Sunday, October 23, 2011

Talents, A Fair Approach II

It baffles me that the developers apparently consider having to look up boss strategies a no brainer, but having to look up speccs worth revamping the talents every other expansion.

Please read the post below first to be able to understand this one.


The conclusion of the earlier post was that if you want to have meaningful choices of equal efficiency in a game, they need to be about style. Examples show that differences in style need to be very strong if the choice between equally efficient options is to be considered meaningful.

If we try to use this insight to design WoW talents, we run into several problems.

(1)
On the one hand side we would like to start the iteration during development with radically different options. On the other hand, we need to consider that having six choices of three options each, creates three to the power of six possible combinations. That is 729 combinations. But since we have 33 speccs in World of Warcraft that makes 24,057 combinations that the designers need to think through if they want to make sure that no unbalanced combination emerges later on.

Of course, the designers don't want to think through every one of these combinations - especially not when considering possible combinations with racial abilities or any content they offer later on. That's why they want the options to be similar! That's why e.g. the first choice in the feral talents at level 15 is all about movement. This way it is easier to think it through. Situations in which movement doesn't matter, for example, can be ignored while balancing this choice.

Another reason to create choices between similar options, instead of radically different ones, is to make sure that every druid has some 'movement changing ability'. Otherwise the player might feel like he missed out on some of the 'content'.

(2)
To keep the balancing manageable it is also useful to homogenize the endgame. I think this is a fallacy, but hear me out. If every boss has an autohit every two seconds and at most one special hit every 3 seconds, some number crunching can be done. For example, the devs can offer the tank the choice to either gain X armor or Y more block chance. Next, they tweak X and Y to be equally efficient while tanking. Of course this only works if encounters are similar enough.

We see that having a very homogenous endgame allows the devs to add more style choices that are equally efficient. If you ever wondered, why the PvE endgame is better balanced than the PvP endgame, that's the reason.

(3)
Unfortunately homogenizing the endgame is a fallacy. What you gain via the more interesting talent choices, you lose threefold by having a more boring endgame.
I recently played a rogue in dramatically unbalanced battlegrounds; and I had fun. And since the waiting times were lower than at max level, I'm apparently not the only person who found this to be fun. How is that possible?

Well, the PvP game consists of a lot of very different situations that appear unpredictably in front of the player. The priest may be instantly dead when jumped by two rogues, but he can also make a fear bomb in the flag room and save the day. That's why the game is fun for him. As long as we don't make a competitive sport out of it, these battlegrounds are perfect fun.

And turning an activity into a competitive sport may not always be the best idea. If it has to be done, the game either requires a lot of unpredictability (Poker) or perfect symmetry (Chess). Unfortunately Blizzard was trying to make the WoW PvE endgame perfectly predictable, not symmetrical (it's PvE) and still competitive. We've seen the boring result of this: the endgame becomes all about execution, “Simon Says”, without any real choices for 90% of the players.

A solution would be to make the endgame less predictable. But that would be a dramatic change that's probably impossible to make. On the other hand .. they just added Panda Kung Fu and Pokemon; nothing's impossible, I guess.

(4)
Looking at the preliminary Druid talents, for example, you can see that they are mostly completely irrelevant for the raiding endgame. And where they are not irrelevant, they are mandatory. If “cenarion ward” increases healing output, it is mandatory. If it doesn't take Nature's Swiftness.

It is extremely difficult to create style choices in the raiding endgame, because endgame players, especially hardcore raiders, just don't care about style. Just recently someone told me that he could care less whether the tank is a Panda or a Pink Gnome. To introduce a meaningful choice about style in that environment is all but impossible. And if a choice is not about style and not about efficiency it is meaningless.

Remember, the act of choosing is a journey towards a goal. That goal can either be more efficiency or better style. There just aren't any other goals concerning character development that I am aware of.

(5)
So, what would I do if I a were a dev? Two things.

First I would differentiate between two kinds of choices: The ones that are meant to make the player ponder for a long time whether he has made the right choice, and those which do not. Next, the choices that are meant to make the player ponder for a long time, need to be very high level choices, like choice of class or specc. They need to change dramatically how a player plays the game. They need to change the style dramatically. They can either lock the player in (e.g. class) or allow a frequent repeat of the choice (e.g. specc).

The choices that have a final answer would still be left in the game. They are fun for new players to figure out and if new players are unable to figure them out, they will receive help from other players and online resources. It baffles me that the developers apparently consider having to look up boss strategies a no brainer, but having to look up speccs worth revamping the talents every other expansion. The devs create an entire leveling game that is wasted once the players reach the endgame. But if the talents become uninteresting for players in the endgame, they somehow think that's a big problem.

