Iron Chef Carnage

PyroDragon's option #2 gets my vote.  You're in a tight spot, trying to share a game world that allows client-side time dilation.  If it'll be a small game, with just a handful of dudes playing in the world, then you should make it kind of turn-based.  You do what you're going to do on a given day, and then when everyone's done with that day, the day advances and everyone gets to start the next day together?  If you're on a three-day journey, you'd just hit the "Stay in hyperspace" button and tell everyone else to hurry up on Teamspeak, making fun of their moms and making yourself a sandwich until they finish with their spreadsheets or whatever.  That way, if there are two players on the server and they're both warping to the same station, they could just poke the, "Next" button until they get there, then hit the "Wait a day" button until their buddy arrives.

Ideally, let players use the dead time for something productive.  You're loading your ships three jumps from our rendezvous and I'm on a six-day flight there, so I'm looking at at least three days of, "Push button, wait for pal," gameplay.  Let me set up some market orders, browse commodities markets remotely or optimize my ship and crew configuration during that time.

I hesitate to suggest a play-by-mail option here, but it might be worth thinking about.  I have a ten-day flight ahead of me, I can just set it to autopilot, then log out for the night, and my ship will make the trip on its own.  I might even be able to set up a system whereby I program the ship to fly to a given destination, load/unload cargo, perform a preset transaction with another player, fly to another site, do some work there, and then head back to HQ.

For instance, let Player A and Player B draft a contract whereby Player A trades ten widgets for five of Player B's gizmos, and then Player A sets up a route that takes him to Planet 1, docks and waits up to four days for Player B to arrive and meet the order.  If Player B shows up and meets his side of the bargain, then the order is processed and Player A's ship continues with its instructions.  This way, you reduce the amount of time players have to spend waiting around for each other, give them something to do while they're waiting (set up waypoints and instructions for the AI) and get a little closer to the ideal, "Hit button, instantly teleport to a future meeting" ideal that you describe in your top post.

One of the big hang-ups with MMO permadeath is that the player is the base unit of the game, not the character.  Social organization and cooperation are the core or a lot of MMO gaming, so if I go on vacation, and come back to find that all my mates have died and been resurrected a dozen times and I'm the odd man out because I'm still specced for the team we ran two weeks ago, then I'll feel like my investment has lost its value.

Maybe you could simply remove the social aspect.  There's a weird social experiment MMO, called The Endless Forest where players are anonymous non-human animals populating a server with an interface that is limited to basic emotes and actions.  You can give your little guy a sigil, some unique symbol that will allow others to keep track of individuals in a group, but it isn't a symbol that appears in any alphabet, and there's no way (last time I checked) of making any kind of a "buddy list" or keeping in touch with acquaintances.  No guilds, no punk lists, no e-peen, the only thing that matters is what you're doing at that very moment.  That game is a tranquil environment, where there's very little to accomplish, but the act of meeting people, interacting with people and then leaving them and forgetting them entirely is perhaps the social equivalent of a roguelike.

I just played a couple rounds of Starcraft II, and the matchmaking there resonates as well.  I get chucked into a match against some random dude based on opaque criteria, we duke it out, I do a good job making SCVs but forget to keep up with viking production, I lose air superiority and his tanks faceroll me.  End match.  My stats are modified minutely in a way I can't really discern, and then I sign up for a new match.  New opponent, new map, new chance.  I harass with a probe, I expand, I mess up my micro and lose my immortals, then I just go balls-out on stalkers and colossus and win.  End match.

No how about a game where you get to make a character, engage in a cooperative/adversarial contest and then resolve the session in one of three ways:  You die, you retire or you "stable" the character for future use?  It may or may not be set in a persistent world, but getting saddled with a random team based on matchmaking heuristics and then being called upon to play without any kind of direct communication (no voice, no chat line, at most maybe contextual utterances like you see in co-op Portal 2) would allow a player to immerse himself in the game and then emerge from it without a strong sense of loss if/when their character is destroyed or abandoned.  You might add visible accolades and designations, like the insignias in Call of Duty, or you might black out all indications of who you are playing with and against.  More or less complex challenges, different gear or options, all kinds of stuff could be "unlocked" on a per session basis, so a handful of experienced players might find themselves in a hopeless situation, on a decompressing spaceship full of angry aliens, and the episode resolves when one guy tells the others to get bent, shoots his buddy in the back and takes the only escape capsule to safety.  Sweet.

