Thursday, January 27, 2011

I love Cho'Gall

We got Cho'Gall down for the second time this week and the kill felt pretty solid and repeatable.  There are some bosses where wiping feels terrible but CG isn't one of them; I think it is because there is so much going on and so many opportunities for small errors to add up that it feels okay to wipe to him.  Also it may be that I really enjoy this fight, particularly because there are so many odd abilities we get to apply to challenges he presents that are normally unused in raiding.  In particular Worship forces us to really get creative and think about all the ways our raid has to interrupt people.  We have a rotation where the mage will dragon's breath us if he isn't worshipped, and if he is then the DK will hungering cold us, and the priests will Psychic Scream if those fail.  We have lots of other single target interrupts that we can use too, but setting up a system of AOE interruption for when we get Worship felt great because we really got to use all the parts of our various classes.

It brings me back to when we first pulled Battleguard Sartura in AQ40.  Stunning things was a really iconic paladin ability and yet it was pretty much totally unused in all of MC and BWL as every mob was immune to everything aside from simple damage spells.  I loved the idea that I actually got to use my stun against a boss and that it did not trivialize the encounter at all, in fact without using stuns properly he/she/it would have been exceptionally difficult to defeat.  I really enjoy getting complex challenges that can be solved in a number of ways but which require planning and thought to solve in the most efficient way.  Another good example of this from Classic is dealing with adds in P1 Razorgore in BWL.  You could CC the adds, dps them down, kite them, tank them, or any crazy combination.  Some methods were easier than others but we got kills using a wide variety of strategies and heard of even more being successful, which I consider a real victory for encounter design that people were able to find so many different ways to win.

Back to CG we see that same thing happening.  There are a lot of different ways to deal with the problems he presents.  There are single adds that have to be positioned, interrupted and controlled.  There are many small adds that need AOE slows, AOE damage and knockbacks.  There is obviously the aforementioned Worship that needs a huge variety of different abilities to deal with and the final phase has more adds that need interrupting, stunning and killing.  I don't think there is any other fight in the current tier that comes close in terms of getting us to use many different abilities and giving us many options to deal with the problems the fight presents while still being tightly tuned.  That for me is the kicker:  CG is tough.  Not brutal, but you have to play well and several small errors will get you killed.  It is a sign of tremendous raid design that there are so many ways to approach the fight mechanics and yet even once you have a good strategy the fight required good damage/healing/tanking numbers to beat as well as solid execution.  This is one of the best designed fights I have seen I think, with but one exception:  You can't bring lots of melee (except broken frost DKs) or you lose to the AOE phase.  Melee suck.  :P

Tuesday, January 25, 2011

What to do after you win

I have been doing a lot of CiV modding this week and have run into a real dilemma, both in terms of philosophy and execution.  When you capture a city you have three choices:  Burn it down, puppet or annex.  Burn it down is obvious, of course, but the others have some really fine points.  Puppetting means the city will be stupidly managed.  It will be like a normal city except it will do everything in its power to make gold ahead of any other resource.  This means it will grow slowly, build slowly, and end up being fairly weak.  It also can't be used for any military production and you can't control it.  Annexing a city means that you have to build a courthouse to make the city function and you can completely control it from that point forward.  The trouble with annexation is that the city then counts towards 'cities you control for real'.  This is important because there all your social policy costs scale by 'CYCFR', and to build the powerful National Wonders you have to have a specific building in each CYCFR. This means that annexation is simply not worth the cost in most cases - you are much better of puppeting the vast majority of your conquests.

I don't like this situation much because having a big empire and being a conqueror is fantastic when you puppet everything.  You get lots of social policies, more even than a small, focused empire, and you have drastically more science and gold and can make National Wonders just as easily.  The question is, how much do I want to reward attacking?  In CiV the attacker has many disadvantages.  The defender can see all the attacker's units and positions, heals faster, can use the local roads for mobility, has a city to do damage, and several policies/wonders give defenders huge advantages.  So if attacking and capturing territory is not particularly advantageous then the player would be well advised never to do so; just hold your initial territory and let the computer throw its hordes of dudes against your defenses.  This is also great because the computers get angry at you for destroying other civs, declaring wars and taking out capital cities.  That would mean that militaries were just for defending against dumb AIs who attack and I don't think I like a Civ game where attacking is just pointless.  Question is, where is the line?

