New Erf Rules Overhaul

[Nihila]

Maybe one way to balance Ranged would be to have it impose an attack penalty after the Ranged combat round?

That way, after inflicting initial casualties, an Archery stack would suffer more in melee than an enemy stack.

[LTDave]

We had another very popular series of games here, "Erfworld Empires" (both this series, and TBfX were started by LTDave, who deserves props for his game systems) in which units were all exactly alike, except for speed. Infantry moved 2 hexes per turn, cavalry 3, fliers 4.

Thanks for the props, whatever they are.

I've written it here many times, and I'll write it again: Complexity Kills Forum Games (and RL Games as well).
I'm playing in the Titanic Mandate, but I'm not sure I understand what it is I'm doing. It's still fun.

I still reckon that my Erfworld Empires v4 deserves more attention than it recieved here, but, sniff sniff, I can hack it.

[BLANDCorporatio]

Thanks for the props, whatever they are.

Props, kudos, congratulations, internets, cookies etc. I meant that your games were consistently fun, and wanted to express my admiration for the homebrewing and GMing skills.

Maybe one way to balance Ranged would be to have it impose an attack penalty after the Ranged combat round?

That way, after inflicting initial casualties, an Archery stack would suffer more in melee than an enemy stack.

That's more or less what I'm considering. The big problem we have is that we cannot do battles by ordering units individually to move around, a-la Heroes. If we could, then Ranged, while still useful, wouldn't be overpowered on its own (as anyone who plays Heroes knows. Tank units or some magical support is crucial to make the ranged units truly effective).

I think I'm on the way to cooking something that we can discuss, but I'll have to post it later.

[BLANDCorporatio]

I'll need to leave soon, but let's see if I can get this ball rolling.

1. "I like Heroes of Might and Magic" will be a leit-motif. Don't be afraid to suggest your own. Such TBSs have been around for a long time, they've had some rules kinks sorted out. Good stuff to pattern our brews after.

2. Since combat will be multi-round, drastically reduce Attack Skill cap. No unit is to have a larger attack skill than, say, a third of Hits. (Example, Heroes: level 1 units may have 10H and 5 Attack, but by level 7 you get 250H units with only 50A).

3. As I said, reframe "Attack" stat to mean "Attack Skill". Base damage dealt by a unit will depend on the weapon type it uses, and is also capped at a fraction of Hits.
Consequence: it will be difficult for one unit to kill another in one blow. Since a combat will consist of those units exchanging several blows in the same turn, this is ok here.

4. Ranged weapons will do less damage than non-ranged ones, and revert to weak melee weapons when melee opponents have closed in.

5. A-la Disciples, I'm imagining that a side has its stacks in a hex arranged in a "two-row" fashion: you'd have the more fragile ranged stacks in the back row, with beefier Melee tanks in the front row. Everyone gets to choose which melee stack screens what.

6. I need three derived stats: Shock (a function of Hits and Move), Steadfastness (a function of Hits and Defense), Agility (a function of Defense and Move). "Shock" is a unit's ability to plow through opposition, aka, to go through screening stacks. Steadfastness is the ability of a unit to resist Shock (not the magic, the physical effect of having a ton of horse and knight galloping at you). Agility is a unit's ability to maneuver to avoid confrontation.
Explanation: ranged units can take targets from any of the rows in the field. Melee units need to get through the opposing front row however to strike at ranged units. How many combat rounds they need to do this will depend on the above three notions of Shock, Steadfastness and Agility.

Actual detailed operation of the three pending.

[Harry1991]

I'm going to have a look at doing a rules system myself, i think you have good ideas but making the game too complex like LTdave said makes it less enjoyable and causes people not to play.

LtDave, would you mind if use some of your rules as source material? it looked fairly good to me. (will give full credit)

[WaterMonkey314]

BLAND, how are you expecting to calculate results for such a complicated combat system? The Arkentool War / TBfMI systems already got to be exceedingly complicated (it took me on average 10 minutes to calculate each battle and make sure I had no errors; I think it took me over an hour to run the fall of Cameria).

