IBM MMOG Roundtable Rundown

Plaguelands has up a rundown on the recent IBM MMOG Roundtable, with speakers such as Steven Reid, Raph Koster, and Geoff Heath putting in their two cents on the growing massive industry. Krones is not shy about voicing his opinions as regards the speakers and their effectiveness. From the article: "Continuing on, despite my subjective disagreement, Steven Reid; Directory of Community Relations NCsoft Europe stepped in after Heath and he pretty much spoke general edification about mmo communities. His presentation was average, not up to the quality of articles seen from community specialist Jessica Mulligan, but I believe he is well qualified in doing what he does and has an excellent head on his shoulders. The defining part of his presentation on community building is that community leaders should be local and native from that community. This is crucial for many reasons... including the most important, cultural differences." Also includes links to streaming media of the event.

To summarize

[CableModemSniper]

The author of the article thinks everybody sucks. Except Raph Koster. Yeah he rules. Also he says that die-hard server nerds will "probably eat this shit up" as far as Patty Fry's presentation is concernced. In conclusion, I think everybody should just watch the thing instead of reading this guy's stuff.

Re:To summarize

[Alkaiser]

This is hilarious considering how badly Koster screwed the pooch on SWG.

He basically establishes that his model is a complete real-world failure, and yet, people keep inviting him to speak at places.

Re:To summarize

[Dachannien]

Apparently, people place value on learning from the mistakes of others.

Re:To summarize

[RaphKoster]

I must say, I have always been fascinated by the different behaviors taken by game fans towards game creators than by fans of other media's relationship to entertainers in other media. Nobody says that Stephen King is a retard because he screwed up an alien invasion novel--I mean, how easy a home run is that? Clearly, we should never listen to a word he says again. Besides, he bears sole responsibility for what he did, it's not like there was a team writing the book.

I'm not whining or being defensive here-

Re:To summarize

[Dachannien]

I think part of the reason is because there has historically been a channel between MMOG players and developers. Since MMOGs are generally works in progress, players see an opportunity to take part in the design process. Unfortunately, many of those players are motivated only by enhancing their own ease-of-gameplay, even at the expense of other players. Since that goal is pretty much directly opposed to the goal of balance held by the game designers, some players view it as a battle to be fought (a sort

Re:To summarize

[RaphKoster]

I've made the politician comparison myself many many times, so I don't cringe at it at all. :) That tension is interesting, though. We are, in the end, entertainers as well as politicians. Something like the post above is decrying the entertainment value (I think) not the management as such. At least, that's the way I generally read it. I won't pin blame for failings of SWG on the players, nor will I take credit for its many successes (which posters like the above tend to overlook). What's interesting to

Re:To summarize

[servognome]

In a sort of dramatic irony, part of the reason that some of the more ambitious features of SWG fell flat is because the players, driven by greed and the desire to be first at something,

Damn those players for being human.

They should be blaming themselves for the failures in SWG's design, instead of blaming the designers. It's really the players' fault tht eliminating "the grind" is probably the biggest challenge in MMOGs.

Sounds like Sony logic of "it's a beautiful design, it's the user's fault it does

Re:To summarize

[Dachannien]

Designing these games is a lot like designing the fitness function for a genetic algorithm. You spend some time writing the code, and let it run for a while, only to discover that you have to start over because the GA found a degenerate solution to the fitness function you posed to it.

