Wednesday, July 23, 2014

The big shift

Awhile ago Val introduced me to the game Six.  They were looking for somebody to test their skills as most of their opponents weren't able to put up much of a fight.  I think they weren't trying terribly hard in the first game or perhaps I was just lucky but either way I got through the midgame and then realized that there was an opportunity to get a hard lock on them in the endgame.  The first hard lock I found turned out to be against the rules (which was a good thing, it was ridiculous) but I found another one quickly and won the game.

Val ended up beating me something like 6 games to 4 because they are much more practised at finding the geometries that force wins than I am.  They expressed some dismay at what I had done to the game afterwards though - I have apparently ruined Six for them.  The game originally was all about making trying to arrange tiles to form one of the winning structures and after playing against me it was suddenly about something else entirely.  In order to win Val suddenly had to play a new game that just wasn't as much fun and once you see that new dimension to a game it is pretty hard to ignore it later.

This same sort of transition has happened to me many times.  In particular I remember it happening when playing WOW.  After going through a raid with a huge number of people and getting shiny and amazing gear I just couldn't feel the same way about killing dorks and levelling up by myself.  The challenge and reward of raiding was a whole category up from the rest of the game and once you have tasted that the simple things just lack something.  That transition happened again when I lead raids.  After directing forty people to do complex tasks and having to see everything everyone was doing all at once just taking orders and bashing faces paled a bit in comparison.

You can't go home again.  Except with dragons, and shiny loot.

You can't go back to the giant pile of purple armour soaked in blood in the dragon's lair?

That lacks in panache, I think.  At any rate I apparently wreck games in this way.  Val is happy to play games well but doesn't invest in breaking them the way I do and as we have seen once a game is broken in that way it doesn't get unbroken.  It isn't true of just me of course - many or even most of my friends from university do this same thing to some extent or other and we spent many an hour doing it with each other taking games to whole new levels.  At this point it is a reflex, done automatically.  I can't not see the breaking points and playing a game without trying to win is unsatisfying.  Sadly it means that there are some people I can't play games with because I either have to refrain from breaking the game which makes it no fun for me or I break the game and then it is no fun for them.

Monday, July 21, 2014

Unbalanced

Tobold talked recently about how he thinks DnD Next is the most unbalanced edition of DnD yet.  Of course he calls in 5th edition and the designers just call it DnD, so I don't even know what we should be calling it.  I will just stick to Next and eventually that will sound ridiculous but hopefully people will have sorted out a widely accepted nomenclature by then.  At any rate Tobold thinks Next is super unbalanced because casters still get to be incredibly overpowered at high levels but they get nice things at low levels too whereas in the old editions casters were useless at low levels as a method of 'balance'.  Balancing classes by making it such that for the first year of play one class is good and for the second year of play the other class is good is terrible though I suppose it is a slight improvement over one class just being superior all the time.

However, Tobold's assertions about the way Next works are completely wrong.  He talks about wizards getting a cantrip they can cast at will that does 1d10 damage as if this is a big thing.  Archer fighters get a cantrip that does 1d8+4 damage and has a much higher chance to hit - it is called 'shoot my bow'.  Those same fighters also have the option of dropping their bow and chopping up the enemies with swords if they are so inclined and then they get to do a lot more damage.  They also get way more HP, better Armour Class, and amazing passive healing.  At low levels wizards are weak and flimsy in combat but do have a couple spells that will occasionally be useful - this is not different from older editions but the difference between the two classes is not nearly as pronounced in Next.

At higher levels fighters continue to do the most damage.  Even on the rounds that the wizard casts one of their best single target spells like Disintegrate it is extremely likely the fighter does more damage and the fighter can do that all day while the wizard runs out of gas quickly.  The only combat situation where the wizard is really unfair is AOE - there is no question that they dominate the field when there are a ton of dorks to destroy.  In terms of straight up combat damage math there is no question that fighters deliver more damage than anybody against a small number of targets and that holds true from level 1 to 20.

