Saturday, November 30, 2019

Fist to the head

Warrior monks are a staple of movies and stories.  Spiritual folk who combine magical seeming powers with physical discipline beating people up with their fists are a thing a lot of people like.  In Civ 6 the devs decided to put Monks into the game, which is a fine thing as far as I am concerned.  I don't know about the realism of people fighting fist vs. firearms and winning, but it is fun so who cares?

But as usual there were some serious implementation problems.  Two problems leapt out at me, once I began to look at Monks in Civ 6 - First, that they suck.  You don't want them.  Second, that you can't even get them if you wanted to.

The reason you can't get Monks is that they require a specific belief, and the AI is hardcoded to grab that belief first.  The AI won't build Monks once it has that belief though, so you won't see Monks in action.  It just takes that belief, reserving Monks for itself, then refuses to build any.  It feels almost like some developer was desperate to spite another one.  To put in those art assets and unique promotion trees and everything just to make sure that players don't ever get to see Monks in action... it is absurd.

You *can* build a wonder that gets you four Monks, but it comes up much later than Monks do otherwise so by the time you build it those Monks will be 20 Strength below other units and they will be weak and useless.

The reason Monks are terrible is that they have special rules.  They are vaguely like melee units in cost and strength, but melee units can benefit from Great Generals, get bonuses vs. anticavalry, and can use battering rams.  Monks get none of this.  Monks also can't upgrade or be upgraded into, so they have weak promotion possibilities.  Even if getting access to Monks was practical, you don't want them.  This is extra sad because it means that the belief that gets you access to Monks is terrible compared to the other associated beliefs.

But I can fix this!

All I have to do is buff Monk Strength, which is easy, and trick the AI into not taking them so the player can have a chance.

First off I tried to swap the names of beliefs around to trick the AI into taking something else.  This failed.

Then I tried to swap the modifiers around, thinking maybe they programmed the AI to look for a particular modifier.  No good, the AI still takes Monks and still won't use them.

Then I tried to change what the modifiers did at the lowest possible level, but this also failed.  The AI is hardcoded to look for the ability to get Monks, take it no matter what, and then refuse to make them.  Seriously people, what the hell?  Do you hate money?

But there is another solution:  Give Monks to everyone who gets a religion!  The belief that normally gets you Monks can give your Monks bonuses instead, so it still does a similar thing.  This one actually worked, so now in my mod every religion gets Monks and they are stronger and thus worth building.

Unfortunately giving Monks bonuses proved complicated.  The coding itself is trivial, but knowing exactly what stuff I have to insert into what tables is not at all clear, and my attempts to give Monks bonuses based on beliefs all failed.

So I went on the steam workshop, and looked up Monk mods thinking I could learn from them.  I found a mod from someone who went through the exact line of reasoning I did, right from 'Monks suck and are too hard to get' to 'give all religions Monks and buff them from the belief'.  This mod worked and was done already, so I gave up my attempts and just used it instead.

Yesterday was apparently a lesson in 'Before you do a job, see if someone else already did the same thing, but better.'

Monday, November 25, 2019

Whose bullshit reigns supreme?

In my last DnD session we had a showdown to see whose ridiculous bullshit abilities would carry the day.  As you level up in the game you get more health, more attacks, and more damage.  Those abilities aren't bullshit.  Bullshit abilities are ones that let you defeat opponents by denying them any chance to act, and rendering their abilities and stats irrelevant.

For example, we went into a room that had a weird shadow spider thing in it.  Its stats were based off of a shadow dragon, but it had at least one extremely important additional ability.  Shadow dragons are CR 13, and this one was located in a deep dungeon so its extra powers related to being in shadow were all in effect.  With us being level 9 it was extremely deadly, as it had a breath weapon AOE that hit for 60% of our health and the ability to hide for free on its turn and then run away so we couldn't target it.

All of this is nasty.  We started up the fight, it went first, and it breathed on us, taking off 40% of the party's health.  Then it zoomed back into its lair.  We ran in but couldn't see it, so we summoned some bats to use their echolocation to hunt it down.  I was impressed with its bullshit hiding ability, but we had a trick in mind:  As soon as it appeared, one of us would cast insect plague, an AOE that does damage every round, and then I would cast a spherical wall of force around it.  The plan was that it would sit there in the wall, unable to leave, and slowly die to the insects.

That is some bullshit right there.  I am going to use that combo on other nasty creatures, have no doubt of that.  We set up our combo and waited for it to die.

