QOTD: I nearly died

"When people tell stories about their greatest moments, they often revolve around nearly dying. In games, what's really special for people is not, 'I killed the bad guy and I was perfect', it's, 'I nearly died, but I just managed to kill the bad guy.' How do we set out specifically to give them those experiences? That's a great challenge for us as game makers."

-- Brian Fleming, via The seduction secrets of video game designers

Guild Wars 2, with their downed, defeated, and rally mechanics, looks to be making a smart move. In contrast WoW's battle rezz (in various forms) seems to have missed the point - you are dead, a red smear on the landscape, and btw quite possibly by some mechanic that wiped out 50% of your health in one hit.

A little idea to toss into the mix - make damage and health work on a reverse logarithmic scale. The big nasty blows at the start of the fight when you're full of pep and bluster will knock 5%, 10% or more off your health, but the same blow when you're nearly dead only knocks off 1% or 2%. It might take 10 hits to get you down to 10%, then another 10 hits to get you down to 1%, and finally 10 more knock down drag 'em out hits to finally put you down. Down, but not defeated of course ;-)

It's all smoke an mirrors, but it'll work in the same way that players like having a 200% bonus for rested XP vs a 50% penalty for unrested XP.

Help players tell their stories

Chris Hecker has a fascinating GDC talk up on his site: "Achievements Considered Harmful?". Its long, but totally worth viewing.

In the Q&A section at the end is a notable question, about the 61:40 mark, which is about how sometimes players will take a video of something really cool they did in the game and post it online, sort of like asking for a verbal reward from other players. The question to Chris was then "is there a way to have players decide when it is time for a reward, and what would be the result of that?"

I find this interesting for a couple of reasons. Firstly and foremost is that this is an area of game development which, in my opinion, has been largely unexplored.

What support is provided in the game? In some you can take screen-caps, some have built in video capture. I forget which FPS it was, but some even provide instant replay video capture - that is, you don't need to actively and consciously turn on the video capture facility, you can instead say "omg, that was cool, what the hell happened" and press the instant replay button. I wouldn't be surprised if there's a game that lets you do an instant replay and show it from multiple camera angles, not just the one you saw.

Linger a while, and hear my story

But if the npc's would talk to each other and talk about events, use turn of phrases and even react to the player inquiring about these things, now that would be something.
They do this to some degree. It's really not too bad. Problem is that there is no character building in the story line. So you don't feel that any of this is important.
-- Nils' MMO Blog

That's an interesting point, and with the accelerated questing and leveling and fast travel and all the leaving zones behind stuff that's becoming the norm ... no wonder players don't bother reading quest texts at all anymore. They just don't get the chance to care for the characters they meet.

I'd prefer a slower paced design, where players stayed for longer in the same place, and experienced more story with the same characters.

Looking back at a few quests in WoW and I see a common pattern: even when there's a bit of story developed between a couple of NPCs, they are nonetheless isolated from other NPCs and other quests. You play through their tiny set piece of story, and then move on to new quests from new NPCs, never to return.

Such a shame, because there's so much more potential there once you let Metcalfe's Law kick in, and you'll get to see more sides of the same characters. Maybe the lovestruck frontier warrior also has a shadowy gambling habit with a traveling goblin trader, is estranged from his father, and is also being groomed for promotion by the local captain. The captain who sends you on the inevitable series of kill-ten-then-repeat quests, also enjoys taking off with that goblin trader to do a bit of quiet fishing, and thinks he's left behind some unfinished business at his previous posting. And that gambling, fishing, traveling goblin trader ... well, best not to ask why he's got such itchy feet.

Then again, maybe I've been watching too many daytime soaps.

Would you cross the street...

Would you cross the street to gain a skill improvement that buffs your key performance metric by 1%?

Would you travel to the other end of the world to gain another 1% improvement?

Would you set off on a journey which would take you several hours of game time to get a 0.1% skill buff?

I'm pondering the design of a skill based mmo set in a massive world, one where there are a small number of common 1% skill buffs, available pretty much everywhere, and then a large number of 0.1% buffs scattered about the world.

A rookie warrior could stay in the one general zone and pick up all the skill training buffs for a total of (say) +50% damage, and then set off into the wide world seeking out 0.1% here, and 0.1% there ... all adding up to another 25% again. It would be like traveling from Paris to Tibet to learn the ancient art of mumble-fu from a secretive clan of Mountain Warrior Monks (0.1% buff to unarmed combat), and then deep into darkest Africa to track down a mysterious Zulu Exiled King to another 0.1% buff to berserker stance combat.

All without teleports or any other fast travel. And even then, the Mountain Warrior Monks want you to prove your dedication and virtue by [killing 10 rats], the Zulu Exiled King generally mistrusts everyone and doesn't give training away for free, and so on.

