Tuesday, May 03, 2011

Optimality

Nils and Tobold are talking about inefficiency and the fear of being sub-optimal. Both of them seem to feel that the drive to optimize is negative.

I disagree.

For a decision to be meaningful, there must be a right choice and a wrong choice. Taking the right choice makes it easier to be successful. Taking the wrong choice makes it harder.

Players optimize because there is a chance of failure, and no one wants to be the guy who is deliberately handicapping the rest of the team. The game is already hard enough. Why make it harder on the rest of your group?

Trying to stop optimization is futile. Removing the ability to inspect or dps meters just means that you don't get any feedback and can't see what's going wrong.

You can remove the decisions entirely. Remove talent points and different stats on armor. No more optimization. But that makes the game less interesting.

You can make the game easier. If there's no chance of failure, there's no point in optimizing.

You could make content less predictable. But this is no guarantee of optimizing. PvP still has optimal builds and, more egregiously, optimal group compositions. It just changes the focus on optimizing to be the most damage for the most situations.

And I don't think that PvE players like unpredictable content as you think they would. Faction Champs in ToC and Prince Malchezzar in Karazhan had unpredictable elements, and there was a lot of grumbling about those.

As well, unpredictable content can still be optimized. Think of Blackjack or Poker. It's unpredictable. Yet there are still strategies and optimizations for those games. Only now you have to take probability into account, which makes the optimizations far harder to execute.

You could rotate content such that the "optimal" build in Fight 1 becomes sub-optimal in Fight 2. Which is promptly followed by the playerbase replacing the players from Fight 1 with different players for Fight 2.

In reality, I think that most people who are against optimizing aren't really against optimizing per se. Rather, they are against other people optimizing for them. If they were the ones to come up with the "optimal" build and reveal it to the world, then they would be happy.

But it just doesn't work that way. My rule of thumb is to assume that there are people who are ten times better at this game than I am. And there are people who are ten times worse. The better people are going to come up with optimizations faster than I will. The worse people will not see what I see, and just perform at a much lower level, making grouping with them destined to failure.

For any meaningful decision in a game, someone will determine the "right" answer sooner or later. The good players in the game will figure it out pretty quickly. The bad players will never figure it out.

The best an MMO company can do is make sure that each "playstyle" is close enough to the optimal so that it is viable. But otherwise, it is impossible to have meaningful decisions in group play, and yet not have optimizations.

18 comments:

spinksville said...

I think the real issue is that people take the minmaxing out of all proportion. If your group is failing on a boss, it's not necessarily because you don't have the minmaxed possible spec and gear, it might easily be execution. But people don't seem aware of the idea of 'good enough' gear/spec. For them it's either theorycrafted minmaxed or else it's rubbish.

But (for example) you don't need minmaxed raid specs or rotations or full top level enchants to beat heroics (not talking about the new ones). You also don't need dps who do 15k. This won't stop players from complaining if their companions don't have them.

I'd compare with LOTRO where however rubbish I may or may not be on my main (and I'm pretty rubbish), it never stopped us killing bosses in raids and no one ever asked me to optimise further because the raid group in general is good enough. And also, no one uses damage meters, so utility and execution sides of the fights are more emphasised.

typhoonandrew said...

I fully agree that we (or I at least) play in a manner which is optimised toward my goals, and that optimal is desireable. However some of my goals are "social" goals which would be viewed as wasteful by Gevlon and some of the other folks who dislike the peripheral game activities. Progression raiding should be optimal...always.

The "unpredictable content" is however still frustrating. There is a significant difference between an encounter which has rules and guides, scripted actions; vs an encounter that has so many random events in layers that its just a mess. Blackjack is random, not chaotic - it has a ruleset which it must apply, and by consistent application a player can develop a strategy.

Prince in Kara was borderline, but the Infernals(!) sometimes made the encounter impossible - and that is a design mistake. The challenge should be rational for PvE, else we'd embrace the chaos and play against other players.

I didn't like Faction Champs in ToC, but it was still an encounter where a combination of strategies could be developed; therefore acceptable. It took a lot longer to get used to, a lot longer to skill up for if you had not done pvp, but players were able to learn.

In this discussion about optimal performance, where is the place of forgiveness?

Is it acceptable for a player to make a single mistake in a boss encounter? Miss an interrupt. Not CC fast enough. Not taunt, etc. In 10 man it seems there is less forgiveness than 25 man, and therefore the players must be closer to the ideal optimal performance to succeed. Is that the correct way to design encounters?

