Showing posts with label Domains at War. Show all posts
Showing posts with label Domains at War. Show all posts

Monday, December 21, 2020

ACKS Mercenary Mixes: Dwarves

Context: Previously, I experimented with reducing the types of mercenaries available in each market, while preserving the total wages of mercenaries in that market, in order to make assembling whole units faster and reduce complexity in play.  Today, an application.
 
It always bugged me in ACKS that you could hire dwarven mercenaries (and you should, because they're good), and you can build a dwarven domain, but if you wanted to hire mercenaries in your dwarven domain, they were mostly human.  A reasonable judge would surely fix this, but it would probably be ad hoc.  But now, thanks to the wonders of hacky python, we have answers.

Let's take a ratio of 4 heavy infantry to 2 crossbowdwarves to 1 mounted crossbowdwarf to 1 fury.  It might be fun to do two dwarven mercenary mixes, one civilized, with crossbows and maybe musketeers, and one uncivilized, with furies and...  I dunno, maybe slingers or javelineers?  Disrupt enemy formations before the furies crash into them?  But let's just start with this, under the assumption that furies are uncommon in dwarven society but not that uncommon.

Raw counts





TypeIIIIIIIVVVI
Heavy infantry286.979.2341.0312.626.12.18
Crossbowmen143.4539.6220.526.313.051.09
Mounted crossbowmen71.7219.8110.263.161.520.54
Furies71.7219.8110.263.161.520.54







Squad





TypeIIIIIIIVVVI
Heavy infantry47 per month13 per month6 per month2 per month1 per month3 months
Crossbowmen23 per month6 per month3 per month1 per month2 months6 months
Mounted crossbowmen23 per month6 per month3 per month1 per month2 months6 months
Furies11 per month3 per month1 per month2 months4 months12 months







Platoon





TypeIIIIIIIVVVI
Heavy infantry9 per month2 per month1 per month3 months5 months14 months
Crossbowmen4 per month1 per month2 months5 months10 months28 months
Mounted crossbowmen4 per month1 per month2 months5 months10 months28 months
Furies2 per month2 months3 months10 months20 months56 months







Company





TypeIIIIIIIVVVI
Heavy infantry2 per month2 months3 months10 months20 months56 months
Crossbowmen1 per month4 months6 months20 months40 months111 months
Mounted crossbowmen1 per month4 months6 months20 months40 months111 months
Furies2 months7 months12 months39 months79 months221 months

Bam.

This led me to realize that I didn't compute rounded total monthly costs of operation for dwarf units in Simplified Mercenaries (nor elf units, for that matter).

TypeCompany kgp/moCompanyPlatoonSquad
Dwarven Heavy Infantry2.52500625125
Dwarven Crossbowmen3.253250825175
Dwarven Mounted Crossbowmen3.753750950200
Dwarven Furies2.252250575125

Man, this was an easy post.  I wish I had more trustworthy tooling to compute BRs though - I had an error in computing monomerc cavalry BR because there was a cell in my spreadsheet that I forgot to change, and it has me wondering if the same error has occurred with any of my previous cavalry, particularly the nomads.  Maybe I'll script that up next.  More ambitious would be to build a pipeline going from ACKS stats to some intermediate representation of Battles stats, BR, and wages which can be rendered several ways for human consumption for different purposes, or referenced by market mixer ratio files...

I guess elves would also be a natural next thing to do with my current technology.

Friday, December 18, 2020

Simple Mercenary Mix Experiments: The Battle of Crecy

Last post, I concluded that compressing mercenaries down to just one type was infeasible, but was thinking about going down to 2-4 units by culture, and then making those the only units available in markets of that culture.  If you preserve the total wages of mercenaries available per month, this will mean that more of each of those types will be available every month, so you can build up useful whole units much more quickly than under ACKS' defaults where every market has tiny numbers of every type of mercenary.  The tradeoff is that you lose that variety, so you may not be able to build certain unit compositions depending on where you're recruiting.
 
The English and French seemed like a fine place to start.  My aim here is less to build a super-high fidelity model of French and English army compositions and more to experiment with techniques for building these sort of cut-down unit mixes and figuring out their availability.

I don't know if the army compositions at the Battle of Crecy were representative but they seemed like a fine starting point.  I like that there's some vagueness to them.