I would plan the design such that it would take several months to travel the world and gather the full 250 x 0.1% buffs. And all while you're doing this you're not in your home zone contributing to the development of your village/estate/guild/kingdom.

But it wouldn't be necessary to do so. The Big Bads would be balanced around players having only a total of 50% – 51% skill buff. That would be all the common training, plus a dozen or so of the nearest rare skill buffs.

The design intent is that there would be a lonely few insane individuals wandering the dusty roads of the world, perfecting their personal skills; and meanwhile everyone else is putting time and effort into improving the quality of the smithworks, securing high grade ore, and other activities which give players additional bonuses.

Reputation design patterns

These Yahoo Reputation Design patterns are very interesting, particularly the table further down the page. They are specifically for managing reputations within online communities, like forums, but they could also be applied to game design on further thought.

On the community front, this suggests to me that CMs could do well to be proactive in managing their community, and not just reactive to posted comments. Something like a kinda-weekly wrap up of good posts, handing out the plaudits. This would be different from the case of a CM weighing in on an existing thread.

It also puts paid to the idea that you can't design community, that forums are fated to be terrible, and that designers/publishers are not responsible for a lousy community.

Will they storm the caves?

Sure, the soldiers of your faction and the Earthen Ring might still be in trouble, but again, the zone depends too heavily on telling us that our allies are in danger without ever showing us how they are in danger. Do these naga that we need to be afraid of ever storm the caves we get our quests from? Do they send out scouting parties to see if we have become a threat yet?
-- Ardol, WoW Philosophized - “Oh dear, Vashj'ir”

Yup, I had the same sense of non-danger.

What I'd like to see is some kind of counter, showing the number of naga outside, gradually ticking over and growing. Once it hits 100 then a swarm of naga come busting in .. if you don't beat them off then quest givers and vendors die. Don't worry, they'll come back in five or so minutes. Five minutes, that'll feel like an eternity if you're trying to level.

That "number of naga" mechanic would of course also be affected by players killing naga, reducing the number. This could give rise to interesting dynamics - say you're returning to hand in some quests, and see the naga count inching close to 100 .. do you detour to kill a few first, or rush in hoping to get your hand-ins done before they attack?

Make it impossible to document these locations?

What is wrong about exploring a really big forest that is filled with random spawns? You earn EP while you do it, you do it only if you like to explore (important) and the designer can even put a few things in there to find. (But make it impossible to document these things in internet sources!)
-- Nils

OK, how do you make a MMO where the locations of things are not trivial to document (saying "impossible" might just be a bit much).

My first strike would be to eliminate x,y coordinates entirely. Don't show them on the map, and don't let them be available thru any in-game API/scripting. Of course, if you have a map, then you've got a de facto coordinate system: the pixels of the image of the map are arranged into a regular grid.

What next then? OK, have the map you see be different from the map I see (thus not a standard shared image). I've written before about mapping as interactive game activity.

You'd still want it possible to find your way back to the thing you found. It might need to be a case of "head up the mountain path until you see a rock shaped like a dog, then climb down the slope until [yada yada landmark landmark] and there you have it! The lost treasure of Sierra Madre!"

What about you ... how would you tackle this design challenge?

Solving the prisoner's dilemma of Tol Barad

The Tol Barad battleground has apparently been a big fail with lots of win trading going on. This was due to the massive honor being awarded for successfully attacking and scant rewards for defending. In some cases win-trading was coordinated or at least discussed and argued for on the forums, but on some servers it just naturally and organically occurred. Why defend this time if instead you could just come back later and attack, and gain (literally) 10 times the rewards.

There are changes incoming, but they sound like simple tweaks of the numbers and not a change to the underlying system.

The problem is that, apart from the rewards of the battle itself, there is no difference between successfully attacking and successfully defending. Win-trading is still an optimal path to rewards, whether planned and coordinated or simply by herd instinct.

Consider though if successfully defending multiple times in a row gave some additional benefit, and continuing to successfully defend increased those benefits further. One successful defense might give (say) a +2% buff, two successful defenses (in a row) might give a +4%, and so on, and on. Now, the defenders have a reason not to engage in win-trading because the longer they can hold Tol Barad, the greater the rewards. (Whether it's +2% cumulative or +5% cumulative, has diminishing returns, is capped etc are parameters that can be tweaked. Lets just use +2% for sake of example.)

Tricky thing though is that you can't make this buff apply to PvP, otherwise you've got a positive feedback loop that will eventually make it nigh on impossible for the attackers to succeed. Especially since fewer and fewer potential attackers would even bother queuing.

Thankfully, Tol Barad includes a raid dungeon which only unlocks for the faction that won the last battle. If the defender's buff were to apply to that PvE content then you don't have a positive feedback loop but you do have a motivation for defending, a motivation that grows and grows.