Azuriel said...

I agree that railing against optimization in videogames is dumb, for all the reasons you listed.

A scenario in which Tobold would be correct, however, is in good ole fashioned D&D. The pejoratives for optimizers are all there: munchkin, min/maxer, rules lawyer, etc. "Optimization" was bad in there because the point was not to blow through encounters the fastest possible way - indeed, the DM's job was to generally scale encounters to always challenge you regardless of power - but rather to X... where X is socialize with friends, roll some dice, kill some time, roleplay, etc.

With MMOs though, it does come down to efficiency, by design. You are being metered by month. I don't know about you, but in single-player games (or multiplayer games without a subscription) I try to stretch out the enjoyment as much as possible, perhaps by deliberately not optimizing, looking up FAQS/guides, etc. Meanwhile, in WoW I know that it will take X number of Valor Points to get Y gear, meaning Z number of hours for dungeon runs, etc.

It may simply be a psychological thing. If I make a one-time purchase of a Zelda game for $40 and play for 80 hours, that would be $0.50/hour. Meanwhile, playing 80 hours of WoW in a month would come out at around $0.18/hour. And yet I would feel more freedom to explore in Zelda than WoW.

Then again, everything in WoW is designed around planned obsolescence. If I took a break for 2-4 months, it's unlikely any of the work I put into gearing beforehand would allow me to progress once I returned - I would need to re-grind those items to be a useful contributor with my peers.

Christian Riesen said...

Azuriel has breached the actual underlying cause of it, which everyone so far has ignored by large. The actual reason for optimizing is a simple desire to get the most out of it. If that is only possible by optimizing it to hell, then that's what players do. Personally, I call this a design flaw, others might find it the perfect game. If you are pushed to optimize because otherwise you can't keep up, or if you take a break you would be completely lost, then that is bad design, pure and simple.

Yes it might keep players playing, just as they keep playing Farmville and many other things. But it feels to me more like another Job than actually having fun.

Kring said...

> For a decision to be meaningful, there must be a
> right choice and a wrong choice

I couldn't disagree more. You only have a real choice if all choices are valid.

It's like democracy. Just because you're allowed to vote doesn't mean you also have a choice. You only have a choice if there are multiple good and relevant parties.

Back to WoW. It isn't a relevant nor a fun choice to pick your talents as mentioned on EJ. The only choice you take is to follow a guide.

But there are choices left in the game:
- Do you level a paladin healer or priest healer?
- Do you level a mage or a warlock?
- Do you play your druid as tank, healer or DD?

Those are good choices BECAUSE they don't matter. Neither a paladin nor a priest is "the right choice" or "the wrong choice". It's a matter of taste.

We need more irrelevant choices! You shouldn't decide between destruction, affliction or demonology based on the Simulationcraft thread. You should decide based on which play style you like. And to get a real choice, all choices must be right.

Nils said...

I agree with you and actually made a post about it even before having read yours ;)

Shintar said...

Nils linked it again in his latest post, but in case you haven't seen it, this Extra Credits video basically elaborates on what Kring said above: Situations where there is a clearly right and a clearly wrong solution are calculations, not choices. For a choice to actually be a choice, it's got to make you choose between things where no option is clearly superior - or between different things that you both want.

Yaggle said...

There is something that I don't think you are including in your equation. People being able to inspect and look at you on armory certainly does give you better feedback and better chance of meaningful feedback to improve your chance of success EXCEPT that end-game encounters are created and modified. End-game is made harder because players are better, because they got this good feedback. If armory and inspecting never existed, average player's ability from equipment alone would be lower, and encounters would be made less challenging also. At first, you may say, "Well that's good!", but I say, it's not good, it's all relative, and all the same EXCEPT that everybody spends more time looking at the armory, inspecting each other, spending extra gold on changing gems and enchants, and reforging, to reach the heightened "optimum", and then having the exact relative experience. It's not good, it's a meaningless, relative treadmill that we don't need to be on. It also fails to reward the outstanding player who can figure out all the optimization of gear according to their team's goals WITHOUT feedback from gearwhore players.

luro said...

Something I think we are missing in the discussion is the fact that game designers cater to the min/max-ers. All the damage meters, all boss spells that are data-mined, all the underlying numbers that players are able to calculate are given to us by the game's developers. As long as the player base is given these things to analyze, there will always be the group that theorycrafters for fun. The group that tries to munchkin the hell out of their spec simply to be the best. I don't think we can fault the community for this "epidemic" without faulting the game developers more for letting it happen.

sam said...

