Thursday, February 24, 2011

No grind or no content?

There is a new MMO coming out very shortly called Rifts.  It is vaguely like the king of MMOs, WOW, in that it has a fantasy setting and many of the controls and interface elements are the same.  As usually happens when a good looking new game comes out people talk about how it compares to WOW and end up saying really ridiculous things.  In particular I have seen people talking about how Rifts doesn't have the reputation and gear grinds that WOW does and generally they are very happy about that.  I think this comes from a fundamental misunderstanding of what grinds are and how they are useful.

Essentially what it comes down to is an MMO has to provide many challenges for players on an ongoing basis.  Those challenges can be extremely trivial or brutally difficult or any place in between.  There are constraints, of course, in that if everything is trivial players get very bored, accomplishments are meaningless and anyone who is skilled and persistent will easily complete every single challenge the game has in short order.  This may well be fun for awhile but it will not lead to people continuing to play and that is failure for an MMO.  If everything is very hard then players will have to practice constantly on everything and play over and over to learn how to beat each challenge.  Again, this will be a huge problem for subscriber numbers because there are huge numbers of players who want to log on for a short time and do something useful and interesting - if everything requires a ton of practice and skill they can accomplish nothing if they log on for only an hour and they will quit.  Given these extremes it is clear the best route is a mix of trivial, medium and challenging tasks.  This allows players with little time to do something, and in particular to choose a challenge based on their mood.  It also means that no matter how skilled a player is (or isn't!) they can find something that pushes their limits.

Great, so we need a mix of difficulties.  However, here is another stumbling block:  It is very hard to design very challenging tasks.  Testing and tweaking and debugging really hard things is very finicky and good players will beat tasks drastically more quickly than they can be designed.  If it takes developers a month to design a single challenge like a dungeon and good players take 3 hours to beat it then what exactly do the players do while waiting for the next challenge?  The answer of course is incremental improvements.  WOW certainly does this, though it by no means discovered it.  Doing a dungeon over and over allows a character to get more gear from the dungeon, get more reputation with a faction and get points towards even better gear.  This means that running the same challenge over and over, well beyond the point where it was actually hard, becomes useful to the player and keeps them interested.  This has a couple of effects, firstly it means that trivial content can actually be used to keep players busy and secondly that players will feel a consistent sense of accomplishment when they use these incremental advantages to defeat challenges that were previously beyond their skill.  Giving players incremental advantages for repeating challenges, even when those challenges are extremely easy, keeps people busy, keeps the community full, and gives people who enjoy the game a feeling of accomplishment.  This keeps subscriber numbers up.  Letting players slowly improve their power through easy tasks and thus overcome greater and greater challenges gives them something to look forward to.  They will not have to face the fact that they have reached the limit of their skill and can go no further; just a little more work and preparation and a new challenge will be defeated.

So if an MMO wanted to eliminate the 'grind' they would have to completely remove the incremental benefits of doing things more than once.  If you give people benefits for doing things over and over they will do them and that then complain about the 'grind' that is mandatory.  However, any MMO that does this will have to face down the problems outlined above:  Players will run out of things to do extremely quickly, having beaten every challenge, and players will run into brick walls, challenges they simply cannot defeat.  Both of these things are disastrous for a business model that keeps people playing and paying for years at a stretch.  Much as people whine about the grind I don't think they really understand why it is there and why it is necessary for this type of game.  I should note that it is entirely possible to build a MMO without a grind at all, but it almost certainly has to be a purely pvp game.  The sort of game that most MMOs are, which focus on challenges for characters from the game engine rather than other players simply has to have some kind of grind to keep the players playing.

Tuesday, February 22, 2011

Little bugs in the code

This week I found a few very amusing bugs in the code in WOW.  Both relate to Divine Storm (DS), a spell which caused much controversy during WotLK, in particular because of the glyph.  DS heals for 25% of the damage it causes and the glyph increased that to 40%.  My guildies and the other ret paladins on EJ had many fights about whether or not that glyph was worth using over a really weak dps glyph.  This discussion seemed like it had been buried forever when 4.03 came out and the glyph of DS was removed and DS was changed from a core part of the ret rotation to a bad AOE spell that wasn't even worth speccing into most likely.