The difference is that with MMOGs, there's an outlay of millions of dollars and multiple years involved, and just "starting over" isn't really feasible. Since you can never really know for certain whether you got the game d

Re:Handcrafted content

[RaphKoster]

You're missing the point... it's about cost of development, not about the hours alone. The reason why the MMOG content tends to be more basic than the stuff in a single player game is because the manhours spent to develop one hour of content are radically different in the two games. Single-player games can invest more in a given hour of gameplay, basically. In the more polished games like Half-Life 2, there's weeks spent on a single MINUTE of gameplay. MMOs, because of the extended play time that people demand from them, must supply more content, and by necessity it gets stretched thin. This is going to get much worse with time--EA has publicly mentioned teams of 200+ for next generation sports titles, and how many hours of content does Madden have, really? In the long run, this is not a good trendline for the industry. Not only will it lead to more overwork and quality of life issues like the ones that came to the forefront this year, but it will also cause budgets to continue to rise. As they rise, you're going to see a few effects:

• the required sales figures for break even may push prices up
• the high sales figures required for break-even will force blockbuster titles to greater conservatism
• more publisher consolidation as those who cannot afford the price tag get swallowed up
• a greater emphasis on hit-driven business, potentially leading to fewer choices for consumers

You will probably also see some other effects:

• More indie shops bypassing publishers altogether. Not everyone will be able to pull off what Valve did with Steam, however.
• The continuation of the flight of developers from console and AAA development and towards casual games and indie games (a major trend in the last few years)

Extrapolate from the film industry if they had to keep reinventing the camera every year, because that sort of Hollywoodish business model is what we're talking about here. The solutions I put forth are hardly unique to me, by the way. They're the underlying point behind what Will is doing with Spore, for example.

Re:Handcrafted content

[vulgrin]

Easy. Build a game engine that lets the players or amateur developers provide the content and / or build a game system that builds itself. (i.e. a game system that just knows the rules behind how it should build itself: physics, evolution, behaviour - ala Spore)

I think in 10 years MMOGs are going to look a lot more like the Matrix (the world exists based on a set of rules and pretty much runs itself) vs. Dungeons and Dragons (all of the rules for everything are spelled out in detail and every construct

Re:Handcrafted content

[Elsebet]

I can agree with this. Look at the success of Neverwinter Nights. Look at all of of the homegrown maps/mods for Unreal, Half Life, etc. Gamers are bending over backwards to create their own content for a company's base game.

I honestly hope to see an MMORPG which will allow this, with proper bounds checking obviously. There is a concern of allowing items/money into the persistent game world from user created modifications. Player mods could start in a sandbox area then move into the developer's own QA

Re:Handcrafted content

[Teh Suq]

The Saga of Ryzom is working on just that sort of thing.

Check it out here: http://www.ryzom-ring.com/ [ryzom-ring.com]

Pathfinding

[ockegheim]

On the tech aspect, something I didn't know Koster mentions is that 40% of the cpu processing is utilized on pathfinding. Yes, fuckin' pathfinding. A fuckin' decade, and almost half of the potential processing powers developers are allocated is used to fuckin' pathfinding. And you know what? Pathfinding is a joke, it could use a lot of work.

I suspect a lot of work has already been done on pathfinding (the optimum legal way for a monster to get from A to B?). An algorithm that delivered a performance bo

Re:Pathfinding

[RaphKoster]

Everybody basically uses A* or variants thereof. The issue is that the environments have gotten more complex (bigger and more detailed graphs to search), the behaviors demanded by consumers have gotten more complex, and there's just plain more AIs to run. BTW, the comment was largely MMO specific. As such, the pathfinding is happening on the server, not on the client computers. You buying a faster computer won't help much. ;)

Re:Pathfinding

[Dachannien]

How slowly can the pathfinding be updated and still keep things believable for the player? That seems like an easy way to reduce server load. Are people using other optimization techniques as well, like only searching the local portions of the graph, bounding obstacles with large bounding polygons having few vertices, and such?

Re:Pathfinding

[RaphKoster]

The MMO servers I have seen run anywhere from 100ms to 250ms frame times. They all amortize CPU load across multiple frames, of course, to keep player responsiveness as high as possible. So as load increases, AI is usually the first thing to get deprioritized. 250ms seems to be the slowest you really want to run an MMO server at.

All the optimizations you cited are pretty standard. In addition, there's a lot of preset paths, use of client-side steering (in other words, only making it LOOK like the creature