Problem solved? No. A strong defender faction has greater motivation to successfully defend, which in itself demotivates the attackers. They failed in their last attack, and they know the defenders have even more reason to defend for your next attack. We still need some reason for attackers to get in the game, and this is even without the defender's buff further motivating the defenders because, by definition, they are the weaker faction.

A PvP buff needs to be provided to the attackers, one that is also cumulative with successive failures. For simplicity sake say it's a cumulative +2% to all attackers for damage, healing, health, movement, and so on.

This way a generally weaker faction will be buffed and buffed to the point where they can actually win the battle and control the island. The buff to movement could be really interesting: eventually the attackers could chase down lone defenders with ease, zerg from point to point faster than the defenders, and easily escape getting overwhelmed by a numerically superior pack of defenders.

As additional incentive, a successful attack could result in the attackers receiving the full PvE buff the defenders had for one round, before it resets back to the initial +2% for a successful defense. If the defenders have dominated Tol Barad for six successful defenses then the attackers know they would be receiving a +12% buff to raiding if they win. Plenty of motivation to show up and try harder.

Note though that if the recently defeated defender's were to rally the next round and successfully attack they wouldn't regain that +12% buff. They'd start over at no buff, otherwise the buff will only rise and rise and rise, just changing hands every now and then, and that would be untenable. Mechanically, call the cumulative PVE buff for successively defending the Defender's Buff and the transferred PvE buff for successfully attacking the Victory Buff, and the rule then becomes easy to express: successful defenses grant a cumulative +2% raiding Defender's Buff, successful attacks grant a Victory Buff equivalent to the losing defender's Defender's Buff, but the Victory Buff only lasts one round (ie. gets replaced with the cumulative Defender's Buff).

Would this work? In what nasty underhanded ways can you imagine this design could be exploited and abused? Would the forums asplode in flame?

A telling assumption

By its very multiplayer nature, a MMORPG can not allow a player to really change the world, because other players need to have the opportunity to play through the same story.
~ Tobold, on story

Players need the opportunity to enjoy the game, that is all.

They can play through different stories, even make their own.

QOTD - Danc on building systems for stories

Don't build games in order to tell a single story. Build meaningful systems that create an explosion of culture, spread by the players who are absolutely thrilled to share what they've learned.
~ Danc, on story as evolutionary success or failure lessons

Quote of the day

They simply forgot to balance their game around gameplay and hence jumped to the conclusion that it was the quest structure itself that was the issue and NOT the structure of the game. So instead they create quests that have nothing to do with the MMO or RPG aspects of the genre (a.k.a. meaningless mini-games that don’t involve ANYTHING you have done with your character AND don’t involve other people).

~ from a comment over at Keen & Graev's blog

Designing systems, vs designing levels

There's a fascinating post over at Lost Gardens, one which gets me thinking whether that approach could also be applied to MMOs. If it could, then it could well be the trick to overcoming the 900 lb gorilla in the room.

For contrast, there's this over at Hardcore Casual.

... what are your thoughts?

World centric or player centric?

Via Syp over at BioBreak:

“How many times will a developer promise me that my decisions will impact the world before I actually see it?”
~ KvanCetre @ Massively

... when the world itself is designed to be the game, with player participation a secondary consideration.

What I mean by this is first design a world, populated with whatever is in that world, and design it such that those things are determined by the other things in the world. Make the world a dynamic model, even absent of players. Design and develop the world such that, even if no players ever logged in, the fates and fortunes of the world inhabitants will ebb and flow.

Now, add players. (Then fix all the bits that get broken by adding players).

Contrast this with what you've likely seen a few times now: design a world as a static landscape, add players, and then add mechanisms for players to affect that world. The design of such a world is entirely player centric - without players exploring their personal destinies (aka "consuming content") the world effectively ceases to exist.

Kill 10 rats ... or kill 4 specific rats

Which type of quest would you prefer to do:

• kill 10 rats, in a nearby farmyard with lots of rats
• kill a red rat from the barn, kill a black rat in the forest, kill a white rat in the house, kill a brown rat in the yard

Certainly, the first quest could be more efficiently completed, but the latter has more potential for a varied experience.

Fine print: they both would take about 5 minutes to complete, both reward 55 silver and 1,250 xp (from mob kills and quest completion).

A lack of talents

Is it just me, or does anyone else feel that WoW's new talent system only facilitates defining your character in terms of which talents you don't take in a given tree, rather than the ones you do take?

Supporting explorers

Bartle defines explorers as players that seek out information about their world, and share it with others. Part of the reward for explorers is the private delight in discovery, but often an equally important reward is being able to pass on your knowledge to others, or even to share it together.