But not so fast!  Tier 10 had the benefit that it reset the cooldown on DS (proc on autoattack) back in wrath but this benefit was clearly not going to work when DS was fundamentally changed and removed from the rotation.  The T10 benefit was changed to provide 5% more damage instead but they didn't bother removing the old benefit!  This may have made sense at the time because DS was originally changed to be a finisher and to not have a cooldown at all but shortly after that they changed DS again to its current form:  A replacement for Crusader Strike.  It seems like the developers didn't bother to remove the old DS effect because it would not affect the new DS but they made that decision and then never revisited it after iterating more on DS.

This is compounded with the fact that the old glyph of DS was removed and replaced with the glyph of Templar's Verdict.  (TV)  They didn't remove the old glyph though, but rather just changed it to have the new effect but again forgot to remove the old.  So now every ret paladin who is properly glyphed has DS healing for 40% of the damage it causes - we can't avoid using this bug to our benefit without compromising our basic abilities in other situations.

The reason all this matters is because DS got monstrously buffed in the last patch.  It went from 80% of weapon damage to 100% of weapon damage + 42% more magical damage.  It turns out that a 77% boost in effectiveness changes DS from a joke to a powerhouse.  In a raid situation DS will hit for 10k per target and heal for 2700 per target under normal buff conditions.  There aren't a lot of AOE situations I get to be part of but whenever I do get to hit 10 guys with DS I have a instant cast spell that does 100k and heals for 27k.  That spell also has a 3.74 sec cooldown that is very regularly reset by autoattacking and drops to 3 seconds under heroism.  This is *not* a reasonable situation.  On heroic Halfus I was able to put out as much healing as the healers did while the whelps were up (which is the critical time) and do absolutely devastating dps.  The nutty thing is when a fight happens to work such that all cooldowns including heroism are popped at once.  This drops the effective cooldown of DS to about 2.5 sec and because the healing effect from it double dips with my AW I can easily expect to be healing for ~50k on each cast and doing 150k damage.  I don't think that a dpser delivering 60k dps and 20k healing/sec with just *one* of their abilities during a AOE burn phase is at all a reasonable thing.  Whether or not Blizzard is going to fix these bugs is unclear since they are really pretty rarely used but their power when they are called up is just over the top.

Thursday, February 17, 2011

Kicking people out for profit

In Cataclysm the queues for dps classes using the dungeon finder tool have been very large.  There are a couple of reasons for this, probably including raids needing less than 1 tank per 5 bodies, tanks being forced into a leadership role and people not being comfortable with that and pvp players not having tank as a role but still wanting to do dungeons for gear.  There is also the obvious truth that doing a dungeon with a badly geared tank is quite hard but doing it with 1 badly geared dpser is no problem as they can be easily carried by a strong group.  This means that new players and alts are generally going to shy away from tanking.  Healing doesn't have the same reasons for it being scarce but nonetheless it is - healers have very fast queues and dps languish for 30 mins or more.

Some people have been abusing the LFD tool by queuing up with a friend who is a hybrid class and then having the friend drop group right away.  The person queuing up as dps then gets a free ride into an instant dungeon which makes the queues even that much worse for those who don't abuse the system in this way.  Another trick is queuing up with 3 people and having your hybrid queue as a tank/healer; once you get into the dungeon you votekick the other two people and requeue as triple dps and get the first tank/healer available.  Obviously these tricks are underhanded and nasty but nonethless people did it.  Blizzard came up with some wonderful solutions to these problems just yesterday:

  • Players who are outside a dungeon for more than a few minutes are now immediately able to be kicked.

  • The first one is great as I have had problems a few times with people going afk outside the zone and the group being unable to remove them for quite some time.  Obviously if someone is in the zone and refusing to play you still have issues but this is a nice change.

  • If queuing as a group with a tank or healer, and the tank or healer drops group (or is kicked) soon after joining, those that queued with them will also be removed from the dungeon.