If, as I suspect, you're planning to use some sort of computer support, you'll also have to come up with an efficient interface, or else you'll spend just as much time hacking players' orders to fit into the computer.

[BLANDCorporatio]

BLAND, how are you expecting to calculate results for such a complicated combat system? The Arkentool War / TBfMI systems already got to be exceedingly complicated (it took me on average 10 minutes to calculate each battle and make sure I had no errors; I think it took me over an hour to run the fall of Cameria).

Running battles wasn't at all hard in TBfMI actually, and I speak as the GM of that game. The last few turns had some horrenduous feats of logistics and combat, but if anything, it was mount exploits that made it difficult to track where things were. The actual battles were very easy with the Unit abilities spreadsheet and the Combat spreadsheet as they were.

What was horribly time consuming were the scouting orders.

If, as I suspect, you're planning to use some sort of computer support, you'll also have to come up with an efficient interface, or else you'll spend just as much time hacking players' orders to fit into the computer.

Indeed, I plan to ditch spreadsheets for something more friendly to quick cross-referencing. Once I have a good idea of what the rule system will be, I'll build my assistant around that and provide some screenshots. I plan to do so within a week.

=====

On the issue of "this system is too complicated": this system doesn't exist yet.

And I'd say TBfMI wasn't a complex system either. The combat system was almost as old as TBfGB, the very first or second iteration of the rules. Most rules were the same. I disregarded any screening rules or such in that case. The one thing that was different was the unit cost formula, which was basically (H^2)*(A + 8/(8 - D) + M) + Specials. As complexity goes that is peanuts.

Anyway, back to trying to cook up a combat system.

[WaterMonkey314]

BLAND, how are you expecting to calculate results for such a complicated combat system? The Arkentool War / TBfMI systems already got to be exceedingly complicated (it took me on average 10 minutes to calculate each battle and make sure I had no errors; I think it took me over an hour to run the fall of Cameria).

Running battles wasn't at all hard in TBfMI actually, and I speak as the GM of that game. The last few turns had some horrenduous feats of logistics and combat, but if anything, it was mount exploits that made it difficult to track where things were. The actual battles were very easy with the Unit abilities spreadsheet and the Combat spreadsheet as they were.

What was horribly time consuming were the scouting orders.

Oh God. I'd completely blocked that out of my memory... Even the auto-scout system was kinda finicky.

If, as I suspect, you're planning to use some sort of computer support, you'll also have to come up with an efficient interface, or else you'll spend just as much time hacking players' orders to fit into the computer.

Indeed, I plan to ditch spreadsheets for something more friendly to quick cross-referencing. Once I have a good idea of what the rule system will be, I'll build my assistant around that and provide some screenshots. I plan to do so within a week.

=====

On the issue of "this system is too complicated": this system doesn't exist yet.

And I'd say TBfMI wasn't a complex system either. The combat system was almost as old as TBfGB, the very first or second iteration of the rules. Most rules were the same. I disregarded any screening rules or such in that case. The one thing that was different was the unit cost formula, which was basically (H^2)*(A + 8/(8 - D) + M) + Specials. As complexity goes that is peanuts.

Anyway, back to trying to cook up a combat system.

It wasn't so much the specials and what not that took up time for me; it was confirming that 1) the units ordered existed; 2) the units could execute orders as given; 3) looking up unit stats; 4) (a biggie) plugging in unit stats into the spreadsheet (this got problematic when there were mixed stacks of different units with different levels, some possibly wounded); and 5) resolving the end result. I tried to build my own assistant to keep track, but it got overly complex (I could represent units fairly easily, but I couldn't figure out a way to sensibly split the workload of resolving combat between the computer and myself.)

If you can get an assistant that 1) aids in scouting / auto-scouting and 2) helps with unit movement and combat, I could probably start TAW back up again. I haven't simply because I don't have that much time anymore.