Finally, I would add more unpredictable content that makes all kinds of absurd decisions valuable every now and then. This situation already exists in the casual PvP content. In the long term, there's no reason to not evolve the PvE content in this direction, too.

17 comments:

  1. Re:unpredictability: They might go in that direction for PVE Scenarios. Some randomness - which side the wave gets from, what set of units comes with each wave, maybe variable mini-bosses - will ensure much higher longevity for fairly minor development cost.

    ReplyDelete
  2. I agree, Shalcker. But keep in mind that randomness is the cheapest way to introduce unpredictability and often makes a game frustrating.

    Players don't have a problem if another player does someting unexpected. But if a boss uses a random number generator to decide whether to be difficult or easy today, that's frustrating /annoying.

    ReplyDelete
  3. Example in case: Rhyolith HM.

    The random placement of volcanos + the tight requirements for a HM kill (pre-nerf, at least) meant that thing could go fantastically well or horribly bad depending on the volcano placement.

    Usually, when you work on a boss, subsequent tries go better with time, but there it could be from 2% to 80% Hp left on the nest try just depending on how bad the volcanoes were situated. The alternative would be to lower the requirements, but then it becomes a faceroll for any hardcore player. Or increase the role of randomness, and then it's rolling dice and not fighting a boss.

    ReplyDelete
  4. Why are you still blogging about WoW if you've deleted all your characters and quit the game?

    You seem to care an awful lot about a game you no longer want to play. Just curious.

    ReplyDelete
  5. 'Hell No', I would absolutely agree that blogging about a game that one hasn't played for several months is a mistake. I critizized bloggers before who did this.

    But I see no reason to not blog about a game I played just a few days ago, really. I still have enough knowledge to do that, I think.

    Besides, just because I quit WoW doesn't mean that I don't care. If I didn't care I wouldn't make a rage-quit video in the first place.

    ReplyDelete
  6. Well you've quit the game, so whatever happens to it now has no impact on you whatsoever, yet here you are still complaining about how it's being handled and prophesising doom and gloom before any real meaty information about the expansion is available.

    I'm curious, if they turned around next week and said "Ok, we were wrong, MoP is cancelled, so enjoy 2 more years of Cataclysm while we develop a new expansion" would you be happier? Would you resubscribe and shout from the rooftops of Orgrimmar about how awesome it is that we aren't getting any stupid lame pandas and pokemon anymore?

    ReplyDelete
  7. Well you've quit the game, so whatever happens to it now has no impact on you whatsoever

    This is not true. I have a long-running MMO blog and generally care about this business. Whatever Blizzard does to WoW has an impact on me even if I don't play it.

    ---
    yet here you are still complaining about how it's being handled and prophesising doom and gloom

    I disagree. In fact I have already made a post in which I pointed out what I like about MoP. E.g. more focus on the open world, more endgame activities, no flying while level, etc.
    I think you are a victim of selective reading here.

    ---
    I'm curious, if they turned around next week and said "Ok, we were wrong, MoP is cancelled, so enjoy 2 more years of Cataclysm while we develop a new expansion" would you be happier?

    Hardly. I consider MoP worse than Cataclysm, but that doesn't mean that I would be happy for two more years of Cataclysm.
    This is like asking me whether I would be happy if I only lost one leg instead of two. I would prefer the one over the other; but I wouldn't be happy about it.

    ---
    Would you resubscribe and shout from the rooftops of Orgrimmar about how awesome it is that we aren't getting any stupid lame pandas and pokemon anymore?

    Yes, I probably would - and I would enjoy it :)

    ReplyDelete
  8. Hence why I said Happier instead of Happy, looks like I'm not the only person with the power of selective reading :)

    Anyway, your post about the good points of MoP consisted mostly of your snarky comments about how you've been saying these things all along, or how Blizzard isn't admitting it made mistakes, and how even if they do make (x) positive change they'll ruin it anyway with (y). It was hardly a positive article, you have to accept that.

    ReplyDelete
  9. I was not trying to make an objective post. I was not writing for the New York Times.

    I said that I like some things about MoP, but my feelings were clearly visible in that post. And that's not unusual for blogs, I think.

    ReplyDelete
  10. Looking up a boss strategy is much easier then looking up specs.

    First, boss strategies differ from raid to raid. They are mostly the same but every raid does something a little bit different. But there is only one single correct "spec of the month".

    Just go tou youtube and search for your boss and you'll find a sufficient accurate description for every normal mode boss. Watch that movie and you're sufficiently prepared.

    Now show me the page where you can look up skills. Leave out the pages where you need a degree in statistics to find the relevant information on page 7, 12 and 47 of a thread. You get a bonus point if that page also explains the rotation which corresponds to that skill. I'm quite sure such a page doesn't exist.

