Slashdot Mirror
Molyneux On Future Of Game Design
Thanks to GameSpy for its interview with Lionhead boss and Populous creator Peter Molyneux, part of a talk held at the DICE Conference in Las Vegas. While admitting that "this industry tends to bumble through innovation", Molyneux noted freeform games could be less interesting than they looked, lamenting that "he always dreamed about making games that give people total freedom, but what he discovered is that complete freedom in a game world is boring after about ten minutes." He concluded by suggesting alternatives to the 'sandbox' approach, arguing: "People like to have challenges, goals, and something to struggle toward."
"People like to have challenges, goals, and something to struggle toward."
Don't forget rewards. Back in the day the reward for beating a game was twofold.
1) you got to see the end of the game
2) you got to tell all your friends you beat the game, hahahaha
Nowadays #2 is jax3d because everyone beats every game that can be beaten. That's why multiplayer games like CS are so popular. Since there is no challenge anymore in playing against the computer people play against each other.
Whenever I look for quality in game design I always look to Zelda games. You can go anywhere and do stuff, so its freeform. But you have a predetermined quest, to get all the triforce pieces and beat ganon. Don't make a linear game in which the places you go and the things you do are in a predetermined order, that hasn't been fun since Mario 1 and 2. You have to at least do things like Mario 3 did or like Mega Man did if not what Zelda did. But be sure to give the player a goal, make sure that they know what the goal is and make sure that you give the player enough information in game so that they will be able to figure out what to do.
Make it hard, but don't make it stupid and arbitrary. FF:CC had this problem. The way to get lots of items and things and cast different spells requires the players to pretty much randomly figure stuff out. There were no puzzles to solve like in a Zelda game. You don't make a series of logical thougts and arrive at a conclusion that gives you the hidden shiny. You pretty much either find it randomly, or you read a FAQ. That is piss poor game design right there. I just can't understand. Everyone recognizes certain games as the best, but so few people learn from those games and apply that learning in designing new games. They just make the same design mistakes over and over. Hopefully indy developers will save the day.
The GeekNights podcast is going strong. Listen!
GTA is the perfect example of implementing a "sandbox" approach and still giving the player something to do; the player can, at any time, cut loose from the mission structure and spend some time with General Mayhem, or they can just go forward mission by mission.
I suppose Morrowind is like that, too, only with way too much walking/running for my tastes (even with the Boots of Blinding Speed).
But yeah really really open games are impossible at the moment. Even if you limit the world it seems an impossible task.
Lets examine his own game Black & White (widely reviewed in reviews as an excellent games wich is prove enough game reviews are not worth the paper they are printed on) an "open" game. Except it isn't is it. You were constantly tehtered to this rather crap "city" and its braindead inhabitants. So instead of exploring and teaching your creature you were constanstly trying to manage the population. To much food and overcrowding, too little and they starve and impossible to get it right. My solution, grow plenty and have my creature eat the population. Actually I tried to be nice at first but one of my houses was build wrong with the door apparently under the ground so lots of people were trapped inside. Didn't notice this until I destroyed it since and they all came out burning and I went from good to evil in the time it took my ape to get out the marshmellows.
To many of the current games still require idiotic micro management and restrictions because the developers seem unable to make a functioning AI or world model.
Exactly how many times have you started a mission to save the entire world from thousands of heavily armed bad guys with 1 pistol and 2 clips of ammo? Worst offender recently was Silent Storm. Crack team of commandos at the height of WW2 and you are even told by the guy in the armoury "we got the finest of the british army at your disposal". What you get? 1 submachine gun, 1 rifle a few clips of ammo oh and a bandage. Wheee!
As call of duty and medal of honor have shown some of the best moments in a fps aren't you with 1 pistol against 100 enemies but you and 100 allies against 100 enemies. Now that is the kinda stuff I want to see. Not me the single superman but playing a character in world wich behaves real without all the inconvenient bits of actually being in a war.
Of course this is a lot harder. Give you a single pistol and you will be to busy ducking and running to notice that the dozen or so enemies are not actually a match they just got insane firepower and health points.
So for the next Black & White (still in production?). Get rid of the insane micro management. Make it easier to control the creature (how often did you punish or reward for the wrong thing because the interface lagged?). Make crap like mouse gestures optional or at least talk to people who know how they work (opera does them great for me). Here is a tip. If you draw a cirle too fast it looks like a square to a computer. (to few point of reference)
MMO Quests are like orgasms:
You may solo them, I prefer them in a group.
Peter Molyneux has done great games but lately he seems to have forgotten gameplay. Black and White should have been an excellent game, if only the pet had taken less of a center stage: it's not because you have one (not so) original (tamagotchi) idea that you have to sacrifice everything to it. Black & White could have been a good RTS with a twist.