I think you missed the point. In a game where it is not unlikely to be booted from a group because one talent point is in the wrong place the Devs have propped up and supported the "cult" of efficiency.

I suspect if they'd remove the armory at least half the complaining and angst would go away. Of course raid leaders and guild masters would scream but I think they'd be surprised to find out that without that tool they'd end up with better players because they'd have to pay attention like we did in vanilla.

Hyperian said...

The Infernals in kara..... (flashbacks begin) man i hate those things.

Oestrus said...

One thing that always upset or annoyed me is when people confuse "optimal" with "possible." Because something is not optimal it's seen as not being possible or not worth doing, even when you can still do a thing at equal or slightly lower capacity.

For example, I'm a holy priest and I love to tank heal. I've done it on various 25 mans and 10 mans and I have done it a few times on H Halfus. I do it well and feel that I can do it and it can be done, by priests at large. I feel that we have been given the tools to do it, with our spells and talents and there is no reason why we shouldn't.

However, if you take a look at various priest resources and specifically Elitist Jerks, there is a vast concensus and a very vocal majority that feels because discipline is better or stronger at it that holy has absolutely no place tank healing. I couldn't disagree more. The argument that people toss around a lot for this is that it's not optimal, so it shouldn't be done. I don't feel the two go hand in hand and it makes me sad that people feel that way.

I guess I see nothing wrong with the concept of something being optimal, but not to the extent where nothing else will do or to the extent where people refuse to entertain the notion of something else being worthwhile.

Masterlooter said...

It sounds like Rohan and Nils believe that the end goal is to make sure your gear is setup correctly. I believe your gear is just a tool to help you complete your goal.

When fighting the Lich King, or Nefarion, or whomever else, they could care less about which gems you have in your gear, and exactly what talents you have. All that matters is who dies first.

If you have the "optimal" setup, and I have a "sub-optimal" one, but I do more damage than you (with equal gear), then I am more valuable to the raid. But since my setup is considered to be "inferior", I may not even get an invite to the group in the first place.

Additionally, nearly any failure in a raid (read: death, wipe) is not caused by gear in the overwhelming majority of cases. Think about what caused the last wipes your group had. Standing in fire, not dispelling, not picking up adds, failure to "dance" correctly. None of these are solved by gear nor talents, thus making your gear and talents meaningless for their ability to help prevent a wipe.

Judge players by their performance - their ability to act or react to what can actually cause a failure (wipe) in a raid(crazy, idea I know) - including their DPS or healing output.

Quit perpetuating arbitraty definitions of skill.

Stubborn said...

Well, someone beat me to it, but I think you're wrong about the meaningful choices. Most choices are far more complex than just right and wrong, and meaning can come from the variances between the types of right and the shades of wrong. I'm no big proponent of "gray areas" arguments, and I like moral certainty as much as the next guy, but if there's a "right" and a "wrong" choice, then really there's no choice at all, whereas if there's choices that benefit X at the cost of Y and choices that benefit Y at the expense of X, well, that's when the cost-benefit analysis comes in, and it's that which makes a choice "meaningful."

Take smoking for example; it's clear that choosing to smoke is the wrong choice for your health, yet many people do it anyway. These people aren't just flat stupid; some are incredibly intelligent, but they've simply analyzed the cost and the benefits and come to a conclusion that doctors disagree with. That doesn't make the doctors wrong or that they think the doctors are wrong, it's just them making a choice that's "wrong" in the eyes of a group with different priorities (health). The benefits of anyone's particular choice may be hidden from others, but that doesn't mean there aren't any benefits.

At any rate, when it comes to WoW, a person's priorities are what determine the benefit in any cost-benefit analysis. Just because some don't want to hear about them doesn't mean there aren't other priorities than raiding. That doesn't make these priorities any "healthier" than smoking, but the people who choose them do so because of perceived benefits (which is true of raiding, too), and no amount of discussion is going to change what they like. You might as well try to argue the spots off a leopard.

Nor does any of this mean that optimization is "bad." For a majority of players (admittedly, I'm only guessing it's a majority, but I think it's a safe guess), optimization is a very positive element of the game, and their priorities shouldn't be derided, either.