A robust, versatile engine would be called for, and I think it would be totally acceptable for the vast majority of sessions to end in anticlimactic death for all parties.  There's a Pen & Paper RPG called All Flesh Must Be Eaten (which I've never played), where it seems that every game must inevitably end with the death of all player characters, and the fun is in authoring your own last stand.  Maybe you get devoured in the first four minutes, cringing in a corner, and then go watch a movie and drink a beer while your buddies wrap up the round.  Maybe you go down under a pile of undead, pulling the pin on your last grenade while your buddies escape into the sewer to fight a little longer.  Maybe you get a call from your mom and shoot your brains out so you can go eat dinner.  Maybe you're the last man standing, with an empty pistol and a bent machete, pushing that Sisyphean rock until the DM tells you that it's 4am and your character had an aneurysm from hyper-elevated badassitude and it's time to go home.  Any way you look at it, you need to be able to shape the game and the game world to meet and defeat the players.

The bottom line is, an anonymous matchmaking session-based MMO roguelike would work if the player could sit down, fire it up, have a rewarding experience and then walk away.  Being able to share a persistent experience with people can be undercut by player mortality.  But by eliminating the persistence of the characters, permadeath can be convenient and fun.  I swap Skyrim stories with co-workers, and we enjoy the shared experiences despite the total lack of multiplayer in that game.  Same with Dwarf Fortress--I'll talk to a buddy for hours about the time his complex clockwork drawbridge system malfunctioned when a cat fell onto a pressure plate and squashed all his archers.  Make a game that provides an endless parade of fun, engaging vignettes, and you'll please players again and again.

I'd recommend an aftergame lobby, where the dead can meet to talk about what happened, exchange insights and maybe even peek in on what's happening since they died.  Just be careful not to introduce metagaming opportunities that will corrupt the formula.

A simple solution might be to have "animal feed" as a resource, and allow land to be purposed for that.  Adjust the food requirements for each type of livestock, and you'll be able to give different animals a different "footprint" on the estate.  You're right that a small number of chickens won't have a significant impact on land usage, but if you're raising hundreds of chickens, you'll need to supplement their feed because there's only so much they can forage.  With pigs, waste will help, but if you have a bunch of them, you'll have to find something to feed them.

You could have a certain amount of "free" feed for chickens or pigs, determined by the size of the estate and represented as a small number of "maintenance free" animals.  A couple pigs, a half-dozen chickens, no cost in land usage.  If you want a hundred chickens, you'll have to grow and harvest corn to feed to them, and that'll require acreage.  That would preserve the idea of low-cost chickens, but still leave players the option of building a chicken-raising empire within the rules of your game.

Whoa.  I just got hit with a wave of nostalgia.  I like to think that the guys who developed BattleSwarm saw this thread, perhaps while they were learning to read, and were inspired to start a career in game design after studying the ancient text that you just posted in.

Short teleport.
T1000-style blob motion.
High-speed phasing.
Smash that sucker with a sweet shoulder check.
Travel forward in time to a moment after your now-unnecessary parkour move.
Move the wall behind you.
Quantum space-hole.
Something with lasers.
Merge with the wall, then emerge on the other side.
Do that thing David Copperfield did with the Great Wall.
Split into multiple copies of yourself, then do all of the above simultaneously.
Invert the world along a fourth axis, turning the whole space inside-out such that your location vis-a-vis the wall is reversed.
Make one of those games where a knee-high wall is an insuperable obstacle.
Bear blasting.
Cast the wall into a temporary limbo, then jog on through before it returns from the infinite beyond.
Blow a hole in it with a bazooka.
Hand-held plasma torch.
More lasers.
Step outside of the corporeal dimension and walk around it in some kind of dream space.
Match its molecular resonance and pass your molecules right through it.
Dissolve it with acid.
Psionically project your astral shadow to the other side.
3D side-scroller?  Go around it in that handy third dimension.