The English had a ratio of somewhere around 2 men-at-arms (heavy cavalry, fighting dismounted in this case - although notably chain barding was not used, which would lower the unit AC to 5) to 4 longbowmen (longbowmen A) to 3 hobelar light cavalrymen (light cavalry C) to 3 spearmen.  Not clear from wikipedia how the spearmen were armed or armored; I'm inclined to go with light infantry B or E, though I did see some things while googling that suggested the Anglo-Gascon armies had Irish kerns with javelins.  I think Light Infantry E has the encumbrance to carry 3 javelins but that would increase their BR and wages up to 9gp/mo, so let's not do that.  I like that Light Infantry E is Formed even though it's lightly-armored; it makes sense in an army where so much of the infantry is longbowmen that you'd want the rest to be Formed, for some backbone.

The total wages of one of these "to ratio" blocks of 2 heavy cav, 4 longbowmen, 3 light cav, and 3 spearmen is a nice round 300gp/mo.  120gp/mo in heavy cav, 72 gp/mo in longbowmen, 90 gp/mo in light cav, and 18gp/mo in spearmen.  That's cavalry-heavier than ACKS' baseline assumptions, where about half of expected wages are cavalry (here it's more like 2/3).  It probably makes sense to make spearmen more available than these ratios would indicate, and they're so cheap that it barely changes anything to do so.  It also makes sense that the poor melee infantry would have suffered the greatest attrition and desertion during the campaign leading up to Crecy.  Finally, if the men-at-arms are AC5 instead of 6 due to lack of barding, that drops their wages down closer to 50gp, and we can get back to 300 by adding about three spearmen.  It's probably even more reasonable to go up to something like 8-12 spearmen, which would then give each of the four classes closer-to-equal shares of the wages pie (100gp for men-at-arms, 90 for the hobilars, 72 for the longbowmen, 72 for 12 spearmen)

So let's take 2 mounted men-at-arms, 3 hobilars, 4 longbowmen, and 12 spearmen as our ratio.  Total wages for one of these "blocks" is 334gp/mo.
 
We should expect to be able to hire about a third of one "block" of 2 knights, 3 hobilars, 4 longbowmen, and 12 spearmen per month in class VI, just shy of a whole block per month in class V, just shy of two whole blocks per month in class IV, six blocks per month in III, 12 per month in class II, and like 40 per month in class I.

If you work this all out, you end up with these raw counts of mercenaries available per month:

TypeIIIIIIIVVVI
Men-at-arms77.3121.3511.063.41.640.59
Hobilars115.9632.0316.595.12.470.88
Longbowmen154.6242.722.116.83.291.17
Spearmen463.85128.166.3420.419.863.52

I should really figure out how to translate these into dice, but picking rolls that have the right expected value is sort of annoying, and these go higher than any of the mercs in stock ACKS, so I can't just crib those.
 
If you turn those raw numbers into squads, platoons, and companies like in Simplified Mercenaries, you get these:

Squads:
TypeIIIIIIIVVVI
Men-at-arms25 per month7 per month3 per month1 per month2 months6 months
Hobilars38 per month10 per month5 per month1 per month2 months4 months
Longbowmen25 per month7 per month3 per month1 per month2 months6 months
Spearmen77 per month21 per month11 per month3 per month1 per month2 months

Platoons:
TypeIIIIIIIVVVI
Men-at-arms5 per month1 per month2 months5 months10 months26 months
Hobilars7 per month2 per month1 per month3 months7 months18 months
Longbowmen5 per month1 per month2 months5 months10 months26 months
Spearmen15 per month4 per month2 per month2 months4 months9 months

Companies:
TypeIIIIIIIVVVI
Men-at-arms1 per month3 months6 months18 months37 months103 months
Hobilars1 per month2 months4 months12 months25 months69 months
Longbowmen1 per month3 months6 months18 months37 months103 months
Spearmen3 per month1 per month2 months6 months13 months35 months

I think these results are OK.  Whole units are available about two to three times as quickly as they were under Simplified Mercenaries, and you also have fewer unit types to manage.  It also adds an additional reason to travel for mercenaries; rather than it just being about hitting big markets to get large volumes of troops, you might want to travel to get access to a different mix of troops.

Speaking of which, the French.

Again, estimates are sketchy, but let's work with 8000 men-at-arms, 6000 crossbowmen, and 14000 spearmen as our estimate of the army at Crecy.  This works out to 4 men-at-arms, 3 crossbowmen, and 7 spearmen as our "block", with total wages of 296 gp/mo, the vast bulk of which is the cavalry.  Let's bump spearmen up again to 12 per block.

There's only one type of crossbowman in Domains at War, and unfortunately he doesn't have a pavise.  Oh well.  Let's stick with the AC5, 50gp/mo men-at-arms and light infantry E for the spears (same as the English).

