Inner Peace (Through Superior Firepower) – Episode 026

[name lips]

I think the units under her command disbanding has to do with a lack of funds to provide upkeep. Maybe.

And maybe there's a bonus to being a Croakamancer where units in your personal stack don't cost upkeep every turn.

She has no schmuckers to pay for upkeep, so they just automatically disbanded.

Wait maybe she does have schmuckers. She got a green gem in the hat as well, perhaps that's the remainder of the Treasury, what's left over after her promotion to Heir.

So I don't know. Just brainstorming.

I suspect she is about to be captured and Turned to Olive's side, where she will genuinely be a "poison pill" and manipulate that side's destruction. Then somehow she will end up with Faq... and the Prediction that she will wield an Arkentool will be made. Then she will, again, manipulate events so Stanley will attack Faq, on the assumption he will fall and she will get the Arkenhammer. That plan will backfire, and seeing that Faq is being destroyed, she will switch to genuinely betraying Faq instead of pretending she was, and she will abruptly Turn and join GK...

[Balerion]

If I had to guess, Haffaton just reused their tunneling ambush strategy.

You mean they already have a mole?

I was referring to when they hit the siege stack way back at the start of all this. They had tunneled underground and hidden the entrance to do that ambush. I assume that is where they were hiding off turn, move closer if necessary next turn to get in single turn strike distance. (I am also aware you may have gotten that and made a joke i just missed ><)

[raphfrk]

Well, it was concealed under the "bonus stops at 8" thing, which may be interpreted as "stays constant after 8".

It could also be the total bonus stops at whatever you get for 8 other units. If you have a stack with 16, then they will all only get 50% of the effect of the bonus (rounded down maybe).

So, for > 8

bonus(N) = bonus(8)*8/N

[BLANDCorporatio]

It could also be the total bonus stops at whatever you get for 8 other units.

Yet another way that was attempted to ensure the 8-stack is optimal is indeed to say "only the first 8 units actually contribute to the fight; the rest do not influence the outcome" (except maybe to become casualties if the first 8 are killed).

If you have a stack with 16, then they will all only get 50% of the effect of the bonus (rounded down maybe).

So, for > 8

bonus(N) = bonus(8)*8/N

If only the top-8 units mattered, then yeah the 8 extra units are just dead weight.

What you suggest is something a bit different, which is for the bonus to decrease by a factor of 1/N for N above 8. This was the first fix that Erfsims tried actually. It turned out to not be powerful enough*, for various reasons, which is why sharper decreases are used.

EDIT: *: obviously, if all you want is to maximize the amount of damage you get from stack bonus, stacking in formations of 8, under the quoted rules, is the way to go. It would be disastrous to stack 16 units, for example. Stacking 10 or so however became routine, because of the interplay between retaliation rules (which favour larger stacks). See, depending on the kinds of units you mixed, or how many you had available, it was sometimes the case that stacking more than 8 would produce a punchier stack, despite the stack bonus decreasing after 8, because any one unit would contribute more raw damage than was lost by decreasing the stack bonus.

[DevilDan]

Puttin' on the Ritz!?

Rob's genius continues to blow me away.

[BLANDCorporatio]

irt. raphfrk:

No, wait, I spoke too soon. The first attempt from Erfsims was a linear decrease. Your system is different, and better. Under linear decrease, it was possible to produce stacks with more damage, even as the stack bonus decreased. Under your system,

if I have N units all with the same attack x, the damage after stack bonus is Nx*bonus(8)*8/N = 8x*bonus(8) ie the exact same damage as if the stack had 8 units. So the extra unit adds nothing, unless you wanted an extra meat shield for a squishy warlord that you have at the end of the stack, or if you wanted the stack to survive at all because retaliation rules say that the opponent is fatigued only if your stack survives etc.

Your system may be enough of a pressure to avoid exceeding 8 units per stack, most of the times, but the thinking went, let's make that pressure even larger. (Hence 1/N^2).

[Sir_Dr_D]

We knew this was going to happen. We knew that Goodminton would be destroyed, and those characters croaked. But it was still shocking and emotionally sad when it does happen. Probably because it happened so suddenly and at a time we weren't expecting it. Good update Rob.

[CNagy]

I always saw the stack bonus as very simply linear, and due to that obviousness and simplicity the 8 stack was the default way to sort units.