The other confounding factor is that it is rather tricky to penalize puppet cities with the tools I have available to me.  I know that it is possible to assign flat penalties to puppet cities using LUA code but I don't know anything about that yet; learning it would be a long process.  Other coders are doing pretty much exactly what I want to do but I don't want to use other people's mods because they aren't *perfect* dammit!  Now that I have invested all this time into building my own system I can't deal with the choices that other people made that aren't quite the same as mine.  I am at a crossroads of challenge - I don't want to spend the hours and hours and hours it would take to learn LUA coding to make this happen, I don't want to just use other people's stuff and I don't want the game to be imperfect.  Something's gotta give, not sure which way it is going to go though.

Friday, January 21, 2011

Cho'Gall is down

This fight was a good one.  There have been a lot of fights in Cataclysm that were either very slanted towards lots of healers due to being very forgiving (Valiona and Theralion) or had a straight up berserk timer that simply caused you to lose at 6 minutes (Atramedes, Halfus).  I don't particularly like either of those mechanics as having bosses that are readily beatable once 3 people have died seems very weak and it always feels a little bit silly when a boss suddenly decides to use his instant death attack that he has been holding back on for the entire fight.  Cho'Gall, however, has a really good enrage mechanic that doesn't revolve around a hard berserk.  Once you get him to 25% he starts stacking up corruption on the raid and as your corruption gets higher and higher you start to do all kinds of really awful things and eventually you go crazy and die.  You cannot beat the timer without all your dps alive, particularly since he also summons adds to increase the burden.

We are the first guild on Vek'nilash to down this guy, or any endboss for that matter.  Granted some guilds are killing him in hardmode but in our little corner of the world we are the undisputed champs, for the moment at least.  It feels much better than it ever did before because there isn't the artificial 'well, we only did it in 10 man mode' tacked on to every boss celebration.  We didn't beat the easy mode, we just beat the fight, and finally we can just say that with pride and without qualifiers.  It must be a brutal challenge to design fights that are of roughly equal difficulty in both 10 and 25 man modes but since we are seeing most world first kills in 25 man and some in 10 man I assume they are doing a reasonable job by and large.

It is certainly notable how much harder Cataclysm raiding is than Wrath raiding.  In Wrath we had every normal mode beat within 2 weeks of launch and were wondering what to do with ourselves.  In Cataclysm we are 6 weeks in and we still have 2 normal modes to beat, 12 hardmodes and 1 super hardmode and we are the best on the server!  They certainly put in a good set of bosses from where I sit, with the first bosses being not too hard and the endbosses being a good challenge.  I can only assume the hardmodes will be absolutely brutal, which is fine, since casual guilds can see some content, good guilds can beat all the bosses, and superb guilds can beat all the hardmodes.  Sounds good to me.  Now we just have to find out if OGT is one of those superb guilds or not.

Edit:  I just saw the best achievement ever.

"I can't hear you over the sound of how awesome I am."

-Reward for beating Sinestra with no deaths in 1 try.  We need to beat Cho'Gall heroic and go get this RIGHT NOW!

Tuesday, January 18, 2011

How hard is it really?

Back in the first week of Cataclysm I had a lot of problems with heroic Stonecore.  My guild did some runs, usually with 1-2 puggers due to not having enough 85s yet, and we wiped to Corborus a lot.  We had huge troubles beating the adds quickly enough and the healer usually ran out of mana around the third burrow and we died.  Today we did heroic Stonecore with a new 85 who just barely scraped by with the minimum 329 gearscore and we mauled Stonecore.  We had some deaths, mostly due to people being extremely aggressive and unconcerned about dying, but we beat the place up easily.  It is amazing the difference in difficulty between those experiences.  Of course in the first week everyone was buying equipment they couldn't use and carrying around BOEs just to hit the 329 minimum requirement so we were really running at ilvl 325 or so while most of us now sit at 352.  That is a solid 29% more stats on our gear, which obviously translates to a ton more health, avoid and damage.  It is easy to see how that makes us better, as the dpsers end the fight in 80% of the time, the tank takes less damage and has a much bigger margin of error, both of which allows the healer to heal harder without running out of mana, which means we all live instead of die.