From a narrative point of view, all-powerful wizard-type dudes tend to be crazy and highly devoted to their field of study, to the exclusion of all else.  They also tend to be pretty old, physically frail and dependent on their magic.  Here's one way to look at it:

Have the "base character" of the mage be a total weenie.  Pathetic stats, incapable of using armor, no weapon affinities.  Then let him build up magic powers on that skeletal framework.  Look at it like bonus counters in Magic: The Gathering.  Treat it like a bag of tricks, an inventory of gimmicks that can either be cast using a mana pool or expended on a single-use basis, perhaps with a cooldown or recharge cost.  So, yeah, my wizard dude can surround himself with a shield that makes him impervious to physical harm, and then use some telekinetic magic to toss around a warrior-type dude while cackling maniacally, but if the totem that powers the shield is smashed, or if my link to the arcane source is severed, or if the spell just wears off before I can come up with a way to actually kill the guy, then I'm just a weedy dork about to get the wedgie of a lifetime.

Even if I win the fight, I'm going to have to go do a few mundane, thankless quests to revisit my Lazarus Pit, expose my talisman to the light of a Blood Moon, refill my decanter with the water of the moonshot spring and meditate atop my array of neodymium magnets before I can pull that kind of shit again.  So the mage must be an economist, understanding the value and merit of his smartbombs and lifelines, and expending them in a responsible and profitable way.  You won't see dudes blasting fireballs at strangers in the newbie zone, because those this have a real value and are hard to regain.  On the other hand, if you think you're getting the drop on a mage, and you have him cornered, he'll bust out some cataclysmic juju to get out of a tight spot, rather than get whacked and go to his grave with ammo on his belt.

Fighting a mage, then, would me like the fisher fighting the porcupine:  Keep moving, keep taunting, get it to stick its quills into logs until it's tired and weak and largely disarmed, then knock it over and tear its guts out.

Yeah, I'd go with text-based puzzle games.  No graphical assets, simple inputs, and they can be written on a graphing calculator during lunch period.

A little more sincerely, if you want to see the simplest, easiest games to make, look at the kind of "games" that show up on banner ads.  "Punch the Monkey", "Tug of War", "Spot the Difference", these kind of basic exercises are chosen because they cost virtually nothing in terms of money and resources to produce.  Hit up Newgrounds and look at the games that get mediocre scores, you'll see a lot of basic ideas getting recycled there as well, when students have to crank out some functional piece of supposedly entertaining multimedia in the face of a short deadline and no compensation.

You can cut corners, too.  Making a game hot-seat multiplayer obviates the need for AI or network coding.  Abstracting graphics can allow you to make a whole game with a half-dozen sprites, or none at all.  Omit sound.  Leave out control remapping.  Use a Windows pull-down menu instead of writing a GUI for options.

There's probably a distinction to be made between lean development and half-assery, but if you want to keep the load light, look for shortcuts.

Ever play any Dwarf Fortress?  In the Fortress mode, you get waves of immigrants.  Many of them are completely freaking worthless, like a cheesemaker or an apprentice glassblower or a miller or something else that either a) anyone in the fort can do just as well, or b) you don't need in your fort.  "You make soap, eh?  I've got thirty years' worth of soap in here.  Welcome to the militia, Recruit.  You don't get any armor."  Sometimes you get a really good recruit, like a skilled weaponsmith that saves you a ton of money on training, or a terrific brewer.  More often, you train up a dude to fill a critical role.  A grandmaster mason that can make ornate doors and tables from raw stone, or a legendary armorsmith that imparts a huge bonus to the defenses of your elite warriors, or the elite warriors themselves, who have been training and battling for years and can crush raiding parties that outnumber them four to one.  These guys get nicknames, and fancy bedrooms, and opulent burial chambers.