  • The second is simply addressing the exploit I talked about earlier.  It does so pretty decisively so I approve very much.  It seems remotely possible that someone could be spited out like this but they could have been kicked the normal way anyway so it hardly matters.

  • If three or more players group queue with each other it will require an additional vote for them to kick anyone they did not group queue with.

  • The third is great.  It was always a problem that a group of three friends could walk in and kick out everybody else so requiring a fourth person to agree with them is a great change.  I suppose they might end up trying to bribe or convince one of the random folks to go along with some sort of dastardly plan but that seems like a poor gamble.  The ability of 3 dpsers to abuse the system right now is tremendous and this at least gives the people they end up with a way to deal with the situation.

  • If a group queue of 4 kicks the one person that they did not group queue with they will each receive a more severe penalty to their ability to initiate future kicks.

  • If someone initiates a vote kick for someone they group queued with they will not incur a penalty to their ability to initiate future kicks.

  • I am honestly not sure what the last two are intended to accomplish.  If a group of 4 queues up they already get an instant queue since they simply must supply a tank or healer from their own group so kicking people is unlikely to be exploitable.  They can be annoying for no reason but generally there is a need to control behaviour that is beneficial to the jerk but controlling people just being jerks randomly is mostly not necessary and extremely hard to do in any case.  The final one is puzzling as I don't know why they want it to be easy to remove your friends from the group exactly.  I don't have a problem with it but I can't figure out what it is trying to solve.

    All told, thumbs up.  There were some really unpleasant exploits out there that mostly affected the solo dpsers trying to get a dungeon done and these close them up pretty thoroughly.  People can still abuse the dungeon finder to be jackasses but they can't do so profitably anymore, which is the key.

    Tuesday, February 15, 2011

    Crushing some dreams

    Everything should be awesome.

    This is the mantra Blizzard uses when designing talents and I think it is a good one for building any sort of game.  When you have a choice in a game that essentially boils down to being good or being bad players will choose to be good (mostly) and the choice ceases to be one.  This has been my essential guideline in redoing a bunch of the units in the modern era in my CiV mod.  In the early game it is fairly clear that when you first get Horsemen or Swordsmen they are very strong and your opponents will fear them.  In the later game the options are much more widespread though and it is much more difficult.  I must make sure that bombers, fighter planes, submarines, antiair guns, artillery, surface ships, missiles, tanks and basic infantry *all* have this characteristic that as soon as you get them you immediately want to build a bunch of them and go crush some dreams.  I don't entirely understand why I love the phrase 'crush some dreams' so much but it really epitomizes to me that moment when everything looks to be going well in the game and then you see the one thing you really, really didn't want to see and you think 'Oh no, not like this'.  Your dreams get crushed!  I want to deliver that moment to my opponents and I want every unit that can be built to have the feeling that if you get there before the other guy you are going to make these units and crush his dreams.

    It is challenging to do this because the units interact in a lot of complex ways.  Submarines should be good at wrecking ships, ships should be good at controlling waterways for land units to cross and to bombard coastal positions, bombers should be good at smashing cities, fighters and antiair guns should wreck bombers, etc.  There are so many different interactions that it is easy to have just one small thing be wrong and suddenly whole classes of units are utterly rubbish.  In the base game the standard modern infantry unit (Mechanized Infantry) has 4 speed just like a tank.  The trouble is that once tanks aren't faster they are useless since they lack many things an infantry has.  This means that no one should build tanks, so anti tank guns and helicopters are useless.  Artillery are overpowered and are the best solution to breaking down cities so bombers are junk and when bombers are junk there is no reason to build fighters or antiair guns and no use for carriers or other heavy ships to support the bombers.  Even a small miscalculation in this giant mess of dependencies can make nearly all units not worth building and reduce a very complex rock/paper/scissors type matchup to a simple game where everyone builds artillery and infantry and nothing else.