There are a lot of people complaining that heroics are too hard.  They are complaining from the perspective of 'I got to 329 ilvl and the dungeon is still really hard' most of the time, and that is entirely true.  Of course these dungeons are designed to be done with a minimum ilvl of 329, so everyone should expect that if you are barely even allowed to zone in that you better play damn well or you won't be able to succeed.  Crafted gear, rep drops, BOEs and quest rewards, not to mention drops from regular dungeons can get anyone up to a solid 340 ilvl without ever touching a heroic dungeon or raid but it does take time.  Most likely people are just bitter that they can't ding 85, hit the minimum ilvl and get carried by people in overpowered gear to massive rewards as they were used to at 80.  Heroics just aren't that hard when you actually make use of all the upgrade paths available prior to them but the necessity to wait and prepare is not something the latter half of Wrath taught us.

As an amusing comparison, consider the difference between 'just ready for heroics' gear level in wrath and someone in half raiding gear, half heroic blues.  We already saw in Cataclysm that difference is about 29% more stats, in wrath the curve was much, much steeper.  A basic set of mixed 187 blues and ilvl ~170 greens give you approximately 120 stats on a bracer slot, more with the blue, less with the green.  A mix of 213 raiding gear and heroic 200 epics has about 200 stats on a bracer slot instead, a 67% increase.  Combine that with the fact that heroics in Wrath were relatively easier and it is obvious why heroics became a complete joke - it was very feasible to drastically outgear them within 2 weeks of hitting 80.  The very act of making epics have the same scaling with ilvl as greens/blues has had a very powerful impact on reducing gear inflation, making sure that raiders have better gear but that a fully geared out raider in t1 content won't end up with a ridiculous 90% increase in stats from when they first hit heroics.  In Cataclysm if you were geared in the same fashion, clearing all normal content and having all badge purchases you would end up with an ilvl of 359, a 33% increase over entry level heroic gear.  That is going to mean that although better gear drastically reduces the difficulty we simply aren't going to see the level of disrespect for heroics we saw in Wrath and heroics are going to stay challenging for PUGs for a lot longer.

Thursday, January 13, 2011

Hardcore gamers making bad games

I read a post by Tobold yesterday talking about how hardcore gamers end up making games for hardcore gamers and casuals for casual gamers.  His point seemed to revolve around the idea that if you hire hardcore gamers you often end up with extremely deep games that take a ton of time to play and learn but which have limited appeal to most of the market.  There are hardcore gamers out there who will buy it but his idea seems to be that you hit a much bigger market when you build a casual friendly game.  I think this is partly right and partly overly pessimistic.

It is true that casual games outsell hardcore games by a substantial margin.  Appealing to those who want to play a game that is quick to learn and play is important if you want to make money, but the problem comes when you make the assumption that hardcore gamers can't make those games or when you think that games can't appeal to both sides.  It is easy to make a game that casual players find utterly unappealing - see Advanced European Theatre of Operations, which is a game with a rulebook that could be used as a blunt object for subduing ones enemies and certainly would require dozens (hundreds?) of hours  to learn all the rules for.  I remember playing a wargame with The Gentleman years ago that took us about 6 hours to get through; 3 hours to skim the rules and 3 hours to puzzle through the simplest of scenarios.  By the end we were still pretty clueless about the rules and had concluded that we needed to take a day each to read over the rules a few more times to be able to play the game reasonably at all.  I don't think anyone would consider this a good game for the mass market and yet certainly there are a few hardcore war gamers who play and enjoy it.

Thing is, you can make a good game that is both appealing to casual players and hardcore gamers alike.  The trick is you have to actually take both into consideration and do a good job to achieve this.  When people make wargames with 500 page manuals they aren't considering casuals at all and when Farmville was conceived the creators weren't trying to get game geeks to play, but rather just people who like to click on pretty things.  Just because a game can fail completely to attract one portion of the audience doesn't mean it has to however.  Scrabble, Chess, Texas HoldEm and Plants Vs. Zombies are good examples of games that are played by both casual and hardcore gamers alike.  It is entirely possible to explain and understand the rules of the game in very little time and still find a tremendous challenge in mastering all the subtleties of play.  The key is in Blizzard's mantra, "Easy to learn, hard to master."  To make a game that has simple rules that are quick to learn but yet has tremendous competitive complexity is not trivial but we should not give up on doing so.

That was pretty much exactly what I tried to do when I built FMB.  I wanted to have a game that I could teach to my family (who are generally pretty smart but aren't hardcore gamers) and also be one where really serious math/game geeks could sit around and argue about the optimal play for hours.  I think I achieved that, and if you look around you can find games all over that manage to be quick, fun, simple to learn and incredibly deep strategically.  It takes more effort and most importantly it takes a creator who genuinely wants to achieve both those things but it isn't a function of hardcore gamer or casual gamer but rather skilled game creator vs. unskilled.