The thing that causes problems is when combat ends and the wizard teleports the group across the world, knocks down a castle wall, goes ethereal, flies around, and becomes invisible.  The fighter watches all this and sighs "So, I can jump pretty far.  Anything that needs jumping over?  No?  Cool..."  To be honest if the group is playing Next the way it is explicitly intended the wizard is probably fairly weak in combat because there are many fights in each day and they need to use many of their best spells out of combat.  Also they have to know ahead of time which spells they might want to use and that will reduce their efficacy way below theoretical maximums.

So while I don't want to play Next and I think they could have done much better I do think it is better balanced by far than first, second, or third edition.  Fighters are the best at fighting.  Everyone else is decent at fighting and has other tricks going on that make them useful.  Spellcasters are dominant at high level but not *nearly* so dominant as they were in the past.  Let's give credit where credit is due - if you want old school DnD feel and you want as balanced a game as possible Next is definitely the game for you.

Friday, July 18, 2014

Mook with a crossbow

While roleplaying last night MattInTheHat pointed out one notable flaw in Heroes By Trade. The trouble that he saw was that combat and noncombat encounters were so strongly differentiated. The simplest example was one that came up in a previous campaign: The characters opened a secret door and found a villain hiding in a small room with a crossbow pointed at them. The crossbow fired hitting the lead character for 10 damage and then the villain was quickly captured. The injured character had 30 HP so they were not really injured and rapidly healed back to full since in HBT the only way to take serious damage is to lose all of your HP first. The trouble with all of this is that there is simply no danger whatsoever in a random villain with a crossbow. Everyone can just laugh, take it in the face, and move on. The potential drama of the moment is lost.

 I wish that this wasn't true and that such encounters could be tense but the solution of making the mook with a crossbow potentially dangerous is fraught with issues. If Big Fucking Heroes can be really threatened by a single mook then one must assume that if they end up in a battle with four mooks they are reasonably likely to die. I want a system where Big Fucking Heroes can walk into an ambush set by a handful of bandits with crossbows and survive and these two scenarios cannot reasonably coexist. To be fair, there is a way to get close to it, which is to have very small chances for very extreme outcomes of combat actions. I could for example set it up so there is a .1% chance of an instant kill on each attack which would definitely make a single mook more dangerous and wouldn't change the odds much for a band of mooks but it would mean that the characters would just be outright killed on a semi regular basis and that holds no appeal. If combats are somewhat predictable and characters have a chance to react to bad things that happen then a mook with a crossbow just can't be dangerous.

 Other systems solve this problem by having magical healing. In DnD you likely won't lose to a mook but if he hits somebody you have to expend resources to heal that damage. It isn't a particularly big deal but it means that the characters have *some* reason to be concerned. I really hate the idea of ubiquitous magical healing though and HBT is built around the idea that it doesn't exist, or at least that it is so limited that it doesn't cause the problems normally associated with magical healing. One possible solution is to change the way HP regenerates after a battle. Right now it all comes back without any cost at all but I suppose I could change it so that HP regenerates only if the character spends actual Wound Points to make that happen. For example, if a character takes 10 HP damage during a fight they might have to take 1 Wound Point damage to get their HP back to full. This would add a new layer of complexity to healing but would certainly make people hesitant get shot at with a crossbow because although it won't kill them it could cost them resources. It is a possibility worth considering at any rate.

Friday, July 11, 2014

Danger Zone

When running a roleplaying game as the GM you have a responsibility to make the players feel excited.  The surest and simplest way to do that is to convince them that they are on the verge of losing and usually that means nearly getting killed.  The trick is that if they are constantly about to die and you use dice to determine outcomes they are eventually going to all get killed and the campaign will end and that kinda wrecks the tension of the moment.

Old school DnD achieved this sense of imminent doom by setting up the game such that any random enemy could roll well and kill a character in a single attack.  A single bandit with a bow was a terrifying threat because that bandit could easily roll to go first and kill two party members before anyone in the group got a turn.  Lots of other roleplaying games were the same and I have plenty of examples on my shelf - All world of darkness titles are definitely on the list, as are Cyberpunk, Shadowrun, most or all Palladium games, etc.  In all those games pretty much any combat was terrifying because you could always just die.