[BLANDCorporatio]

Right, so I've been mulling a bit on ways to allow unit movement during combat, while avoiding to ask players every time about which stack should move where and attack what.

Unit movement during combat is what allows a very chess-like combat system for things like Heroes. But just like chess, it requires players to consider every move, and that will really drag things down in a play-by-post setting.

So one way is to replace that with automatic movement/atacking rules. The problem is, under a system that looks like these Erfsim rules, that will be very byzantine indeed. Therefore I'm thinking of something else.

Won't describe the exact system I have in mind (but I do have one), I'll just list a few key properties and ask is this acceptable?:

1. Stacks will have at most 8 units; mounting is the only way to increase that, but not by much.
2. Those units will usually be very different from one another. (You may replace "unit" with squad if you like. The stack won't contain an Archer and a Piker, say, but a "squad" of Archers and a "squad" of Pikers, maybe one Dwagon as well etc, if it makes you feel that's more Erfy, but the squad is indivisible and is de facto a unit)
3. Any battle will be between an Attacking and a Defending stack.
4. A player will, when sending a stack out to battle, provide orders like "Maximize damage dealt", "Minimize damage received", "Kill that particular unit then {retreat, or max damage dealt, or min damage received}". I'll ask you to trust that
....a. the system is such that it's easy to write a computer program that accurately optimizes as asked
....b. stack construction, and picking which stack to fight which other, and under what orders, actually makes a difference.

[LTDave]

LtDave, would you mind if use some of your rules as source material? it looked fairly good to me. (will give full credit)

Sure, go for it. Just make sure you credit Ronaldo as well.

[WaterMonkey314]

BLAND, I just realized this (possibly related) issue: we still haven't really settled what the proportion of general infantry to heavies / warlords / special units is.

If you look at TBfGK1 and 2, you see pretty much 6-7 grunts to each commander. By the time you get to BfMI, you have crazy Generican mega-stacks of 40+ archers with 1 commander. And in the last chapter of TAW, with the multi-engage rule, you have fairly huge columns with 50-60 units per each commander.

In the latter cases, where there's a lot more general infantry, making stacks homogeneous makes sense and wouldn't cause too much trouble. In the former cases, I think things might get rather more problematic - would you still be able to, say, protect a Diwi by stacking it behind other units?

[BLANDCorporatio]

That's one reason why a hard limit on stack size is appealing.

But before I try to

Spoiler: show

copy the Disciples system

in an Erfworld setting, I have yet another idea to float. Hear this out.

Units have 4 stats: Hits and Move are just as before in the Erfsims, TBf<> etc. Attack and Defense act as combat skills and provide a bonus to base damage: base damage receives a bonus (A/D). So for example, if an attacker unit has Attack 10 and defender has Defense 5, then the Attacker's base damage is doubled.

(Obviously, defense needs a lower cap of 1.)

Base damage is a result of weapon choice. A unit must choose one and only one weapon to use. Various weapons would have different strengths and weaknesses.

So far, this isn't too different from the old TBf<> rules actually, if you think about it, except the separation between Attack stat and base damage.

Ok, so the formula for the Combat bonus is Attacker's Attack/Defender's Defense. This will always be the case, with one exception. First, the normal cases:

Melee opponents whack each other instantaneously. Bonus is A/D.

Ranged opponents whack each other instantaneously. Bonus is A/D.

Now the exception. Melee units vs. Ranged. At the start of a battle, the Ranged units will fire on the melee units. The bonus to their damage is A/(D+3*M) where M is the move of the melee units (or rather, the move of the slowest unit in the stack).

Next, imagine that there are 10 hexes between the melee units and the ranged ones. The melee units need to cross these 10 hexes to get to the ranged units, and they do so with a speed of (M - Mr/2) where Mr/2 is the move of the slowest ranged unit. When the melee units close in on the ranged units, the battle becomes a regular melee: the stacks damage each other all at once, the bonus is A/D. But, the ranged units fight as if their attack was one half of what it was previously. Example: if an Archer has Attack 6 while shooting arrows, it will drop to 3 when he fights melee.