Raw counts:
TypeIIIIIIIVVVI
Men-at-arms158.4143.7522.666.973.371.2
Crossbowmen118.8132.8116.995.232.530.9
Spearmen475.23131.2567.9720.9110.13.61

Every french village has a knight!  I like it!  It's a little silly that you can find crossbowmen out in the sticks (they were actually Genoese) but I'm not going to worry about rural vs urban units yet.  Or maybe ever.

Squads:
TypeIIIIIIIVVVI
Men-at-arms52 per month14 per month7 per month2 per month1 per month3 months
Crossbowmen19 per month5 per month2 per month2 months3 months7 months
Spearmen79 per month21 per month11 per month3 per month1 per month2 months

Platoons:
TypeIIIIIIIVVVI
Men-at-arms10 per month2 per month1 per month3 months5 months13 months
Crossbowmen3 per month1 per month2 months6 months12 months34 months
Spearmen15 per month4 per month2 per month2 months3 months9 months

Companies:
TypeIIIIIIIVVVI
Men-at-arms2 per month2 months3 months9 months18 months50 months
Crossbowmen2 months4 months8 months23 months48 months134 months
Spearmen3 per month1 per month2 months6 months12 months34 months

So the French can raise knights about twice as quickly as the English, but have no light cavalry and much weaker archery.

So now that I have tech in place to make generating market availability for subsets like this easy, the question seems mostly like "how do you pick reasonable ratios"?  Both of these army compositions definitely deviate from the guidelines in Domains at War: Campaigns that you should really only have 10 heavy cavalry per 120 recruits, but I think I'm OK with that.  This isn't "we raised all the fighting-age men and trained them"; this is "among professional fighting men of a given culture, what fraction have what fighting styles and equipment?"  So if you weren't going to qualify as heavy cavalry in Late Medieval France, you were probably just not going to be a professional soldier.

Monday, November 9, 2020

Reading OD&D: Chainmail

Delta's latest post got me interested in OD&D, particularly his mention of a movement point system.  The whole "miles per day with multipliers that don't work out to even numbers of hexes" has been a thorn in my side before, so I'm really curious to see what came before it.
 
Reading Chainmail first.  As usual this is a collection of notes on things that surprised me while reading.
 
Part 1: Mass Combat
 
One figure represents 20 guys, and one inch of table is 10 yards.  Looking at the pictures before the introduction, it looks like groups of 20-30ish figures were the norm, or about 400-600 guys per large block of troops.  So the whole battle was probably around battalion scale, in Domains at War terms.
 
I love that secret orders with simultaneous movement is an option - Domains at War with secret orders and simultaneous movement would be interesting. 
 
Movement rates are already of the form 6/9/12 for armored, heavy, and light foot.
 
Tracking fatigue for individual units, oh dear.
It is interesting that fatigue kicks in after 4-5 turns of action, after which you need a rest turn, and fatigue makes units attack, defend and roll morale one step worse.  Which sounds suspiciously like dungeon fatigue, where you get 5 turns of action followed by a rest or you take -1 to hit and damage.

Using a painted dowel to resolve cannon fire.
 
Morale is big, obviously.
 
No firing into melee.
 
Crossbowmen can't volley overhead.
Landsknechte seem to get a bunch of bonuses?
Knights have to roll obedience or charge any enemy within charging range (preferably enemy knights).
Mercenaries can demand more pay or defect mid-battle.
Peasants have to roll before attempting basically anything.
Religious order knights never surrender.
English longbowmen can plant stakes to ward off cavalry.
Mongols recover from retreats (but not routs) automatically.
Polish troops have elite morale.
 
If enemy units get near your table-edge they can loot your camp.
 
I like the weather state-machine, and the influence of heat on fatigue.
 
Part 2: Man-to-Man Combat (shifting the scale from 20 guys per figure down to 1 guy per figure):
 
I don't see a ground scale for man-to-man combat (where each figure represents just one guy) mentioned.  If you have 20 guys per 30' square, with one guy you'd expect 30' * sqrt(1/20) or about 6.5 feet per inch.  So looks closer to a 5 foot square than a 10 foot square, and engaged in melee is within 3 inches or about 15-20 feet.
 
Crazy rules about how the relation between the reach (rank) of the attacker and defender's weapons determines who strikes first and how many attacks they get.  Between this and the weapon-vs-armor table, picking weapons seems like it would matter a lot.
 
Parrying!  Forfeit your attack to give someone attacking you a penalty to theirs.  Naturally weapon reach/speed plays into whether and how effectively you can parry. 

Berserkers appear here; I guess you just can't get them in bulk.
 
Jousting.