    I think it is the fine balance of making things awesome right away but counterable down the road that is so hard to find.  Submarines should be great as soon as you get them and you should have fun sinking lots of ships (or getting your ships sunk, depending!) but there needs to be an answer.  That answer needs to be decent in and of itself though, otherwise the game gets too swingy depending on if you have the rock to the opponent's scissors.  The trouble with this sort of stuff is it requires tons of testing to get right.  It is really hard to figure out just how high a bomber's attack strength needs to be to make it useful - 40 and it is useless, 60 and it is okay, 80 and it is unstoppable.  You can't figure all this out from looking at a chunk of code, either, you have to actually play it out and see how the game works.  That is the sort of testing that is very hard to fit in with rushed production deadlines of course which is why CiV shipped initially with such incredibly terrible balance in the modern era.  Dozens of units and only two of them were really worth making... such a waste.  I guess they needed to release it when it was done and not before.

    Thursday, February 10, 2011

    Trinket trouble

    Yesterday I got a chance to sort out my new rotation and try out my new trinket.  I do a lot more damage based on the new rotation but the new trinket was a real disappointment.  It managed to illustrate very well the difference between theorycrafting a patchwerk fight and real raid encounters.  The trinket in question:

    Fury of Angerforge:  Melee attacks stack up a buff on you that stacks to 5 and has a duration of 15 seconds.  When it gets to 5 you may remove the buff to give yourself 1926 Strength for 20 seconds (2 minute cooldown).  Not to be ignored is the benefit that I turn into a dragon when I use it and look really cool.

    This seems pretty sweet at first glance.  The 2 minute cooldown and 20 second duration are identical to Avenging Wrath (AW) and Zealotry so I can sync them up to massively boost the power of the trinket.  The problem is that there is an internal cooldown on the trinket stacking up its buff and the proc isn't at all reliable so quite regularly I get it up to a 3 or 4 stack after 30 seconds or so and then it falls off!  Suddenly I have delayed my AW for quite a long time and the trinket won't be ready to use until a full minute into the fight.  Worse, it might not even be ready then as I might need to move or be stunned or whatnot which increases the chance I drop the buff even more.  Certainly last night I was not popping the buff as much as I should as I was trying to sync it up with AW but it ended up with uptime lower than 5%.  AW also ended up with uptime around 13%, which is far lower than it should be due to me trying to get the two to play nicely together.  Even if I simply popped either of them whenever I could and didn't sync them Fury of Angerforge wouldn't be very good because it still would be unreliable.

    The trouble with this sort of thing is that FoA shows up as extremely powerful in a simulation.  Normally simulators just aren't very good at dealing with fight interruptions, movement and target switching and this is no exception; in a simulator the buffs sync up and generate very impressive numbers but in real raiding they require lots of extra babysitting and end up being very lacklustre.  One thing that really shocks me is that this is the buffed version of the trinket... prior to 4.06 it had an even longer cooldown on gaining a stack and was absolutely rubbish instead of simply subpar like it is now.  I am not surprised that Blizzard got the trinket wrong in the first place but noticing that it was bad and then buffing it so that it is still bad just feels sloppy.  Once you know you screwed up in a particular spot it seems worthwhile to take the time to get it right if you are going to do anything at all.

    I also got to test out my new rotation a bit.  Initially I had figured on syncing up Zealotry, AW and FoA all at once for some huge beats but this proved problematic.  Not only did I have all the issues outlined above with FoA but trying to link Zealotry and AW was annoying.  Zealotry is only castable with 3 HP so I often ended up not linking them quite right (or waiting too long on Zealotry) and spent far too much time and brain cycles worrying about it.  I have gone back to my previous style which is to have Zealotry macroed to both my finishers so it goes off without me having to think about it.  This way I maximize uptime and minimize the amount of time I have to think about it. I left AW and GoAK as the cooldowns I actually manage since they are usable anytime and set up my new trinket (which has an uncontrolled proc) to give me a big fat PowerAuras notice so I can punch AW/GoAK when it is up.  Hopefully I can stack on another 1k dps with these changes and lock up first place on the meters.  :)