As Spinks said, I think it's the imposition of one group's priorities onto others that creates the problem. "Bad players" are only "bad" if they're imposing their un-optimized play on a group of people looking to raid. They're only "bad" if they themselves want to raid but can't due to lack of information, skill, or effort. Someone who doesn't care to raid nor does heroics with groups who wish to raid is in no way a "bad" player. In fact, by their own standards, they may be an excellent player, and if they're not interacting with people of different priorities, then that's the only standards that matter.

Great post, interesting discussion.
Thanks!

Talarian said...

@Masterlooter
If I want a cabinet built, I go looking for a carpenter. If said carpenter is using a flamethrower or a broken saw to build this cabinet, I'll take my business elsewhere.

Gear is just a tool, but using the correct tool and ensuring the tool is in good condition are important to getting a job done.

I agree with you that most wipes aren't caused by a lack of gear, or a really poorly optimized spec, but as a raid leader, I only have so much criteria to judge potential raiders on. If I'm digging around for a PUG player because someone in my raid got sick, if I don't do the appropriate research, then I risk wasting the rest of my raiders' time because the PUGger I recruited could not contribute in a meaningful way.

The only such criteria that I have is looking at the tools of the potential raider. The gear, talents, glyphs, enchants, and gems that player has. Now, one could be "perfectly" optimized and still play poorly, but chances are if someone has a completely off-the-wall spec with wacky (or no) enchants, gems and glyphs, there is a high probability that player doesn't understand their class well enough to contribute, and potentially cause wipes.

If the gearing/spec/whatever is a little sub optimal? Eh, I personally won't care. There's so much to tweak between talents, glyphs, enchants, gems, gear and reforges that I doubt there's one be-all end-all optimal setup (and even if there were, individual skill levels would likely wipe out any such advantage).

If a player wants to try something completely wacky, or they don't want to take the time to put some effort into playing the game better, or their class better, all the power to them. They can play however they like, and they're happy and I'm happy.

But if they're trying to be in my raid, then I expect a certain level of competence and I can only measure that in so many ways without throwing them into the deep end of the pool and watch them sink or swim. When I have 8 other people to consider in my raid, I'm sorry, but I don't have the time nor my raiders' time to do that.

spinksville said...

@Talarian: If I wanted a cabinet I'd go to IKEA. Why do you feel you need a master cabinet maker for every single cabinet? What's wrong with just grabbing a functional one and going with it?

Matt said...

> I disagree.
>
> For a decision to be meaningful, there must be a right choice
> and a wrong choice. Taking the right choice makes it easier
> to be successful. Taking the wrong choice makes it harder.

I agree with you to some degree, but I think you're missing the point. I think it might be easier to think about this if you take a step back and think about games in general. Think about games that have stood the test of time like chess or go. Both of these games support different styles of play. Some players are more aggressive, some prefer to lock down the board, others are clever. Each of these are valid play styles and I would argue that's part of the reason the games are long-lived. They have relatively simple rules that allow for complex interactions that may take a lifetime to master and importantly, there is no one best style.

I believe this is where Ghostcrawler and the team want to take WoW. It's a tall order, but if you listen to the devs comments, they're in-line with this goal. They'd like WoW to be a game that lasts for a very long time. I believe they understand that in order to fufill that goal, they need to create a system that allows for meaningful choices, not simply optimal choices and sub-optimal choices. I believe it's an incredibly tall undertaking, one that I don't know if they'll be successful with, but one that I appreciate them attempting.

Adrian said...

well my two cents fly by, watch out or youll get hit ;)

I think kind of all of you are right. You do have the choice to use the "not-quite-optimum-but-still-doing-the-job"

You really DO have the choice, given your raidlead accepts it. It's all a matter of who you play with. Me as a raidlead myself do expect my guildies to use the optimal specc, the optimal lineup will be chosen if available and all that min-maxing stuff. But hey, that's fine, my guildies know that and they knew when they signed up for raiding. So that's how I like to do the encounters, and if they do not want to do it my way, they're not the players I want to play with.

That doesn't mean I don't respect their decision or think they're stupid for playing BM for example.

They just have other goals than me, and that's fine.

People tend to want all and everything but still don't contribute to it. I've seen soooo many players in my guild that wanted to down tier endbosses, but they refused to even gem or enchant, because they didn't want to. Well it's your decision not to gem, but you wont down the boss with a bunch of non-gemmed, non-enchanted players because it's not designed that way.

The fail really is that they want everything but are not willing to invest what they need to, a general failure in our society I think...