An example combat: one Infantry stack with move 5, and an archer stack with move 5; both stacks have Attack 6 and defense 0. Combat would go like this: first, Archers fire on the infantry with a bonus of 6/(15). Next, infantry moves 3 hexes (5 - 0.5*5); they are now 7 hexes away from the archers. The Archers fire again, the Infantry moves again, and is 4 hexes away now. Repeat, and the infantry is one hex away. Repeat, and the infantry has reached the archers, and melee ensues until one stack is wiped out.

What can infantry do to be more resilient to Archers? Two things: bump up Move, or bump up Defense. If they bump up defense, they decrease the Archers' damage bonus somewhat If they bump up move, they increase their survivability against ranged because not only do they reduce the damage bonus from the Archers, they also close in faster on them. OTOH, bumping up move will not help those melee units when they fight other melee units.

So at least three unit archetypes naturally emerge: Ranged (High Attack, Fast, Frail), Cavalry (Melee, Fast, Frail), Infantry (Melee, High Defense) with a rock-paper-scissors relationship between them.

Slow Archers are also probably viable against Fast Archers, since in Ranged vs. Ranged, move is irelevant. OTOH slow Archers would have a tough time even against infantry.

Another observation is that as units can spend more and more on Move, Archers don't get much of an advantage. In my example, I had Move 5 Inf against Move 5 Archers. Infantry needed 4 rounds to close in. If I had Move 10 Inf against Move 10 Arch, infantry would have needed 2 rounds to close in. At Move 20 for both Archers and Infantry, the Archers get only one free shot before the Infantry closes in.

This means that, if you have to fight Cavalry (fast Infantry) with Archers, you're better off doing so with units that have a very high Attack, rather than move. Fighting slower infantry though is where Fast Archers would really outshine Slow ones.

[BLANDCorporatio]

PS: what if I have a stack of Infantry with Move 1, attacking a stack of Archers with move 20?

Setting aside the fact that it's a very one-sided battle, the speed that the infantry closes in on the archers is 1. I lower-capped closing in speed at 1, because I can

PPS: so this way I can have combat be between just two stacks, Attacker and Defender. No "stack screens another" byzantine bullshit. Combat will be multi-round, but by this time we're all agreed that managing things like large scale Erfsims needs software support anyway. And the rounds

a) are easy to compute, as we're basically using TBf<> rules for them
b) it's easy to tell how many there will be, at most. Either the Infantry is killed trying to close in on the Archers, or they close in in 10/(relative speed) rounds. And that will not change during battle. (Even if the slowest unit that caused the move to be as it was dies)

[WaterMonkey314]

Hm. I like the idea quite a lot.

One (semi) nitpick - let's not say that the melee guys are X "hexes" away from the ranged units. That's bound to cause confusion (see City Zones for issues with understanding sub-hexes nicely). Unless you're planning to introduce obstacles within hexes, all you really need is a generic distance unit - say, "clicks". Or "furlongs". Or anything other than "hex", really.

[BLANDCorporatio]

We can dispense with furlongs even. Just imagine there's some distance, no matter how large, between the ranged and the infantry stack, and the infantry is trying to close in on the ranged units while under fire. This means the ranged units get to fire "several times", which effectively is a bonus to their attack score.

Assumptions:

units spend stat/ability points to improve on the template. Improving Attack by 1 costs 1 point, improving Defense by 1 costs 1 point, improving Move by 2 costs 1 point. (Erfsim rules, in other words.)

After some napkin calcs, I suggest these formulas for the Combat Bonus:

A/D for melee vs melee, or ranged vs ranged.

For the bonus of ranged units when damaging melee ones, use:

[A/(D + M)]*[2/max{1; [8*M - 3*Mranged]/max(M; Mranged)}]

What's A, D, M, Mranged?