    Tuesday, February 8, 2011

    18% in big chunks

    In the patch today Blizzard buffed Ret paladins a lot.  Our old mastery effect gave us a chance to get a free finisher from autoattacks, and this had lots of problems with it that I have already gone over; most particularly it is true that adding more attacks to a very full rotation isn't very useful.  The new mastery effect is entirely passive and just makes me do more damage so its scaling is good and the numbers are big enough that it hits very hard.  I worked on updating my spreadsheet for the new mechanics and ended up concluding that today I should do 18% more damage than yesterday.  Part of this is the new metagem changes allowing me to use a 3% crit damage metagem and the exorcism glyph bugfix but the bulk of it is simply mastery changes.

    The crazy thing about this extra damage is that it is in huge chunks.  I went and whacked on a boss dummy for a bit and was getting crits on my TV as high as 78k.  Obviously in pvp people have a lot of resilience, but in pvp I also would have more buffs than the ones I make myself so I would certainly expect to be cracking people for 50k fairly regularly.  Given that I can get several of these in a row I wonder if the burst potential of a ret paladin suddenly got to be just a bit too high.  I also have big droughts with the new system too as they are times when I am building up holy power and get no procs so I really stand there doing nothing for extended periods.  This is fine in pve generally speaking but is a recipe for being excellent in pvp as occasional huge spikes are great for putting people in the ground.

    The other big benefits I got were massively improved target switching and AOE.  Prepatch I would expect about a 11 second ramp up time from engaging before my full damage would be rolling and now I should hit my stride in 3 seconds, as soon as my second autoattack lands in most cases.  This is going to mean I am hugely better at swapping to adds and in fact may completely remove any use for Seal of Righteousness - Seal of Truth simply does so much more damage that even in AOE situations there is no reason to swap.  The AOE news is much less relevant as Divine Storm now does a lot more damage but it is still a very small part of my overall damage.  Gaining 5-10% AOE damage when I can AOE a large number of dudes is nice and all but my damage is still so anemic that I should never *be* AOEing unless it is absolutely necessary.

    Basically I think they got the mechanics right for single target.  Mastery now makes sense and applies itself in a reasonable way and my stat weights look very sensible.  I expect stat scaling to work fine too, so the only thing that is left to worry about is whether or not I do too much or too little damage right now.  Sometimes I was able to top the charts but usually I languished down low so I think 18% will be sufficient to have me competing at the top but not dominating.  I still am hideously bad at AOEing and I don't know if that is a design decision or just something I have to live with.  It may be that decent AOE is something that is going to be reserved for some classes and not others.

    Friday, February 4, 2011


    On my Brightcape blog I got a strange comment yesterday.  It was

    "You people are all <expletive deleted> stupid and your guild sucks."  -Anonymous

    Where does that come from?  It is particularly strange since it was at the end of the Safe Schools post, which is a little bit old at this point.  I deleted it right away though I suppose I could have left it for awhile in light of the fact that I am posting about it.  Clearly the troll who left it is some person who is mad at me for something I did within WOW but it isn't clear at the outset who they might be or why they chose to spew their trash on my blog.  After a little bit of consideration I figured out where it came from.  Note that I have exactly no proof of this but I got trolled by some fool who was mad at me about WOW, today, and who found a link to my blog which is in the post I made in that thread, so I can be pretty confident it was this person Tolur.

    So here is the thing.  I have been a recruiter for a long damn time in WOW.  Back in BC my guild seemed to take tremendous, one might say inordinate, pleasure in watching me roast applicants who posted crappy, slipshod, half-baked applications.  People would come to the guild website and instead of reading the application form and making a reasonable attempt at answering the questions as asked they would ignore the question, talk in all leetspeak, lie, answer complex questions with 'yup' and all manner of other disasters.  I would then politely tear them apart, listing every way in which their character was suboptimally set up and every question which was answered foolishly or wrong.  I think the guild assumed that I was just crushing dreams because it is fun to crush dreams but that is not the real reason I did these things.  To be sure, there were some applications (This one is actually pretty hilariously bad) where the applicant so severely disrespected the application that I really just spent my time roasting them for no reason other than to chastise them in a fashion amusing to me.  Most of the time though my negative criticism had a very specific point to it:  To find a diamond in the rough.