How to support explorers in an MMOG?

First, obviously, you need lots of content for them to explore. Start with adding quirks to your terrain, to reward the player for pushing through the bushes and exploring an otherwise unimportant gully. Simple things like setting up a picnic blanket and basket, but with skeletons scattered about and nothing more – an unusual sight with an unexplained story.

You could also add hidden mechanisms to your world, like ley lines that affect magical powers. Let the explorers figure out where they run, and what effects they have on the game.

Map making and documentation.

Another neat thing you could do is build into your game UI a means of taking and sharing photos of the world. That way you could not only boast about this tragic picnic in the wilds but show proof to others. I don't just mean allow the player to take a screen shot and upload it to a website, I mean collect and share photos in the game itself. They could share individual photos the same way items are shared in chat or trade, they could pay an NPC or a crafter to take their photo and mount it in their home or guild house, and they could publish collections of images with annotations in guide books.

Storm the beach, climb that mountain

One problem many pvp based games face is that experienced players with hard earned gear stick around and gank the new comers. Another problem that can arise is that the smartest battlefield strategy is to mindlessly zerg.

So acknowledge that, and make it work for your game, not against it.

The key issue with newb ganking is that there's a sense of betrayal of fairness, that on the battlefield all should have an equal chance and equal capability. Unfortunately, handing out gear improvements as rewards undermines that supposed fairness.

Thus: design the pvp game around scenarios which have entrenched veterans defending a beachhead, and the newbs being cannon fodder. Make it 40 vs 10 even - a game of asymmetric warfare. It might even be possible to have the numbers on each side being determined by a gear budget, that way if a massively over-geared player queues for an early tier they just might find themselves defending the whole beach by them selves.

Once those players attain sufficient gear they get swapped to the defense team, or can go offense on the next tier of attack. First the beachhead, then the village, then the hill, the castle, and onwards. It's probably important to have that sense of progression in the scenarios too - don't just bump the players up a numeric tier and throw them against the same beachhead (only now facing tougher and more well equipped defenders).

So .. two points: asymmetric warfare, and narrative progression.

Think that might work?

Steam-punk factory

If you were to build your own steam-punk factory, what would you have in it?

I'd have furnaces and steam engines, massive flywheels, steam pipes, steam pressure regulators, conveyor belts, trolly carts on rails, big water tanks, grabby things on crane arms grabbing crates and barrels of incoming supplies, big axes on levers chopping up the crates & barrels and feeding them into those furnaces driving the steam engines filling those pipes.

Lots of odds and ends all joined together into a one convoluted mechanism, taking in raw materials on one side and spitting out manufactured product on the other.

What are the components you'd have?

The wolf is represented by the bassoon?

There's this thing called a Leitmotif. At its simplest, it is an audio cue.

So here's an idea - represent different classes or roles with different instruments, and have various actions or activity levels give rise to longish chords of music. This is more than just a 3 note audio cue which is played every time the priest casts Smite - instead, there would be a multi-bar track of music which is associated with a priest casting lots of offensive spells, with a different multi-bar track for healing spells.

For the tank, as the danger level increases the dramatic tones of his leitmotif could increase. The more adds he's tanking could be represented by a more frantic pace. If the tank's health starts to dip, the music could take on an ominous score. And if the tank dies?

All this is a mile ahead of simple background music on a loop and sound effects tied directly to specific abilities ... so I wouldn't expect to see it in any game anytime soon. Which is a pity.

And no, the wolf is represented by French Horns. It is the grandfather who is represented by the Bassoon.

Grrr .. a rant on bad design

This afternoon we formed a pug with scrubs from /2 and headed into Ulduar to try Freya hard-mode.

It's a big indoor garden room, with lots of packs of trash between the boss and the entrance. We clear all that.

The boss fight goes something like this:

• Phase 1 has six waves of adds which must be eliminated before the boss can be harmed. There are three types of waves, each with their own abilities. Some do a massive AoE explosion on death, some do knock-backs, one wave is just a big tree elemental and some mushrooms the raid need to hide under.

• Phase 2 is less hectic, it's mostly a tank and spank but with the boss throwing seed bombs periodically.

So here's the rant: nowhere before the boss fight do we get a preview of any of the mechanics of all those adds. Not at all. Remember all that trash before the boss ... they don't do anything remotely similar. They even have some mechanics unique to themselves.

The only possible way to learn how to handle the adds in the boss fight is to do the boss fight. And probably wipe, multiple times. And all that trash before? A pointless waste of time.

It would have been better if players would be exposed to each of the mechanics in separate trash packs, where they can learn the signs and the effects and teach any newbs in the raid what the heck happens. Then the boss fight packs them all together in one big encounter in combination.