Imagine basic units with a 1 attack. Say the bonuses increases by 1 for every unit in the stack, capping at 8 units (+7 bonus). If you've got 16 such basic units, your formations could be 16 units with 16 attack (+7 bonus), or 2 x 8 units with 8 attack (+7 bonus), with a final calculation of 23 for the 16-stack vs 2x15 for the 2x8-stack.

Possible problems with this approach include the specifics of combat, but we don't know if Erfworlders know the specifics of combat resolution, so even if the 8-stack is not the most efficient, it still appears that way to them because one of the basic components of combat prowess (the stack bonus) stops increasing at 8 units, and because two groups with max stack bonus are more effective than one group consisting of all the units. If combat consists of resolving battle between opposing stacks (or all opposing units in a given hex), and within that battle between specific opposing units, then there may very well be a different optimum configuration depending on the units involved.

This is, of course, just my opinion, based on the idea that the 8-stack bonus was very easy to understand by a population that considers Mathematics to be a branch of magic. System-wise, I would see the benefit of comparing stack statistics to set up the conditions for the resolution, with those conditions carrying over into individual units in the stack for their own combat resolution. In the end, the acknowledgement of Luck as a force means that the 8-stack could masquerade as the most efficient formation even if it is not, technically speaking, because anomalies that might expose it are attributed to chance.

[Gorky]

Wow that was amazing. This really sets up Wanda's motivations well for the rest of the series.

Also, I'm pretty sure Clay boosted her and cursed the rest of Goodminton in the process. The power of Love.

[BLANDCorporatio]

I always saw the stack bonus as very simply linear, and due to that obviousness and simplicity the 8 stack was the default way to sort units.

Linear, as in how? Linear increase, then hold as the number of units goes above 8? Linear decrease, as the number of units goes above 8?

Imagine basic units with a 1 attack. Say the bonuses increases by 1 for every unit in the stack, capping at 8 units (+7 bonus). If you've got 16 such basic units, your formations could be 16 units with 16 attack (+7 bonus), or 2 x 8 units with 8 attack (+7 bonus), with a final calculation of 23 for the 16-stack vs 2x15 for the 2x8-stack.

Oooh, this is an additive bonus. (A stack of 8 units, each with 100 damage, would also have a bonus of +7).

We never even considered this in Erfsims, because in-canon evidence suggests bonuses act as multipliers.

But let's roll with this for a while. Say there's a bridge, and there are Twolls guarding it. You want them out of the way, and you have 24 Pikers.

Now, suppose that you know that 8 Pikers will not be enough to kill the Twolls, but sending 3 stacks of 8 (each with +7 bonus) is enough to kill the Twolls; and all three stacks will have to fight at least once.

Suppose also that sending all 24 pikers at once could also clear the Twolls. Then the best bet is to send all pikers at once. We know units retaliate when attacked, so attacking the twolls three times means they get three chances to kill your units. If the Twolls, at the start, are strong enough to kill 10 Pikers, then they'll croak 10 of your 24 pikers if you send them en-masse. Or, they'll croak 8, then more than 2 (there's plenty of Twolls left, you need to attack them 3 times, remember?), then maybe even one more. So you have more casualties.

[Whispri]

I bet those Boopers kicked over her Snowman too!

Something else we've learned for certain - you can't just throw a few thousand schmuckers at any given Commander and have them ride off into the sunset to survive as Barbarians. Heirs and Rulers only.

II suspect she is about to be captured and Turned to Olive's side, where she will genuinely be a "poison pill" and manipulate that side's destruction. Then somehow she will end up with Faq... and the Prediction that she will wield an Arkentool will be made. Then she will, again, manipulate events so Stanley will attack Faq, on the assumption he will fall and she will get the Arkenhammer. That plan will backfire, and seeing that Faq is being destroyed, she will switch to genuinely betraying Faq instead of pretending she was, and she will abruptly Turn and join GK...

Speaking of Faq, this 'sudden death from nowhere' thing would have been a lot easier to pull off with the help of a powerful Predictamancer and a powerful Foolamancer. Never mind the powerful fliers.

[BLANDCorporatio]

Speaking of Faq, this 'sudden death from nowhere' thing would have been a lot easier to pull off with the help of a powerful Predictamancer and a powerful Foolamancer. Never mind the powerful fliers.

Are you pointing names and naming fingers? D'you think Jack had a hand in that?