Except it had another whole level of bullshit.  In addition to all the normal shadow dragon stuff, it got to plane shift at will.  So every turn it can hide for free, move away, and then *teleport to another plane*.  Then it can teleport back wherever it wants, whenever it wants.  If it felt like it, it could just plane shift back as soon as its breath weapon was ready again and roast us.  Of course, it plane shifted out of the Wall of Force, and fortunately it didn't come back right away because if it had we would have flat out died.  What are we supposed to do against a monster that is never visible except during its turn, and can leave the fracking plane any time things go against it?

Level up our bullshit, obviously.

Our monk has the ability to stun monsters.  He can fire off four stuns a round when need be, so when the shadow spider came back to finish the job an hour later, having conveniently given us time to rest and heal up, the monk went first, ran up to it, and stunned it over and over.  It finally failed on the third try, so we got to spend a round unloading on it.  On his next turn he stunned it again, and then again on his turn after that.  Before it got even a single action we plowed through nearly 200 hit points even though it was resistant or immune to everything except psychic, radiant, and force damage.

I don't know about high level play against such creatures.  It seems like a silly contest of looking up in the rulebook to see whose bullshit abilities trump whose and the loser doesn't get to do anything on their turn.  Winning is its own kind of fun, clearly, but it doesn't actually feel that great to cast Magic Missile over and over against a helpless opponent until it finally keels over.

It also must feel absurd to be the GM and to have to stare down Insect Plague / Wall of Force or endless stuns.  What is an enemy with powerful abilities but without teleportation supposed to do?  Stand there and die, I guess.

Of course if you add 'plane shift at will' to a powerful monster, it ought to be considered several CR higher than listed to account for the extra power.  I wonder if we will get experience for a standard CR 13, or a full and proper CR 15 reward?  I learned a lot from that encounter, after all, since I cast a level 1 spell 3 times.  But a bigger and badder target for that level 1 spell changes how much I learn about magic!  (XP systems are stupid).

Tuesday, November 19, 2019

Run, a dragon!

In my DnD campaigns over the past few years we have run into a number of dragons.  They have varied wildly in power from baby dragons at CR 4 to legendary dragons at CR 20.  One thing that has been consistent though is that when you run into a dragon it is dangerous for you right then.  Never have I been in a level 8 group that ran into a CR 4 dragon that we could just beat up easily, and I think this is because adventure designers make sure that every time you see a dragon it is set up to be a deadly opponent.

Other creatures aren't like this.  When you are level 1 or 2 an ogre is a dangerous foe, capable of downing characters in a single hit.  By the time you are level 8 you know that ogres are trivial and you can just blast away groups of them with AOE attacks or tear them apart with weapons if you want.

Dragons though, those aren't fixed in difficulty.  Some are trivial, some are terrifying, and they are constantly positioned so that they are a deadly encounter for whatever level the party is at the time.  I think sometimes adventure writers forget how weird this is.  My character knows that they have grown dramatically in power.  They know that previously challenging foes are easy to defeat now.  And yet when we find out that the ruins we are investigating contain a dragon we are supposed to automatically be terrified.  Even if we beat a dragon long ago, this new dragon is definitely something we should be worried about.

It feels strange because our characters know for sure that if we are level 10, there are dragons that we can beat up without worry, dragons that will be a serious challenge, and dragons that will kill us all without trying.  Knowing 'there is a dragon!' tells us almost nothing about what challenge we will face.  Except, of course, that we know that every dragon in an adventure will be a terrifying challenge, because that's how adventures are written.

What I really want to see is an adventure where the characters face a dragon that just isn't a challenge.  Let something else take centre stage, and subvert expectations. 

The actual fight against the dragon this week was an illustration of the issues with the way crossbow characters work in 5th edition.  I am playing a crossbow specialist and my feats allow me to fire crossbow bolts in melee range without penalty, as well as shooting without any attention paid to cover.  We had our two melee characters ready to fight the dragon, and I hid in the far corner of the room behind a pillar.  I could easily blast at the dragon at full strength from there, weaving my bolts around obstacles.  When the dragon flew across the room to bite me, I stood in its face shooting away anyhow. 

When you build a crossbow character your numbers are good, no doubt, but the real issue with the build is how it doesn't care about the fight.  I don't care about distance, terrain, obstacles, anything.  As long as I can see an opponent, I pump huge damage into them.  That is great from a 'be powerful' perspective, but it does have something lacking in the 'fights are interesting' perspective.  I suppose is it still interesting for me to use cover and terrain against my enemies while ignoring it myself, but I think the game might be better if I didn't have all of these abilities to ignore the map.