Games have been moving steadily away from that model.  DnD did so very dramatically with 4th edition and I took a few steps further when I build Heroes By Trade.  I wanted combat to be more predictable and I wanted that feeling of worry to exist without constantly offing people when the dice didn't go their way.  I prefer a system with a lot more granularity between 'totally fine' and 'dead'. Beating that single bandit with a bow didn't feel particularly gratifying or heroic because reverse lotteries where you almost always get nothing but occasionally get a catastrophic result aren't much fun.

DnD Next definitely wandered back to its roots in this regard.  Now it is significantly better in that people don't regularly start the game with 1 hit point (Hopefully you don't get hit... ever?) but there is still plenty of 'okay, the bandit shoots you with a bow, critical hit, take 18, you die'.  It is certainly less dangerous than the old days but I think it is still too dangerous to be the ideal game.  As a simulator it is perfectly reasonable for people to die to a single arrow but for roleplaying purposes I don't much like that and as an engaging combat game that feels weak too.  You just don't get to make very many interesting decisions when the enemies kill you before you get a turn.

Real life is super dangerous and lots of things can instantly kill you.  In the fantasy worlds I want to play in there is a lot more of the heroes being hurt, worried, and panicked and a lot less of them just slumping over dead without warning Wash-style.

Tuesday, July 8, 2014

A hard fight

DnD Next is out to some extent.  The starter book is live and people have begun playing the game albeit with a somewhat condensed set of rules and very little in the way of charts and information for the GM.  In a way this is actually a good thing since I find that DnD Next still has too many charts and lists and such though I will concede that it seems to be far better that its predecessors in that regard.  It does not have seventeen different polearms, say.

One thing that really concerns me is their guidelines for building encounters.  I don't have all the information yet of course but what I do have seems troubling.  Encounters are meant to be designed by adding up the XP values of all the enemies and then comparing those to the XP budget for the party.  Here is one line of the guide for a level 2 character:

Easy - 20
Moderate - 50
Challenging - 140
Hard - 210

The thing that is crazy here is the scaling.  When you have ten times as many monsters in a battle they last ten times as long and do ten times as much damage.  Obviously AOE effects can change this equation but amping up an encounter from an Easy encounter where the characters expect to take a tiny amount of damage, say three, to Hard you could reasonably expect the characters to now take three hundred damage.  That is, if the Easy encounter is the most trivial thing imaginable then a Hard encounter on this same scale is absolutely lethal, no possibility of survival.  Going from a Moderate encounter to a Hard one increases the number of monsters fourfold and even this is enough to mean certain death.

I have a vaguely similar system for Heroes By Trade where I wrote up the character's chances for victory based on how strong the monsters were relative to them.  It was based on both actual play experience and thinking about real life - if I have to fight a random dude the outcome is very much in doubt.  If I have to fight two dudes I get absolutely mauled, no question.  If I get to fight a dude with a friend helping me I win trivially.  This sort of thing happens similarly in tabletop games - doubling the numbers one side has guarantees their victory if the outcome was even remotely interesting beforehand.

They try to deal with this by suggesting that you can't just add monsters in because their effective cost is increased if they outnumber the party.  This blatantly says that they don't have the scaling of XP values for creatures in the right place at all.  If an orc is with 10 XP and an ogre is worth 40 XP then a fight against eight orcs should be about as hard as a fight against two ogres at 80 XP each.  According to their system though the eight orc fight should have a multiplier on it because they outnumber the party, but that multiplier wouldn't come into play if the fight was four orcs and one ogre even though the XP total for all three encounters is the same.

The issue here is that the game designers are trying to stick to old ideas about XP in the face of new math.  The new flatter progression of attack and defence bonuses mean that orcs are relevant forever.  A volley of arrows from a bunch of orcish archers can be devastating even to a high level party.  The reason this is an issue is that instead of making orcs worth 10 XP and a Lich worth 70 XP they are stuck in old school mode where high level monsters have to be worth orders of magnitude more than low level monsters.  It almost seems like they decided that a lich has to be worth 1000 XP and an orc 10 XP and then tried to put together a encounter balancing system from that standpoint instead of actually playing the game to figure out how many orcs a lich was worth.