Spoiler: show

(Attacker's average A)/(Defender's average D) when melee stacks fight melee stacks; or ranged stacks fight ranged stacks.

When computing the damage a ranged stack does to a melee one use this:

((Attacker's average A)/([Defender's average D] + [defender's slowest M]))*[Maneuvering factor]

Maneuvering Factor = 2/max{1; [8*(Infantry's slowest M) - 3*(Ranged's slowest M)]/max(Infantry's slowest M; Ranged's slowest M)}

When computing the damage the survivors of the Infantry stack do to the ranged stack, use the regular, A/D formula.

Combat bonus is to be capped above by 3, and below by 0.2.

As far as I can tell, this results in:

-Slow Infantry beats Fast Infantry;
-Slow Ranged beats Fast Ranged;
-Fast Infantry beats any Ranged (especially effective against Fast Ranged);
-Slow Infantry and Slow Ranged are somewhat evenly matched.

I've tried this imagining this: let's say I create units all starting from the same template: n Hits (unimportant how many), 1A, 1D, 1M. All units get the same amount of points to spend.

Fast Units spend half on Attack, half on Move.
Slow Infantry spends half on Attack, half on Defense.
Slow Ranged spends all on Attack.

Then I jiggled some parameters until the inequalities mentioned above emerged.

[WaterMonkey314]

We can dispense with furlongs even. Just imagine there's some distance, no matter how large, between the ranged and the infantry stack, and the infantry is trying to close in on the ranged units while under fire. This means the ranged units get to fire "several times", which effectively is a bonus to their attack score.

There is one difference that could be important - how you deal with the random rolls.

If you roll once and multiply that result by the number of "volleys" the ranged stack gets, you'll end up with a relatively high std. dev. compared to if you were to simulate successive "volleys" by the ranged stack with multiple rolls.

I don't know what you think about this; I personally think the multiple-volley system is better, on account of reducing the chance of super-aberrant results.

[BLANDCorporatio]

Good point.

OTOH I quite like lumping the various volleys into one mobility factor. It simplifies analysis of the system immensely, because, at least in the interval 0.2 -> 3 I get a continuous combat bonus. As such various handy approximations are useable.

My guess then is that, for the purpose of Ranged against Melee, either the random number should be calculated differently, or we try and see how the system goes without applying any correction to the random factor. Since my analysis was based on the random factor being the same in all kinds of engagements, I expect that, even with possible abberations stemming from very good/bad rolls, on average combat will tend to favour unit types in the manner I described previously.

Ok, next step is to write a rules doc. It won't be from scratch (*phew*): I'll take the Erfsim rules and

-slice time to pop/build
----also, adjust the amounts of popable units. The rule is this: if a unit U costs X points to create (starting from a 0/0/0/0 template), and another, V, costs kX points, then in the time it takes a city to pop one V, that same city could pop k Us instead
-increase income from resource points
-decrease income from cities (slightly)
-place a cap on hex upgrades based on controlling city level
-add "weapon"/"base damage"
-change the combat system: attack, defense, and possibly move, will define a Combat Bonus as described above
-special exception to the combat bonus rules (like Ranged in fortification, flying ranged vs ground infantry): infantry cannot close in, and as such ranged will get "a few" free shots until "they run out of ammo".
-combat system: a side may attack a specific stack only once per turn
-combat system: a side may send one of its stacks into battle as many times as it likes during its turn, as long as there are still eligible targets in the stack's reach
----exception: ranged stacks may only fight once as ranged attackers during their turn.
-combat system: leadership bonus will now work like this: a fraction (level-dependent) of the leader's Attack, Defense, and Move stats is added to all of the units in the stack, and it is these improved values that enter the calculation for the combat bonus
-"leader" means unit with "War Leadership"; example: Overlord, Heir, Chief Warlord, Warlord.
-replace unit templates with templates that: start at n Hits, 1A, 1D, 1M. Stat/ability points then will purchase a weapon (every unit must have one and only one) that defines base damage; points may also improve stats and/or purchase other abilities
-special abilities will mostly stay the same except:
----Flight (cost will now depend on Hits and Attack, and will be much higher)
----add Hover ability: moves like Flight, but without blocking ground infantry from reaching it
----add War Leadership, which is like Leadership but also carries a bonus to stack
----adjust Leadership special: it makes the stack led, and as such allows it more complex orders like refusing to engage, targeted fire, capture etc; it doesn't provide bonuses itself, however
----breath weapons will carry slight stat penalties, not just massive bonuses
-improve Shockamancy spells to become superior to generic caster spells
-remove: Mathamancy (it's useless), Predictamancy (same), Retconjuration (it's called GM Fiat, and I get to do whatever I want coz I is GM, don't need any compendium of that), and any other discipline that I feel is worthless at this time
----making a proper spell compendium is a job in itself, to be postponed for later
-item crafting: all casters may produce items of any type; Dollamancers create items at half cost
-juice replenishes fully each turn
-wild units pop more often
-turning/taming will not require either Turnamancy nor Carnymancy, but without magical support, they will take "long"