    There are plenty of people out there who really don't know that if you are filling out an application you need to actually do a good job or you won't be accepted.  There are also plenty of people who don't know that there are free resources on the internet that can tell you exactly how to set up your character and how to play the game.  Most of the people that don't know this aren't good raiders but there really are people who aren't aware of these things but could be good additions to a roster.  The key ingredient is tenacity.  When rejected some people just walk away and sulk, some go on the blog of the person rejecting them and post vulgar garbage, and some fix their crap up and try again.  That third type of person is exactly who I want on my raiding team.  Life will regularly kick you in the head; that is unavoidable. Your cannot prevent that situation but you can choose how you respond to it.  Someone who knows what they want and is willing to say "Okay, I was wrong, but I can and will do better."  is exactly the sort of person I want working with me.

    To be sure, the vast majority of the time when someone applies and I list all the things wrong with their application they simply leave, delete their app, insult me, or what have you.  It is rare that I actually get someone playing back at me who shows real moxie and stubbornness.  When I do get those people I feel vindicated; I have found my diamond in the rough.

    Tuesday, February 1, 2011

    Who does what

    Raid leadership is an interesting beast.  In particular figuring out who is going to call out what and who is going to make particular decisions is a tricky sort of business that involves balancing logistical considerations with social considerations in ways that can generate a lot of drama (not to mention wipes) if it is done badly.  For example, when I was leading Hounds of War, the guild I was in back in Classic WOW, I did all the raidleading.  I made all the calls, organized the groups, assigned healing and such.  That is, I did that until my head nearly exploded and then I got better at delegating.  It was really hard to hand off healing assignments to someone else and just hope they got it all right and prevent myself from stepping in to fix things.  Sometimes I am sure my idea of how it should be was actually better than the person who was doing it but many times it was just my perfectionist streak and the real difference was negligible.  Obviously sometimes I must have been wrong but my memory records none of that!

    The fun part about choosing who does what is often figuring out who should actually be entrusted to say things on ventrilo.  Some people are really chatty and regularly sing out with information or warnings but aren't consistent; they might simply not say things at the right time, or take too long to say it, or miss it when things get sticky.  Some people simply aren't interested in giving those warnings at all and will only do so under duress.  The fun part comes in when you end up having someone who is very consistent and effective but who just isn't appropriate for a particular task.  Sometimes you want a dpser assigning targets and calling kills and no one wants to step up so a healer does it and sometimes you have a mage telling the tanks when to swap.  I find very often the ability to communicate effectively over ventrilo does not correspond well to someone's skill at playing otherwise.  I have had people who made great delegates, organized groups and assignments and could report things effectively who just weren't that great at playing.  They were often extremely effective when dead since they no longer had any reason to be distracted personally and could give orders and instructions to the whole raid - in fact I recall at least a few dicey kills that only went off because the guy who died happened to be able to direct the whole raid for the balance of the encounter.  Of course, there are way more people of the opposite variety who play really well but can't communicate effectively at all and are utterly useless once they die.

    We had this sort of situation on our Al'Alkir tries yesterday.  We got him to 140k health, or about .5%.  In phase 2 you have to prioritize killing adds correctly and get the timing on downing them right to maximize your damage and instead of a regular ranged dpser calling it out we had our disc priest, who is also the raid leader, doing so instead.  Granted he was probably smiting them some of the time but it seems we really went with the option of getting someone who is extremely reliable and comfortable on vent and not with someone who should be spending their time watching the boss exclusively.  I figure I could do it since my job in p2 is a real joke difficulty wise, but optimally we would have a ranged dps picking the targets and attacking them personally since that smooths out the time to die considerably.  That isn't to say the choice was wrong, as there are plenty of considerations when choosing a person for that role and none of the ranged were clamoring for the job, but it was an example of how you have to mix the priorities of game role and real life characteristics when choosing a chooser.