In this way, each character is both anonymous and unique.  Sure, they all have names, but they're also conveniently color-coded by their job classes, so when you need a floor smoothed or a wall built, you send eight random white smileys to get it done.  On the other hand, if you need a masterwork platinum sarcophagus that menaces with spikes of ruby and is encircled with bands of brass, you use dedicated workshops that are only used by one bodacious guy, and when the vile force of darkness shows up, you hit the button that sends the faceless punks up to absorb arrows and the important bosses downstairs where the canned goods and hookers are kept.

Sounds to me like the first two percent of Dwarf Fortress.  A world isn't a game, as I'm sure you know.  Neither is it a foundation of a game.  There are plenty of games out there that don't even pretend to have a solid foundation in geography, as I'm sure you know.  What you have there is a tool, a powerful element of world creation that can be either envied or totally disregarded.  If you are able to come up with a context for it, it's a huge asset.  If not, it's fluff.

Think about its merit.  Think hard.  Find a context in which that tool is powerful, and then capitalize on the power.  A strong world leads naturally to resource exploitation, and if you can come up with a game that depends heavily on that, you're in good shape.  If you come up with a design that uses your world generator in an ancillary way, you've got a handy utility.  If you come up with a game design that features a world as a backdrop, you've got a ready-made green screen.  Remember:  A product for an audience should cater to the audience's perspective.  It's possible to have thousands of hours of world-building that's only relevant in a three-minute cutscene in the middle of a 120-hour gameplay experience.

Don't be so proud of your world-building utility that you convince yourself that world-building is game design.

What's gained from the Hyrule setting?  Hyrule's looked different in every Zelda game ever, varying so drastically in both size and contents that it can scarcely be considered to be any specific "place" at all.  You'll likely have a Death Mountain, a Spectacle Rock and a Lost Woods, but what else is there that is distinctly Hyrulian that you couldn't get in a totally novel world?  Truth be told, there's not a lot of continuity in Zelda games at all, and its success as a franchise boils down to star power and consistently excellent game design.  You could sell a turd with a triforce on it, because of fanboys, and Ocarina of Time would have been an awesome game even if it was set on Venus and all the characters had been British robots who wanted to form the loudest heavy metal band ever.  Hell, even the pantheon changes from game to game.  Every game features a host of spirits or gods that are ancient and wise, but the games are so far apart on the timeline that in each one the gods of the others are either totally forgotten or haven't come into being yet.  Continents, races, primal forces, the goddamn laws of thermodynamics get altered more frequently than Ganon and Link make an appearance.

As a long-time Zelda fan, what I love most about the games is being Link.  I love the idea that the world is filled with dungeons and treasures and weapons and tools and armor that have been stored away for centuries, waiting for me.  I roll into a crumbling temple, and the sage there is about 150 years old and his family's been guarding the vault since time immemorial, beating back armies of bandits and resisting government interference and living on beans and humility.  He looks at me, and he says, "Oh, sweet fucking Jesus, it's about goddamn time.  This is all for you."  And then I touch some crystal with my hand and the wall creaks open with a thousand years of dust falling off it and inside there's a chest and in the chest there's a 5% armor upgrade that turns my shirt red and I high-five the sage and wander off to look for pennies in tall grass, leaving him with nothing to live for.  Fuck yeah.

The MMO paradigm is dependent on a lot of things, and Hyrule isn't especially conducive to any of them.  You could do it, for sure.  You could have a game that used the intellectual property associated with the Zelda series in a cookie-cutter MMO, and you could put it on the Wii, and you could use the WiFi to hook up with some servers, but I don't think that there's much to be gained, either for the Zelda universe or the MMO medium.  In fact, I think it would be bad for Zelda.  The games (excepting The Adventures of Link, which was an aberration) don't feature "leveling" in any real sense, and equipment upgrades are few and meaningful in Zelda games, so two of the mainstays of MMOs are lost immediately.  I'd say that a Zelda MMO would feel more like Minecraft than anything, with the characters all being on roughly even footing throughout, having to set themselves apart by their deeds.