    ReplyDelete
  11. I disagree, Kring. I used to go to EJ and have a look at the forum of my class. There's always some kind of specc compendium, e.g.

    http://elitistjerks.com/f75/t110326-cataclysm_fire_mage_compendium/#Speccing_a_Fire_Mage

    There you search for 'talents' or 'specc' and voila.

    http://www.wowhead.com/talent#o0hZfchrkbRRsfoc

    Of course, it's even easier to just have a look at some mage who does good dps or even ask someone (careful: social interaction is not supported by WoW)

    .. sorry I failed to surpress the brackets ;)

    ReplyDelete
  12. The point that I think most people miss regarding "Simon Says" endgame, aka The Dance, is that the problem isn't about The Dance per se (it's always been around). It is about not being able to choose who dances.

    The shining example is my guild's first Mimiron kill.

    In today's encounters, not only will your raid wipe pretty much the first time someone dies, but you also don't get to choose who gets the hardest part of the encounter. Defile is the perfect example of modern raid design - if your weakest member gets targeted, pretty much a wipe. In my 10m raid, there were 2-3 people who were fine at their rotation/healing, but they couldn't be trusted with anything else. If I cannot put them safely in a corner while the more pro players handle the tough stuff, then we may as well not raid; those encounters are no longer being designed for us.

    There is nothing wrong about perfectly predictable raid encounters. It comes down to designing them such that things can go wrong mid-attempt and yet success can still be snatched from the jaws of defeat.

    ReplyDelete
  13. I agree that this is a wrong design for non-hardmodes, Azuriel. But I don't agree that this is the only problem. I absolutely consider the fact that the challenge is in the execution alone a big problem.

    That doesn't mnean I didn't like raiding. But raiding would certainly be a lot better if it was less predictable. That doesn't mean that I want more PvP enceounters like in the WotLK colosseum - they weren't good encounters for several reasons. It also doesn't mean that I want bosses throw dice to determine what they are gonna do today.

    But there are ways to create interesting non-static encounters. PvE scenarios might be a tiny frist step, by the way. I have a post about that in the queue, actually :)

    ReplyDelete
  14. The first problem is that EJ is only English but WoW was released in many languages. Most people are not willing or able to look up video game information in their non-native language. That's exactly the reason why game companies make translated versions of their game.

    And even if they are willing to read up in a foreign language they will not understand the guide because every talent and spell has a different name. Often they aren't even translated but just got a similar name and even if they are translated, school-English is not enough to translate these fantasy terms. They might be able to Ctrl+F for the wowhead link, if they already know to look for a wowhead link, and compare icons on the talent guide.

    Youtube has tons of non-english guides for every boss.

    The next problem is that EJ "doesn't write guides". There might be a useful guide for mages but the mage section was always one of the best sections in EJ. Try to find information about tanking on EJ and you'll fail. There you have to go to tankspot or maintankadin.

    -----

    When you watch a youtube video you know that it's correct. Every boss tactic did work to kill the boss and is therefore correct. But everyone can put a crappy talent spec into the Internet. How is a "casually invested player" able to know that EJ is the correct source? The loading screen doesn't mention that.

    -----

    When I played the Witcher I wasn't able to find a talent guide and decided that the game must be beatable with every spec, which it was. I assume many people in WoW take the same assumption.

    I had problems beating the guardian boss in Witcher 2. Guess what, Youtube had a video on how to beat him.

    -----

    > Of course, it's even easier to just have a look
    > at some mage who does good dps or even ask someone

    Now, how could the player not knowing EJ know which mage does good dps? Benchmarking is one of the most difficult tasks.

    -----

    Maybe you should be able to create a [Scroll of Nils' Fire Mage] out of your spec and put it on the AH. At least that would put this part of the game back into the game.

    Actually that would be cool. :)

    [Scroll of Fire Mage]
    Made by Nils
    Used to defeat:
    - Ragnaros 25 HM
    - ...
    - ...

    ReplyDelete
  15. I think they do try to design talent trees so that people can pick abilities that look fun and end up with a serviceable character without looking things up.

    But then they design endgame content that requires minmaxing. And the challenge dungeons will be no better.

    ReplyDelete
  16. That scroll is not even a bad idea for WoW, I think, Kring :)

    On speccs vs boss strategies:

    I think you underestimate the experience and knowledge that is required to properly understand a boss strategy or video on youtube.

    ReplyDelete
  17. I don't think that's related to the videos but to the fact that the dance today is complicated. I was mentioning that boss guides are easy to find (everyone knows Youtube) but skill guides are hard to find. What percentage of player would you expect to know what Elitistjerks is on a German WoW realm? I would guess less than 10%.

    ReplyDelete