If you keep this is mind it explains the scaling craziness above pretty nicely.  The fact that they expect you to multiply the XP budget by four for a Hard encounter over a Moderate is because monsters worth four times as much experience aren't four times as hard.  Not even close.  This means that any time you have a fight against a single opponent that has a reasonable XP budget the fight should be incredibly quick and lethal because the enemy can't possibly be appropriately tough and if you are fighting a horde of dorks you will get mulched.

If you sit down and play with the various monsters it is entirely feasible to figure out how strong they are relative to one another.  Not with perfect precision of course but you can come up with a good ballpark.  When you decide what the numbers are before actually doing your homework you end up with convoluted systems that are barely better than 'just eyeball it'.  Unfortunately this is what the encounter system in DnD Next amounts to - XP values are little better than wild guesses and encounter building is left as an exercise for the GM.

Not that every fight has to be balanced of course.  Having fights just be there in whatever way makes sense in the world is a perfectly fine way to play.  However, if you are going to build a tool for the GM to figure out if a given encounter is winnable you should make that tool properly.  GMs should feel free to ignore the tool and roleplay but the thing should *work* for those who intend to use it, especially if you are including it as a guideline for new GMs who want to make a fun adventure that is challenging but winnable.  We don't want to go back to the good old days of building encounters randomly and hiding dice rolls behind the screen so you can cheat when you accidentally make a fight the players can't win.

Sunday, July 6, 2014

Precision

Sometimes the gaping chasm between what an ability's description implies it does and what it actually does leaves me shaking my head.  In particular I am thinking of the Bladesman perk in Skyrim.  The description is:

Bladesman Rank 3:  Attacks with swords have a 20% chance of doing even more critical damage. (+50% crit damage).

So what does critical damage mean?  The default assumption people tend to go with is that a critical hit does double damage.  Since there are no in game ways to get clear numbers in Skyrim you need to do a ton of controlled testing to find this out and since you can only do that testing after selecting the perk permanently that isn't much help.  If you assumed that critical damage means that the damage is doubled you would be wrong - it is in fact 50% bonus damage which makes this ability extremely weak.

But there is more.

That 50% critical damage only operates on the base damage of the weapon.  For example, a daedric sword does 14 damage so the perk above gives a 20% chance to do 10.5 extra damage, or 2.1 damage per hit.  My regular hits counting all of my ludicrous multipliers end up putting my attacks at around 13,400 damage.  So I can spend three points on Bladesman to add 0.015% to my damage.  Given that perks often add 10% or more to overall damage these three points adding 0.005% each is a teensy bit underwhelming.

This is something that has gotten me thinking a lot about how games write up abilities.  The Skyrim perk above is obviously coded wrong - nobody was intending that it be 2,000 times worse than other perks at endgame.  However, reading the perk really tells you nothing about what will actually happen when you use it.  It reminds me a lot of old versions of DnD and other RPGs that use vague descriptions which require tons of houserules and adjudication by the GM because the writeup simply does not tell the reader what happens in a clear manner.

4th edition DnD was a huge departure from that because they made it extremely clear how abilities were used, what their area of effect was, and exactly what they did to the enemy.  Everything is listed in a fixed order and figuring out what an ability does is extremely simple - little to no adjudication is required.  I loved it when it came out because I always found poring through pages of text on abilities to figure out how they worked frustrating as it dragged the action to a halt.  When I wrote up powers in Heroes By Trade I wrote them very much like 4th edition powers with italicized flavour text and rigid, clear structure to describe the mechanical impact.

The trouble is that even though this newer system is much more clear and avoids rules conflicts it somehow lacks a style that people want.  When I titled an ability Flying Smash and its flavour text talked about leaping across the battlefield people were bitter that it did not in fact allow you to fly short distances.  They wanted to be able to defy gravity at will once they learned this ability and that was clearly a terrible idea since it is available to a starting character!  People have mentioned that they don't like the clear separation of flavour and crunch in the powers and want things to feel more seamless.