Tuesday, January 11, 2011

Melee Hate!

Blizzard has all kinds of different ways to make a boss hard.  One real problem they have is that many of the ways they have to make things hard involve players either being in a particular spot or not being in that spot and ranged classes are drastically better at both those things.  Melee classes have tremendous restrictions on where they can be and this translates to a real problem in encounter design.

Last night Wendy and I were talking about how melee are bad in Cataclysm and trying to brainstorm ideas for solving that issue.  We wanted to talk about melee advantages and came up with only a few things that melee do better:

1.  Unlimited Resources.

2.  Immune to antispell attacks such as silence, interrupt, manaburn.

3.  Free cleaves.

4.  Mobile dps.

The first was supposed to be a limitation that ended up being axed a long time ago.  It was obviously terrible for ranged to be just better for the first 5 minutes of a fight, run out of mana and then be garbage after that.  It leads to brutal stacking based on fight length.  The second has been used here and there but is a fairly gimmicky, fight dependent sort of thing.  Ignis in Wrath interrupted casters regularly, which successfully made melee dominant at fighting him and there were several places in TBC and Classic where manaburns were used and which made casters much weaker.  The problem is that these mechanics really aren't very good these days.  Manaburns don't affect hunters and do affect melee hybrids and interrupts similarly don't affect hunters.  Blizzard cannot balance melee by using those sorts of effects.  Free cleaves were used liberally in Wrath and ended up being extremely overpowered in some encounters like Anub'Arak and ridiculous in heroics but were often worthless.  Blizzard got rid of that as a mechanic and I think they were right to do so.  The fourth point is actually useful.  On Grobbulus for example the boss has to move long distances fairly constantly and melee can be dpsing him almost the entire time.  When ranged have to move to follow him they often lose dps.  The issue is that the amount of movement required for ranged to be bad is very large and the mechanic can't be one that forces melee away from the boss or it just ends up being even worse for them.

So given the current model of unlimited resources for all dps, most antispell effects not working properly and free cleaves being gone, what advantage could melee bring to make them worthwhile?  My personal favourite is self healing.  Specifically in earlier incarnations of WOW Judgement of Light did a lot of healing to the raid and it lightened the healer load considerably.  If that sort of effect were available to melee attackers that could make up a lot for the additional damage they are likely to take due to having to be in close.  It would have to be something that triggered off melee attacks specifically and probably using a PPM mechanic to avoid rogues just being invincible.  It seems a somewhat natural sort of advantage to balance off the penalty of having to eat void zones and explosions and such to continue attacking with limited positioning choices.

Wendy came up with a truly unique solution that won't be implemented but is great to think about.  The idea is that melee can continue to attack outside melee range but their attacks grow weaker with distance.  Specifically some kind of function that reduces autoattack speed and special attack damage such that at 6 yards nearly full damage is done and at 40 yards zero damage is done would work.  The animations would have the character throwing their weapons at the boss and the weapon coming back could signal the timing of the next autoattack - this way the increased distance neatly links the animation with the mechanic.  It would mean that melee would be weak at range but they could do reasonable damage while temporarily at range.  Obviously since pvp balance would be completely demolished and immersion (for many people, not all) would be very threatened this will never happen, but it is a hilarious idea.

Thursday, January 6, 2011

Reversal of Fortune

When Cataclysm launched there were some real issues with Ret paladin scaling.  In particular Mastery was an absolutely terrible stat, as shown by the weights below.

Str - 204
Haste - 80
Crit - 74
Mastery - 38

On the PTR Mastery has been changed to add a substantial % of Crusader Strike and Templar's Verdict damage as holy damage and no longer to give additional attacks.  This is a major improvement in two ways and a major annoyance in one way.  The improvements are twofold:  First, paladins no longer generate Holy Power (HP) in single chunks.  That was a big problem where due to the delay of displaying HP we would often go over the 3 HP cap and would lose HP by doing so.  It also made the rotation feel really manic and shaky because every single strike had a large % change to change what the next strike should be and that % chance was only resolved halfway through the GCD.  Now all HP is either gained via CS in an entirely predictable fashion or comes in 3 HP chunks so the rotation is much smoother.  We can actually plan attacks a few seconds ahead and although we still have to react to multiple procs the interruptions will be less frequent.  The second improvement is simply in overall damage; Mastery is a much better stat now and Ret damage is improved significantly overall, in the neighborhood of 11%.  The new weights look as follows:

Str - 225
Haste - 80
Crit - 76
Mastery - 94

The major annoyance is that I, like most good Ret paladins, have been collecting haste/crit gear.  I deliberately avoided anything with Mastery on it to try to maximize my damage and now I need to go out and acquire Mastery gear instead.  If Mastery had been comparable to the other stats this wouldn't be an issue of course but it is significantly better, enough so that I will definitely want all of my pieces to have Mastery on them.  Of course I shouldn't complain too much since the numbers are *much* more in line now than before and getting a general dps buff is pretty great too but it is frustrating to have to do a complete about face, especially when I need to collect Mastery gear during the period where Mastery is still junk.  It reminds me a bit of the problem in Wrath where Survival hunters got a big buff and suddenly every raiding hunter had to swap specs to optimize their damage.  GC said that they felt like this was a big error:  Better to buff underpowered things very gently so players don't feel like they have to change their gear/spec/style en masse all of a sudden.  Having the worst thing suddenly become the best thing means that you get a lot of people ticked off.

Initially I felt like Ret paladins were heinously underpowered in Cataclysm.  I was bottom of the damage meters consistently and there seemed to be nothing I could do about it.  Last night in our raid I was #2 though, and that on fights that really favour ranged damage in a lot of ways.  I just don't know what is up with that; I have acquired a completely broken trinket and got a few upgrades but there is just no way I should be improving by 50% based on those few upgrades I acquired.  I wonder if there are some changes that have been hotfixed in that I just don't know about.

Tuesday, January 4, 2011

A new beginning, again

The great Sundering of Brightcape has begun.  All my gaming posts will hereafter appear here instead of on Brightcape.  Hopefully this will begin a new era of even better discussion.

Tol Barad.  What a bloody mess.  There was plenty of feedback during the beta that TB heavily favoured defense and yet it managed to go live so badly balanced that defense won at least 90% of the time.  This is the worst possible situation as if it had gone live where offense nearly always won it would at least have meant equitable access to dailies and the PVP boss but as it is one side gets to control the zone for the whole day until a server shutdown or a middle of the night defensive fiasco.

The solution, as we all probably know, is to make offense get 1800 honor for winning instead of 180.  I can't imagine how anyone thought that getting enough honor for a whole new piece of gear for a single win was a good idea but they went ahead anyway.  Unsurprisingly everybody figured out that win trading was the best possible solution to this situation and defenders refused to play at all so they could attack 2.5 hours later.  Clearly no bans  or other punishments are going to come out of this since half the level capped WOW population took part (many or most without any personal wrongdoing) but the rain of free pvp gear was incredible.  Finally a hotfix came in to stop the madness and the 1800 was ratcheted down to 360.  This is enough that is it worthwhile to try to win on offense I think but not sufficient to make win trading reasonable.  If you accept the idea that we are stuck with a ridiculous defense favoured map then this is a good reward system but the whole concept needs a serious rework.

My solution to TB is a very simple one.  The fundamental problem with attacking is that destroying the towers actually provides nearly no benefit considering the time it takes to take them down and that since equal numbers are guaranteed the attackers have a hell of a time taking the fortresses.  The best way to deal with this is simply have each tower that is destroyed grant the attackers a % bonus to the effect they have on capturing bases.  For example, if destroying one tower got the attackers a 5% bonus, two was 15% and three was 30% then after all the towers were down it is inevitable that the attackers win.  The question becomes can they get the towers down and cap all three fortresses before time runs out, the answer to which hopefully is sometimes yes and sometimes no.  30% might be too high and since I have only participated in two TB battles I don't have a really clear picture of how the zone plays out.  However, it is clear that incompetent attackers would lose (probably stymied in their attempts to knock down towers) and very competent attackers would win fairly quickly so the system is pretty much guaranteed to be better than what is in place right now.  It is much better to err on the side of favouring offense rather than defense so I would aim high for the bonus and tone it down if necessary.

Most of Cataclysm has been a real success, minor bugs and hotfixes notwithstanding.  TB however gets a huge thumbs down from me.  Usually I am a big Blizzard booster but they really, really screwed this one up.