won or lost from the start.
And I really ain't no friend of predetermined scenarios like in the beginning of AoS.

But this is still a kind of brainstorming phase: A hell lotta ideas in the sketchbook and baby steps ahead in development. This project will have to grow for a long time.

Next formula iteration is done:

PowerLevel = {[SusR * HP *oMC/eMC * eApR (1 - 0.1 (Def-eStr))] + Mv/MvR} * {[ComR * A * 0.1 (Acc + WM - Cv) * Dmg (1 - 0.1(eDef - Str)) * (1 + 0.2Rng))] + Mv/MvR} + {SqR/Sq} + {SvR/Sv}

Explanation of the non-profile values:

oMC - own model count
eMC - enemy model count
eApR - enemy's average attacks/round
eStr - enemy's average weapon strength
eDef - enemy's average defence
Cv - average cover modifier
MvR - relevance of movement
SusR - relevanve of sustain
ComR - relevance of combat
SqR - relevance of squad activaions
SvR - relevance of survival actions

SusR, ComR, MvR, SqR and SvR are adjustment variables that apply for each model individually

if Dmg >= max enemy sustain
Dmg = max enemy sustain

And I changed the damage system.
It's now weapon's strength against target's defence that generates a random range between down to -50% and up to +50% of the base damage... in simple terms.

Technically speaking it's:
base damage * (1 - 0.1(Def - Str - Random.Range(-1;1))) //for Def and Str being between 1 and 10

Phew, it "starts" getting complex.

Onto the next tests!!!


Automatically Appended Next Post:
Aaaaand it's completely off...

Back to the sketchbook...

Back to the basics I figured out that some of my first balancing suggestions were entirely wrong... ARGH!!!

Thank Tzeentch, I have an appointment with a balancing expert from Wooga (developer for f2p-mobile games; based here in Berlin) in 2 weeks and he's a tabletop gamer.

Sadly I can't find his GDC Talk about balancing formulas. At the end he linked his email adress to help others with their balancing issues...
He almost immediately wrote back... now, the video is down... mhm, coincidence? Oops.

Parametric cost is good because you can recost on the fly.

But be careful with formula for a complex unit - it often gives a false sense of precision and correctness, which can result in a series of systematic errors that can be hard to correct.

No, danger for this one above though.
It just cannot be, that models get cheaper the more models that player has... I simply forgot a few important values.
Right now, it's a puzzle, but I'm learning a lot about my system.
Wow, math can be interesting...