Thundering Charge

Rushing headlong across the battlefield you build up energy and crash into your enemies with a thunderous boom.

Rank 9
Effect:  Take a Move.
Target:  One creature adjacent to you
Hit:  Physical damage +5.

Augment 3
Make the Hit Roll against two additional targets: one before the Move, and one after the Move.

It is a tricky balance to strike.  I love that I can talk about opponents being staggered in the flavour text but have the power simply knock them back two spaces without having to worry about people trying to figure out if staggered means something in particular.  I want to talk about people being set on fire without everyone being confused as to whether or not their equipment is destroyed and don't they have to run away in a panic now?  (No, because it doesn't say they are Afraid so they aren't!)

I don't want to roll things back to the good ole days where when someone casts a spell nobody really knows what the result will be.  I also want the game to feel viscerally satisfying and have descriptions of powers be exciting and engaging.  I think what I am going to end up doing is loosening up the writing constraints I placed on myself for the crunch section and try to just write a descriptive paragraph more than a set of bullet points.  Hopefully it is possible for me to make the powers sound more exciting and feel more natural without making them unduly confusing and unclear.

Wednesday, July 2, 2014

Dodging is bad, just take in on the chin

A new patch for Diablo 3 is in the works and there are lots of changes, most of them good.  The one that interests me the most is that both the monk and the demon hunter are going to get armour from their dexterity instead of dodge.  This is a massive upgrade and should make both classes much more survivable.  Prior to this those two classes really got the shaft because their main stat was poop for defence.  The trouble was that everybody else got free armour or resist from their main stat and basically got to take half damage from attacks.  Demon hunters and monks got to dodge about 35% of the time which is shockingly bad in comparison.

The question that is useful to answer is how much dodge chance would you need to make up for lacking 50% flat damage reduction?  Clearly the answer isn't 50% dodge because consistent reduction is greatly superior to chance based avoidance since unlucky streaks kill you and lucky streaks aren't that exciting.  I suspect the answer is something like 90% since with that much dodge you could really expect to avoid most of the damage in nearly any situation.  As long as the monsters aren't cracking you for a third of your health on every hit 90% dodge is probably fine.  In softcore, that is, since death every so often isn't that big a problem.  In hardcore you need something like 97% dodge to make up for 50% reduction.

Of course some characters avoiding 90% of attacks is a ludicrous situation.  For one you have to worry about people pushing the envelope by stat stacking and getting to 100%.  For another PVP becomes completely ridiculous as the dodge classes would win nearly every fight without a scratch.  There is simply no way to have 90% avoidance that scales nicely with a stat and have that not be horribly broken in one direction or the other.

So Blizzard has finally caved in and accepted that their system which has been broken from day 1 is best salvaged by just flattening everything out.  It isn't as if dodging actually introduced any interesting choices or mechanics so just getting rid of it entirely is a fine plan.  Having a variety of ways to defend yourself is interesting when those ways actually generate thinking and varied strategies and the old system did not do that at all.

However, there is a way in which the old system could have been kept in place without losing the flavour of being dodgy vs. tough.  My suggestion would have been to drastically reduce the bonus other classes got to their defenses from their main stat.  If all the strength and intelligence classes got about 40% of their current benefit from their main stat things would be a lot more equitable.  They would then be getting something like 20% damage reduction vs. the 35% dodge that monks and demon hunters get.  You still might choose the 20% but at least the dodge classes would legitimately be taking less damage instead of taking more damage *and* spikier damage.

The trouble with my solution is that it would require significant rebalancing of monster strength and would been seen as quite the nerf.  People tend to celebrate buffs and curse nerfs even when their relative power to the monsters stays constant.  From a game variety perspective I think my solution is much more elegant especially as it would reduce the power of everyone's main stats and personally I think that might introduce a little more variety and thought into gearing.  However I don't have to content with legions of people complaining about their characters changing so I understand Blizzard's position.  They would rather the dodgy classes suddenly feel a lot tougher and everyone else just continue on oblivious even if it does mean giving up a smidgen of variety.