Enough nay-saying, here's how I would do it:

The Zelda games are always set at a cusp, a time of renewal, when the world has reached some kind of critical stage and is being remade.  It's vulnerable then, and the Triforce and its stewards are agents of transition, scarring and burning and remaking the whole world, so it can start its life cycle over again in a new form, laying the groundwork for a new battle, a new cusp, a new video game.

Add a historical period between games, where the Hero is gone for a while and the world is being systematically regrown and reconfigured by various divine forces.  Fill it up with enough story to account for Link and all his descendants/students dying of natural causes in the time of peace.  Then, set up a system for players to be in charge of resetting the world for the next Hero.  Have the "factions" be the sources of the epic gear that Link, Ganon and Zelda will use next time they manifest in the world.  Players will do quests for the Gorons that will serve their destiny as the makers of super-great armor or weapons.  They'll do jobs for the Gerudo that help them become the premiere experts on stealth and archery.  They'll protect the Kokiri Forest so that the magic of that place can flourish.

Throughout all this, they'll gain access to the skills and powers of each place, being enriched by the energies that they cultivate, so your dude will be awesome, like Link, but he'll exist in a world where awesomeness isn't all that unusual.  I like that a lot of the stuff in Zelda games, and in a lot of fantasy, features powers and entities that come from a lost time.  So, yeah, link's fighting a three-headed dragon from beyond the veil of human imagination, but there must have been a time when that dragon had parents, and maybe there was a whole race of three-headed dragon-monsters who got wiped out, leaving their last son--their greatest warrior--as the guardian of a deep place in an impenetrable cavern.  So rewind the world to when that immortal beast was an infant, have players do battle with his ancestors, and set your game in a world before the hellions were banished, before the gods retired to their mountain springs and their powers grew thin and weak, in need of rejuvenation.

Then, go crazy with it.  There's a lot of lore in the Zelda universe, not much of which is consistent, so you can have all kinds of playable races and wacky adventures for players, and all their quests, all their levels, all their achievements can be deposited in a bank and earmarked for the next cataclysmic event.  The king of Hyrule is doing his best to keep peace and look after the welfare of his subjects, but the world is filled with brigands and monsters and reclusive magic beings who have their own agendas.  Players can build faction standing, they can explore and destroy, but the chief objective is to cultivate the world, to seed it with high-intensity packets of power and influence that will, eventually, be discharged in the world-shattering battle that inevitably comes when the cycle repeats itself.

You could have a whole faction that does nothing but grow and imbue magic crystals, developing the art and refining the enchantments.  Millions of quests could be completed for that faction, and the end result (which is never achieved in the MMO, of course) could be some amulet that Zelda uses in a cut scene of a Zelda game.  A whole nation could flourish and die, leaving a legacy consisting of a handful of octorock-infested dungeons and one hookshot.

It would be a strike against, for sure.  Mentally, when I see a game with just one resolution setting, it makes me think that the development team cut corners on things like compatibility and features, so I go into it expecting the game to have problems that might break the experience for me.  I don't actually make games, and what little I know about the process I learned here, but I assume that ubiquitous features like resolution support is something that can be "dropped in" from some kind of readily available toolkit, like it's some kind of module or something.  Since I believe it to be trivial to code-wizards, its absence makes me think that the game is not the real deal, but instead some kind of lame project that was made in a hurry as a midterm in a college class.

I don't know a formal name for a prompted button press.  It barely qualifies as a minigame, in my mind, since it's just one input, and the "game" consists of seeing the prompt on the screen and pressing the button.  If it's more sophisticated, like Rock Band, I'd call it a rhythm game.  If it's just remembering a sequence and putting it in, I'd call it a Simple Simon game, and if you're just seeing "Press X to not die" and then pressing X, I'd call it an on-screen prompt.

Your wolf scenario reminds me of the Krauser knife fight in Resident Evil 4, seen here.

But it's not the only example, there are others.  The first time I saw it was in Dragon's Lair.

What do you need the name for?

Here's my suggestion:  As Sandman said, store the data for a time.  So for a few minutes, or until the player gets far enough away from the area for it to be considered free to operate without him, hold it in memory and don't let enemies spawn inexplicably or anything like that.