[technojunkie]

Speaking of Faq, this 'sudden death from nowhere' thing would have been a lot easier to pull off with the help of a powerful Predictamancer and a powerful Foolamancer. Never mind the powerful fliers.

My thoughts exactly... Who's to say that FAQ isn't intimately involved in the destruction of Goodminton? I think that would provide a great deal of incentive not to care about FAQ when Stanley eventually shows up.

[Vorteks]

Powerful story, I'm really enjoying it.

So... "end of part 1", eh? What happens now? Does part 2 of the novel start immediately? Will there be a break? Or is it back to our regularly scheduled book 2 comic?

[MonteCristo]

II suspect she is about to be captured and Turned to Olive's side, where she will genuinely be a "poison pill" and manipulate that side's destruction. Then somehow she will end up with Faq... and the Prediction that she will wield an Arkentool will be made. Then she will, again, manipulate events so Stanley will attack Faq, on the assumption he will fall and she will get the Arkenhammer. That plan will backfire, and seeing that Faq is being destroyed, she will switch to genuinely betraying Faq instead of pretending she was, and she will abruptly Turn and join GK...

Speaking of Faq, this 'sudden death from nowhere' thing would have been a lot easier to pull off with the help of a powerful Predictamancer and a powerful Foolamancer. Never mind the powerful fliers.

Not likely. First off, Jack needs to actually be present in the field to work that kind of magic veiling, and he was important to Faq's defense; as such it is unlikely that banhammer would ever send him out into the field with the mercenaries. Second, we know Jack was in the magic kingdom twelve turns ago; this would mean he was in faq just 12 turns ago and thus would have needed to get from Faq to goodmitten in just those 12 turns to join the battle. Third, Given the size of haffaton and the lacking resources of goodmitten, it seems unlikely that haffaton would even feel the need to hire mercenaries; haffaton throughout this whole story seems like they were very much in control of the situation.

[Patso Redneb]

Suppose also that sending all 24 pikers at once could also clear the Twolls. Then the best bet is to send all pikers at once. We know units retaliate when attacked, so attacking the twolls three times means they get three chances to kill your units. If the Twolls, at the start, are strong enough to kill 10 Pikers, then they'll croak 10 of your 24 pikers if you send them en-masse. Or, they'll croak 8, then more than 2 (there's plenty of Twolls left, you need to attack them 3 times, remember?), then maybe even one more. So you have more casualties.

I'm a newcomer to Erfworld and these forums (hello everybody btw) so I probably don't know many obvious things, but is there any evidence in the story, that attacked stacks retaliate to every attacking stack. I would rather guess that every stack can attack only one other stack at the same time, be it attack, retaliation or whatever. In that case many small stacks would be preferred to one big stack. And there could be additional bonuses for flanking.

[Housellama]

Suppose also that sending all 24 pikers at once could also clear the Twolls. Then the best bet is to send all pikers at once. We know units retaliate when attacked, so attacking the twolls three times means they get three chances to kill your units. If the Twolls, at the start, are strong enough to kill 10 Pikers, then they'll croak 10 of your 24 pikers if you send them en-masse. Or, they'll croak 8, then more than 2 (there's plenty of Twolls left, you need to attack them 3 times, remember?), then maybe even one more. So you have more casualties.

I'm a newcomer to Erfworld and these forums (hello everybody btw) so I probably don't know many obvious things, but is there any evidence in the story, that attacked stacks retaliate to every attacking stack. I would rather guess that every stack can attack only one other stack at the same time, be it attack, retaliation or whatever. In that case many small stacks would be preferred to one big stack. And there could be additional bonuses for flanking.

That would make a lot more sense, actually. There's nothing that indicates that one stack can retaliate against multiple units.

There's also nothing that indicates that more than 8 units in a stack get to attack at the same time. In Sword and the Stars, you can have fleets that are bigger than the number of units your Command and Control ships can handle. What happens is that the ones that you don't have C&C points for don't show up on the battlefield. If an active unit dies, then another unit from the fleet takes its place, but you never have more ships on the field than you can control. I've always seen stacks like that. Sure, you can have a big stack, but the other guys back there are going to be useless.