Saturday, November 9, 2019

Playing nice

I have been doing better in Through The Ages against the AI, and just got my best score today.


Barbarossa into Napoleon - I warred and warred and warred some more, resolving a culture war on all three opponents on my three final turns.  Smash!  I do not cooperate, I take what I want.

Sometimes though, I like to look at competitive games and imagine what they would look like as a cooperative game.  For example, last year at WBC I spent some time with people figuring out what the maximum score you could get in Castles of Mad King Ludwig would be.  You have to look at the game totally differently of course, and it is interesing to imagine how differently the game plays when the goals shift.

This week I tried that same sort of thing with Through The Ages.  I wanted to see what the highest possible score you could get for a single player would be.  Clearly the game has a lot of RNG so mathing out the perfect score would be a monumental effort - instead, I just wanted to try playing a 4 player game where the goal for all 4 players was to maximize player 1's score.

Think for a minute on what you think that score would be.

I thought for 15 minutes or so and came up with 2500 as my guess.  The key difference in the game is that it would be *drastically* longer.  You want as much time as possible to rack up as much culture as possible, so everyone mostly just takes the single end card each turn.  Player 1 needs to corral all the culture at the end of the game, but that is easy to do by having all the other players make as much culture as they can and then declare culture wars on player 1.  Since player 1 is the only one with any military, they can easily vacuum up all culture in the game.

I spent five hours running through a game in hotseat mode, where unfortunately I had to watch a recap of every turn before each player got to play, making this a tedious affair.  Here was the result:


Just shy of 3000 culture.  I screwed up my second to last turn though, and should have had over 3000. I forgot to scoop up a random tech, and it cost me.  The other 3 players scored some on the final turn and collected some impacts, so even though they all zeroed out their culture on the final turn they still got quite a lot of points!  I feel good about my 2500 estimate, as even though I was off by 500, I certainly got the ballpark right.  The main reason I underestimated was that I assumed the players would take a total of 5 cards per turn between them, but by the time I got to the end game I realized that no extra cards were worth taking, even if the ordering was really annoying.  Everyone just took whatever the last card was, no exceptions, and that extra game length gave me a lot more points that I had thought.  The table was making ~150 culture a turn, so even a couple extra turns is a huge swing.

There were some things that I hadn't quite counted on.  Captain Cook was an absolute monster, giving me 12 culture a turn for almost all of Ages 2 and 3.   I also set up the game so that I could win some big wars over territory and science in the midgame, but that might not have been such a good idea.  Player 1 got so many yellow cubes from colonies and had so much science production that winning those wars wasn't that great - the other players needed those cubes to maximize their own culture generation!  

I ended up at 222 Strength, which is kind of hilarious, but I could have done it with much less.  The other players had tons of turns to declare culture wars and hand me all their stuff.  Also the game has some glitches and struggles when you do things like this - it turns out that culture numbers greater than 999 cause issues at times, and the displays for colonies and wonders start to be a problem when you have 11 colonies and 10 wonders.  I also discovered that you absolutely have to maximize your civil actions, because when you have 9 wonders already in play you have to have 10 civil actions to take that last wonder!  It was totally playable, but things occasionally looked ... a little weird.

I am sure that with better play and luck I could get a much higher score yet.  I don't know if breaking 4000 is possible with a normal deck distribution, but I am sure if you could lay out the deck exactly the way you want you could get massively higher than that, surely as much as 5000 under ideal circumstances.  Given the way hotseat games work with the app though, I am not looking forward to trying this a bunch of times.  5 hours of clicking through warnings about unused civil actions and corruption is not much fun.

Monday, November 4, 2019

Not quite Alpha

I have been playing a bunch of Through The Ages against AIs.  I like playing against humans in theory, but in practice they spend time taking their turns, don't like it when I just wander away for an hour, and otherwise have needs.

AIs don't have needs!  They play exactly when I want them to.  So much more convenient than meat sacks.

The AI in TTA is quite good.  I have been playing a bunch of Skyrim and Civ 6 lately, so I have been exposed to hilariously bad AIs, which may have lowered my standards some.  Watching Civ 6 AIs send endless troops to their deaths and declare war on enemies they can't even reach doesn't fill me with a sense of awe, and laughing as I slowly whittle down Giants with my bow because they can't climb onto the rock I am standing on is no less silly.

TTA is vastly simpler to program than Civ 6 or Skyrim, no doubt.  Navigating terrain is a huge thorny mess, so it should be no surprise that the simple integers of TTA are much easier to solve. 