After that time passes, though, and it's out of the player's view, don't worry too much about simulating everything that's there.  Don't just reset it, either.  Give each area some abstract stats that can be modeled with very light resource consumption.  Say a player loots and burns a village, then wanders away.  Give that village 2% "Structure" and have it slowly increase back up to 100%.  If the player comes back when it's at 15%, some of the buildings will have started to be repaired, and the villagers will be actively engaged in rebuilding.  At 50% spawn an area that's partially functional, maybe a shop or two are up and running.  At 87% there's still evidence that it's been damaged, but it's pretty much all there, and at 100% it's a new, complete zone.  A town might have stats for wealth, structure, population, military presence etc. and a player's actions toward it can cause it to get richer, more or less military presence, higher or lower population.  In this way you not only present a world that feels like it's operating when you aren't looking at it, but which responds correctly to events in the world.  You don't have to save the location of every dropped sword and burned tree, just make sure that it looks about right every time you go there.

It's a nugget of forum etiquette with which you may not be familiar, I'm afraid.  In my experience, forums universally discourage the practice of "cross-posting", as it's been defined here, because it has two negative effects:  First, it fractures the discussion into two separate conversations, making both less effective.  Second, it clutters boards with threads that might be filled with comments that don't make sense unless the reader is familiar with both of the parallel discussions.

To keep things tidy, a moderator like Tom will usually close one of them and direct interested readers to the other.  I like to see a link in the locking post, which Tom didn't include.  That makes it tougher for me to find out where the other thread is and read/comment in it, but thanks to GDnet's design, it's a simple matter to click your name and find the other one in the list of your created topics.  Rest assured, anyone who read the message and saw that it was locked for cross-posting would have no trouble at all finding and replying to your duplicate thread if they were interesting in participating in the conversation.

This thread does you some harm, however, since you are coming off somewhat petulant.  That tends to garner ill-will in your readers and will make people less eager to help you out with your issues.

GDnet is a great community, and a very mature one.  Since videogames have a wide audience and are very interesting to people who have no idea what game design is, we see a lot of people come and go with hare-brained schemes or vague visions, and we are often skeptical when a new name pops up with an idea or question that is neither novel nor significant.  We immediately think, "Why has this guy started a discussion thread to learn things that can be learned with a quick Google search and a few hours of reading?" or, "Has this guy never played any indie games?" or, "This is a college kid with a big ego and a short attention span who wants to be told that he's great," or, "It's not worth my effort to parse this wall of text in a gracious and diligent manner in order to formulate a reply that will not be read in a gracious and diligent manner by the topic's creator."

Having perused your thread on the For Beginners forum, I'd say the feedback you received was very good, in keeping with GDnet's track record for constructive replies and criticism.  I was going to bust your chops for your shabby typing, but that's been covered.  I was going to suggest that you do some reading, but that's been covered.  I was going to mention that on a design team (or any team, for that matter), contributions and duties have to be diffused throughout, and the end result will not embody one man's genius.  All of that was explained, patiently and well, with helpful links and a professional tone, and you sulked and called names.  Then it went south when your status degraded from "Plucky upstart with newbish questions," to "Hyper-sensitive nubcake who isn't likely to succeed."

With this thread, you've graduated to, "Egotistical douchebag who doesn't understand anything (not even how forums tend to work) and can't be bothered to learn about it." and your goodwill is just about shot.

Pro tip:  In an online community, first impressions are important, and a few faux pas that hurt you here were:

1:  Bad presentation.  I know you don't think typing matters, but the text of your post is your face here, and if you can't be bothered to put forth some effort in your presentation, it shows that you think very little of either your idea or your audience.  Since you obviously think highly of yourself and your ambition seems genuine, your lack of concern for the post is an insult to your reader.  Besides, it makes it hard for us to know what you're talking about and give useful feedback.  Imaginary commas are like sarcasm, the internet doesn't transmit them well.