[BLANDCorporatio]

Right. I do not want to hear any more of "TBfMI is complicated". I've just had a hack into the Caster system (and I don't mean spell attributes). Urgh. All that has gone. I've moved the complexity where it belongs.

Tomorrow I hope to get the last chapters (cities, terrain and hex upgrades) done so that I can post a link to the new rules doc. If it is sometimes looking like a software spec, that's because it is. The next step will be to create the software assistant for GMing. I for one would not dare to GM either version of the Erfsim rules (especially my own) using only spreadsheets and notepad.

[BLANDCorporatio]

Aaaand here is my version of the rules, (very) loosely based on NewErf 0.37.

I'm very proud of the weapon design, you might want to check that and the Specials listing. Despite it containing a few points that will be contentious (like at least one Summoned unit for each casting discipline I kept), I'm reasonably happy with the magic system as well; though almost anything would have been an improvement over the older one (that's a rant in itself, but another time).

If you are feeling really, really brave, check out the combat system. Of course I expect nobody to keep track of all that, except for their own stacks, maybe. But we already know we need a software assistant for things like map drawing with visibility accounted for, and once that is done, the number crunching for combat is easy.

I welcome comments and suggestions for improvement. Meanwhile, I'll get on doing the hex engine, since that's very important and not really very tweak dependent.

[Twoy]

Q: Do units with the simplemancy special have independent leadership like casters? Similar to Archons or do they need to spend points to get leadership by itself.
A: They have to pay for the leadership separately.

Q: For spells such as Hadoken and other similar spells, do they deal the stated damage or the stated damage plus the casters attack.
A: The spell is the unit's action during that phase of combat. They do not get to add their attack stat to the spell damage.

Q: What is the warlord patrol bonus? It is mentioned in the Moneymancy section but nowhere else.
A: The warlord patrol bonus gives extra shmuckers if a warlord spends the whole turn patrolling something. The bonus is plus 10 percent per level. A warlord can patrol a city or an upgrade. The patrol rule was removed because it detracted from the puspose of the game.

Q: For the stacking bonus, does the bonus get added to each unit? Or to the total stack's attack/defense.
A: The stacking bonux is multiplied by the stack's total attack and defense.

Q: Also I believe there is a typo in the level up section. It states that the unit gains 1 HP 1 Attack and 1 Damage. I think damage is supposed to be defense but I may be wrong.
A: It is supposed to be Defense. I knew about this one and still have not corrected it on version 0.38. Maybe I can get it in version .39.

Q: Also as a random question, what do you think a trimancer link with a dittomancer/thinkamancer/turnamancer could reduce the entire side's production. I think at least 2/3rds.
A: I would say at least one-third. Or you could make it for 150 juice it reduces by one-third, or for 300 it reduces production by two-thirds and for 450 it reduces it by four-fifths.