To refer to Bland's Twoll vs Piker example, let's examine two possible scenarios if we assume that only 8 units can attack at a time from any given stack. Either A) the Twolls can retaliate in full to all units it fights, or B) the Twolls can only retaliate to one unit that attacks it. In case A) the best thing to do is to lead a single big stack and just let the reinforcements cycle in until the Twolls are dead. That's better than letting all the Twolls take a shot at every stack as it comes in. Sure, you may lose the same number of men in the meat grinder single-stack approach, but it's unlikely because there will always be 8 pikers attacking the twolls. Individually, each stack will be weakened by casualties and thus will be less effective overall.

However, in case B) you would be much better off splitting your pikers into three different stacks because only one stack will take damage, but all three will deal damage. So one stack will be wounded and possibly wiped out, but the rest of the stacks are untouched, making 8 your maximum possible casualties. If you had sent in one big meat-grinder stack like above, you have no idea what your casualties will be. Maybe you'll get them in 8, but probably you won't because you're doing less damage per troop and you're doing it slower.

Now, let's take a hybrid of these ideas. Suppose you can only have 8 troops attacking and you can only retaliate once, but if you are fighting multiple stacks and wipe one out, you can continue to attack another stack. However, you can't engage more than one stack at a time on the defensive. So if you take out one and are still alive, you can turn to the next stack and keep going. Or, let's assume that you can engage multiple stacks simultaneously, but it splits your offense. So if there are 8 twolls being attacked by three stacks of 8 pikers each, then each would have either two or three twolls responding to them.

In either hybrid case, your best bet as the attacker is still to split your troops. You will be doing more damage faster, and you will suffer less casualties. You may lose more than one stack, but you'll have two stacks attacking unopposed for quite a while, or each stack will outnumber the defenders by at least 2:1. Since Erf isn't a game, speed and event timing does matter. Things aren't summed and tallied at the end of the turn, they happen in real time. So having two stacks unopposed or a 2:1 advantage is a huge thing. Either way, the best play is to have multiple stacks.

All of this is assuming that only 8 units at a time can attack from any given stack, and this is also all dealing with stacks without Leadership. However, based on what we've seen so far, I don't think that's an unreasonable assumption to make. Putting more people in a stack would ensure you get your stack bonus for longer, which might matter in certain cases, but otherwise it's probably best to keep each stack at 8 and have multiple stacks. Which is exactly what we've seen in the comic.

[Lamech]

Recall the situation of Jillian: She was captured effectively becoming a GK unit. Then she escaped and once again went back to her old barbarian role. Wanda should be able to do the same. Chevok's gun has been readied.

[Housellama]

Recall the situation of Jillian: She was captured effectively becoming a GK unit. Then she escaped and once again went back to her old barbarian role. Wanda should be able to do the same. Chevok's gun has been readied.

Different situation. Jillian was a prisoner, not a part of GK. She didn't Turn. She was never really a GK unit. Wanda, as far as we know, completely joined GK, putting her under Stanley's command. Maybe she's allied and maybe once you're an Overlord you're an Overlord forever, but it's not the same as Jillian's situation.

[Whispri]

Speaking of Faq, this 'sudden death from nowhere' thing would have been a lot easier to pull off with the help of a powerful Predictamancer and a powerful Foolamancer. Never mind the powerful fliers.

Are you pointing names and naming fingers? D'you think Jack had a hand in that?

Yep, I believe I do. It helped Fate along, Marie's schtick. It struck with complete surprise, Jack's schtick. And it granted a chance to boost their Treasury with the capture of a powerful City. And then there's the Book's title to consider.

Not likely. First off, Jack needs to actually be present in the field to work that kind of magic veiling, and he was important to Faq's defense; as such it is unlikely that banhammer would ever send him out into the field with the mercenaries. Second, we know Jack was in the magic kingdom twelve turns ago; this would mean he was in faq just 12 turns ago and thus would have needed to get from Faq to goodmitten in just those 12 turns to join the battle. Third, Given the size of haffaton and the lacking resources of goodmitten, it seems unlikely that haffaton would even feel the need to hire mercenaries; haffaton throughout this whole story seems like they were very much in control of the situation.

One word: 'Scrolls'. 'Sides Faq being a hidden Side at this moment in time is not a given. Twelve turns is time enough to cover a considerable distance by air, even without, for example, move boosting magics. And it would hardly be the first time Haffaton's used another Side to do Knife work on their behalf, if they were even involved, which is also not a given. The alternative is that enemies Goodminton knew about somehow slipped a large army within range of their Capital without their boosted scouts noticing a thing.