2:  Etiquette.  A cross-post is a simple mistake, a newbie mistake.  It looks bad, it inconveniences people and it reflects poorly on you, the author.  Similarly, making a whine thread just to complain about something makes you look like a pouty little bitch, and that's not going to do you any favors here.

3:  Talking trash on the moderators.  Tom's got his detractors, sure, but he's done things and written things and said things that people respect, and that means he's got more credibility here than a 3-post account does.  You don't have to lick boots, just don't pick fights with the regulars.  The whine thread ties into this one as well.

I don't know if you can salvage your good name here, but I hope you read the links you were given, I hope you keep learning and working toward your goals, and I hope you communicate more effectively online in the future.  At the very least, this negative experience on GDnet might help you not make an ass of yourself next time you try to learn something from a forum.

Ass.

Co-Op is great, because two players are more than twice as powerful as one player.  Twice the gun, twice the eyes, twice the thinking, and since video games are not as visceral as real life, you can avoid panic and have three players surrounded by dozens of aliens and their chief concern is maintaining headshot streaks and getting combos, rather than the pure pants-shitting terror that would likely accomodate such a situation in real life.  The biggest power players have in the game is communication and consensus.  In a scary movie, a guy says, "Did you hear that?" and his buddy says, "Shhh, I'm trying to peep into the girls' locker room."  In a game, the guy says, "Sounds like a licker, check the windows," and his buddy says, "Right.  I've got the shotgun, so stay in the corner of the room and hit me with the flashlight while I kick in all these doors."

So mess with their communication and consensus.  In general, mess with what they can see.  Is there a little waypoint marker on your screen when your mate is around a corner?  Have it move around or vanish or turn red for no good reason if they've been gone a while.  Is their status shown on the bottom of your screen?  Switch it to "dead" or put a monster face there now and again.  Not a lot, just sometimes.  Make it so that players wonder whether or not they should disregard a piece of information they're given.

Hallucinations, unique to one or another character, are great for this.  Have sounds that only one guy hears, or make lighting different for each player, so they'll hug certain walls and avoid certain corners for reasons their buddies can't see.  Have an enemy that looks to every player like a different member of the team, right down to the little floating name above his head, and have it travel around with you, collecting items and getting healed by your medic, until someone notices that you've got a six-man team instead of the five you left with, or that there are two Jerry Browns restocking at that ammo crate.  And then, which one do you shoot?

How about an enemy that latches onto a player, undetectable to that player, and psychically makes the other team members look like monsters?  You turn away from your team for a second, and when you look back, the three riflemen have suddenly become two riflemen and a zombie.  You open up on the zombie, and then what?  Do they assume you're one of the dopplegangers and fire back, do they realize what's up and yell at you to stop?  Once the situation's resolved, everyone has to take a moment to scan each other and find the parasite, then remove it before moving forward.  And then the next time a zombie shows up in a room, you might think twice before dropping the hammer on them.  Maybe they'll get you.

It doesn't even have to be that specific.  Open a door or airlock and toxic fog sweeps out.  Everyone dives for their gas masks and you go in, but your health is still draining slowly.  You look for another way out, but there isn't one.  You find one tiny alien lifeform in there, clinging to the ceiling, and when you kill it, the fog vanishes, your health is restored, the lights come back on and the doors are all plainly visible.  That's annoying, but when there are eight zombies in that room with you, and you're fighting them in the dark, with your vision occluded by the gas mask and feeling trapped, that's a serious tractical consideration.  Ten minutes later, you open a door and gas comes out and you think, "Is it real?  Do I put on the mask?  Is my health really going down?"

That kind of stuff, that uses the players' fast reflexes and love of headshots against them, can turn the sort of skills and tendencies that make a player so good at games and use them against the team.  It's cool to have a hyper-elite Halo nut holding the sniper rifle, but if he gets mind-controlled at the same time that a demon wave hits the team, he'll be looking at ten targets and not be able to tell which three are his guys.  Does he take his time, watch behavior patterns and pick his targets?  Does he open up and hope it goes well?  Does he take a few precious moments to sling his rifle and get out his medkit and take the antidote?