Part 3: The Fantasy Supplement
 
Dwarves and Gnomes are equivalent, as are Goblins and Kobolds.  Even here dwarves have darkvision (and this is the origin of the "giants have a hard time hitting dwarves" too), but elves don't.  Elves can all turn invisible and have magic swords that wreck orcs and goblins, but use different mechanics from human magic swords.  Hobbits "have small place in the wargame", can blend into terrain, and can throw stones as well as an archer shoots / to the same range which is wild.
 
Orcs are "nothing more than over-grown Goblins" and have to pass an obedience roll or start fighting other orc warbands within charge range, even if they're on the same side, unless already engaged in melee.
 
Heroes fight as four normal men (hence 4th level fighter's level title and the breakpoints for to-hit in Basic, presumably) and super-heroes fight as 8 normal men.  They can also just...  shoot down flying dragons (with a low chance, but it's all-or-nothing for each individual shot).  Rangers are slightly better heroes - this might be the origin of ranger as a weird special class that you had to become rather than something you could start as.
 
Various levels of wizard; they have different penalties to casting depending on how strong they are, but all fight as two normal men and are impervious to normal missiles.  Fireball and lightning bolt are both save-or-die (but you have to pre-declare a range, like artillery fire in Warhammer 40k, so they're easy to miss with) and dragons don't give a fuck about either.  Counterspelling is a totally normal wizard thing to do.  Phantasmal Forces creates an illusory unit.  Haste doesn't give extra attacks, just more movement speed to a unit (likewise Slow just lowers movement speed).  Confusion interacts with secret orders.  Cloudkill drifts at random if not concentrated on.  "In order to cast and maintain any spell, a Wizard must be both stationary and undisturbed by attack upon his person."  The Number of Spells table says "The power of the magic user determines the number of spells he is able to manage:"  This is ambiguous - number of spells known, or number of simultaneous spells he can concentrate on?  Presumably the former but the latter would be an interesting alternate direction to have taken magic in D&D.
 
Optional rule for rolling to cast may have the spell fail, or it might just be delayed a round, which the wizard has to spend continuing to concentrate.  The six degrees of spell complexity look like the beginnings of the spell level system.
 
Wraiths are very ringwraith and paralyze any normal man they touch; paralysis persists until touched by a friendly Elf, Hero, or Wizard, which rather reminds me of freeze tag / capture the flag.  Lycanthropes are only werebears and werewolves, and when near forest terrain they get bonus attacks from allied beasts.
 
Troll / ogre distinction is muddy and they don't regenerate, but they do fight as six normal men.  Their hit points are ablative though - six cumulative hits, whereas heroes need four simultaneous hits in a single round to be killed.  That's a really interesting distinction.  Hit points are great as a resource that can be ground down in dungeoneering but having damage be all-or-nothing / non-cumulative ("it takes 18 damage in a single minute to kill a 4th level fighter") is another strange direction D&D could've gone, like magic users concentrating on simultaneous spells.  True Trolls can only be killed by fantastic creatures but aren't very good at attacking normal men.  Giants fight as 12 men with cumulative damage, can throw rocks like a light catapult, and never check morale (except when hit with fireball or lightning bolt).
 
Dragonbreath is also save or die, and it's not three times per day - it's three times before they have to land and "remain stationary for one turn in order to rekindle his internal fires."  Dragons also attack other fantastic beasts regardless of side, and are immune to normal weapons and missile fire.
 
Wyverns, griffons, and rocs are all under one entry and can transport a single man-sized figure, so presumably these rocs are more like Tolkein's giant eagles than the roc of Arabian Nights.  Djinn and efreet are just normal elementals.  Wights and ghouls are in the same entry; they are immune to normal missiles, their paralysis lasts one turn, and you get a chance to attack them first.  Zombies are a subset of wights and ghouls but slower; they fight as orcs, but I can't seem to figure out what orcs fight as against normal men?  Ah, it's on the table, as heavy foot.  That's sort of interesting, that orcs don't fight significantly better than humans.

Weapon vs armor table.  Weirdly, the single-man scale missile fire rules basically have ascending AC - "Class of armor worn by defender", from 1 to 8, with 1 corresponding to No Armor and 8 corresponding to Plate and Shield.  So it's really surprising to me that we somehow got to descending AC from here, and then back to ascending AC later.

In conclusion: I'm amazed that we ended up with the D&D that we did given this starting point.  I love how all these units are unreliable in their own special ways; knights having to charge, mercenaries defecting, peasants cowering, orcs fighting each other, dragons fighting other beasts regardless of side.  It seems like it would lead to unpredictable but hilarious games.