"He says he watched a 15-year old playtester chat up a woman in town who happened to be the mayor's daughter. He brought her gifts and flowers, talked to her all the time, started hugging and kissing her... and eventually they married and moved in together. Molyneux says he was delighted that a player was exploring this part of the game. Then the playtester talked to the Mayor and asked him to follow him. He took the mayor out to the woods, got him behind a tree ... and killed him! "Why did you do that!?" Molyneux asked. "I figured the mayor was rich, and he'd give all his money to his only daughter," answered the tester. Of course, now the daughter had lots of money, but didn't want to share any of it. So the playtester killed her, too. (Then he moved into the mayor's house!)
I'm anxiously awaiting the release of Fable this summer, but I can't help but wonder: Does freedom turn us into complete psychopaths in the game world, or does freedom allow us to "express ourselves" in the game world without real world consequences?
does this make a good game, or does this make a commercially successful game? And is there a difference? This is definitely a good question--especially considering his analogy. I've read plenty of awesome books which are difficult to sum up into one line--one is going to have trouble doing this with any modern non-genre book.
Peter Molyneux was also responsible for Black & White if you recall.
Guess he learned his lesson.
I currently have no clever signature witicism to add here.
he always dreamed about making games that give people total freedom, but what he discovered is that complete freedom in a game world is boring after about ten minutes.
Not true, I've played both GTA and GTA3 for hours upon hours without ever finishing a mission (or even trying to). Finding creative things to do in the GTA sandbox was enough fun to keep me entertained. Some people don't want missions, or rules. After a days work they just want to come home, and do something fun not virtual chores like fetching Vinnys hookers or something.
It depends on the Zelda game, and what aspect of it you're looking at it.
The first Zelda had a "recommended" order of dungeon completion, offered to the played in the form of those level numbers, 1-9, but overall the player could enter any dungeon at any time, and even complete some of them. Player equipment got in the way of some of that, but in places, if you were good enough, you could penetrate deep enough into a dungeon to get an item early. People have discovered the Magic Key in Level 8 in the first quest is like that, which could be considered a cheat since it makes finding any other keys for the rest of the quest moot.
The second game, however was very linear. There are sequence breaks for it, but they don't seem as sponsored by the designers themselves as in the first Zelda.
Once in the dark world in Link to the Past, the player became able to play and complete some levels out of order. And Ocarina of Time does have a nod to that type of non-linearity -- although the game points you to Death Mountain after completing the Forest Temple, you can actually complete the Water Temple first. (I believe -- going from memory here, it's been years since I've played it.) And of course all the Zelda games (except maybe II) have had little things like extra equipment and heart pieces scattered around to find outside the "main" sequence.
The problem with non-linearity in Zelda games is striking that mix between letting the player do whatever he wants, and hand-holding players who get confused about what to do next. I greatly prefer the former approach, and loved just getting lost in Wind Waker exploring the ocean, but I've coached someone through the whole of that game, who couldn't handle the non-linearity of it. Judging from the sales of hint books, there's a lot of people like that.
I think, in the future, we may end up with more approaches like in the new Metroid, which *appears* to have a strict exploration structure but actually turns out to be EXTREMELY non-linear, almost like the original game. People have written over at GameFAQs that you only really need nine, of the hundred items in the game, to win, that you can beat Ridley before Kraid, that you can skip the Speed Booster, Screw Attack, High-Jump Boots, and so on. It turns out that much of this sequence breaking seems implicitly included by the designers, with the routes that break the "intended" order with less equipment being more difficult to find, and requiring "un-supported," yet still present, moves such as the bomb jump and the wall-kick. It still seems to be impossible to make the player utterly trapped anywhere, and there are special ending pictures for winning with 15% of the items in the game or less.
Here's the thing about Black & White that most of the people who don't like it completely miss.
It's not a game, it's a sim.
I know, it was advertised as a game. I know, it sort of kind of feels like a game because there are all these little challenges in it that you can succeed or fail at. But when it boils down to it, it's not a game.
The key failings of B&W were the interface (grabbing the ground and pulling is ridiculous. You couldn't play the thing well with just the mouse anyway as you needed shortcuts to move from place to place quickly, why not bite the bullet and give it movement keys like a flight-sim, or any FPS with "fly" on), and the incredibly long tutorial that, while it overemphasized all the little problems with the controls, completely negated to tell people how to manage their worshippers.
The key to managing the little worshippers is simply not to micro-manage them. It's hard to do because we're so used to "Demand-Response" interaction in games as being the route to success, not "Demand-Ignore". Those little guys really do learn. If they learn that when they cry "We need food," some mystical force shows up and gives it to them.. guess what they're going to do next time they're hungry? On the other hand, if a couple starve, they soon figure out that if they need food, they'd best go get it.
That Jesus Christ guy is getting some terrible lag... it took him 3 days to respawn! -NJ CoolBreeze