I seem to be able to crush the maximum difficulty AI consistently now, but I have managed to fail twice.  The first failure was just a second place finish that happened because one AI managed to be the punching bag and the other two stole all of its culture before I could resolve my first war on culture.  Some bad RNG on events in addition to poor military card draws left me unable to win. 

The other time was far more clear:  I got cocky and figured the AIs wouldn't beat me up for a round.  I didn't have any military bonus cards, but I had a decent hand size.  Surely they would respect the hand size, right?  Instead of respecting me, they resolved two aggressions, blowing up my buildings and taking my rocks.  Next turn they took my rocks again, killed a population, and declared a war over territory.  I gracefully resigned, realizing that I had entered the death spiral from which nobody emerges intact.

So lesson learned there.  The AIs will build military and kill you, if you let them.  Don't let them.  In fact, this trend is stark.  In my games with humans the scores are often fairly high as some people go for massive culture generation.  In my games with AIs though, the scores are miserably low.  It is always a hardcore military race ended by massive wars on culture with whoever fell behind. 

In fact, I have never yet built a library or theatre building.  Two whole *types* of buildings, and none ever created by me.  How can you afford such things?  If you build them, violent people will show up, tear them down, and equip more violent people with the remains.  My mind is filled with images of Knights armed with violins and armoured with pieces of bookshelves.  Better to build your own violent people to tear down the libraries and theatres created by others than face such humiliation.

It makes leaders that focus on libraries and theatres seem silly.  I look at them and laugh.  Sure, I could get more culture if I had a huge culture engine.  But how am I ever going to have a huge culture engine with all the ravening hordes at the gates?

When I first started playing this game Umbra told me about his Plan A:  Get to 70+ military strength at the end of the game and drop a huge war on culture or two on the last two turns to crush people.  This plan works, and I have been following it avidly.  I feel like if I am going to play against great players I will have to have more things in my toolbox than this, but so far it feels like this is the way to beat the AIs consistently.

Sunday, November 3, 2019

So much blood

Diablo 4 has been announced.  Clearly at this point in the game creation pipeline we have to be careful, as Blizzard likely won't release it until 2021.  Still, the cinematic trailer and gameplay videos are *gorgeous* and I am twitching with anticipation.

I was twitching with anticipation for D3 too though, and that had.... some serious flaws.

I think there is room for optimism though.  D3 had some huge issues that were created by Blizzard trying to solve problems, and they realized that they went too far from their source material.  The Auction House sounded like a good idea to me, because they were trying to eliminate the inconvience, spam, and scamming that were a constant in D2.  A good goal, to be sure, but it ended up a game where the best way to progress a character was to sit on the AH all day farming gold, and then just buy whatever you needed.  That did not work out.

D3 also had the issue of frequency of set and legendary drops.  When it was first released sets were theoretically in the game, but after farming max level content for 100 hours I had found a single set piece.  The timeline to actually assemble a set looked like it would be in the 2,000 hour range, and that is completely absurd.  On the other hand, at this point in D3 you get your sets immediately and in huge quantity and you just farm them over and over looking for ideal stat rolls.  Also those sets increase abilities by 2000%, which throws all game balance completely out the window.  Both extremes are a huge problem.

It seems like Blizzard is aiming for something in between D2 and D3.  They are going back to a skill tree and multiple ranks in skills, which sounds good, but their examples clearly show characters using a variety of abilities while fighting.  D2 had the problem where characters would usually end up just using a single ability, and that gets pretty boring.  D3 had the problem that characters had no permanent attributes, just a level that was shared across the account.  In theory they could combine the best of both of those games and come out with something marvellous.

I know what I would do for D4.  I would have a skill tree like D3, and limited ability to reroll or change point allocations.  I would gate higher level abilities behind total points invested in a particular tree, so that putting points into lower level abilities still makes sense - you have to do that to unlock the higher level ones.  D2's solution to this (passive skill benefits) was extremely limiting on builds, and D3's solution was to get rid of the skill trees completely and I don't like that much.

Between D2 and D3 lies a truly magificent game.  One I would play the hell out of for years and years.

Blizzard can create a great cinematic, and they can make the game look pretty.  By pretty, of course, I mean dark, bloody, brutal, gothic, and terrifying.  But whether or not they can combine the pieces of those two games into something wonderful remains to be seen.

I should be cynical and hesitant by this point.  I have been burned before.  But hope springs eternal.