Slashdot Mirror
Carmack on New id Game, Game Theory
An anonymous reader writes "CNN/Money interviewed id Software wizard John Carmack at the recently completed QuakeCon. Among the topics discussed is Nintendo's recent announcement that today's games are too complicated and hard for players. Carmack, surprisingly, agrees, saying 'I agree strongly with that point of view, but I'm in the minority in the PC space. I want a game you can sit down with, pick up and play. [Role playing games], for example, got to where they had to have a book ship with the game.'"
Pc games are better for things like complex role playing games, internet cames, and even action because the keyboard and mouse is alot more flexible then a controller pad. I can move staffe left and right quicker and create my own macro's. Try staffing left, firing a weapon, and then change to the next weapon on a controll pad at the same time? You can do it but it will take longer and your aim will not be as good when doing it.
Quake3 is pretty easy but it would suck on any other platform. For example even if it was an xbox lan enabled release, I could not download mods or new maps. Are there any and I mean any internet games for consoles?
I am sick of the arguement that pc's are for work only and a console is for real games. I consider the pc a rolls royce of gaming and I am fustrated that most game developers now only concentrate on consoles. This is why dukeNukem continues on the ps/2 and why it was killed on the pc. I think executives who only look at installed units per platform and tell the developers to use only x instead of seeing that a particular game is more suited for the pc platform.
http://saveie6.com/
but i always liked games where you were actually led through the game. obviously, there are games that offer virtually unlimited complexity like chess and go, but computer games are quite different. obviously, it is harder to guide a player instead of just creating a bunch of levels he has to get through (which isn't easy either), and arguing about controls is not the right way to go, here.
although a lot of games *do* include tuturials and training missions, etc, it can be difficult to pick up a game because of it. arguably, what a game needs is that each mission/level require a limited subset of skills, and as the game progresses, combine those learned skills, instead of just throwing more monsters at you.
probably my favorite computer game of all time was freespace 2. sure, i like simulations better than FPS and many other genres, but at the same time, it really gave you the feeling of being a part of a "war", mission by mission. the only thing it lacked was cooperative campaigns.
anywho, a lot of modern games lack fantasy: innovation in game play. RPGs have lots of spells, FPSs get you to shoot lots of people, etc. if someone has been playing FPSs or RPGs for a long time, they can get into a new game of the same genre easily. however, when i see a new FPS, i think of it as just that: a new FPS. i want something original!
look at it another way: you are marketting to tech geeks a lot of the time. tech geeks like to build things (like carmack and his rockets) why not translate this kind of interest into a game? mindrover was great for this reason. you actually had to think a little to be good at this new type of game.
BSD is for people who love UNIX. Linux is for those who hate Microsoft.
Carmack has traditionally taken the stance of environment and fast-paced action over character development and gameplay. This is nothing new.
id has long followed the idea that a game should be build around the technology and not the other way around which is simply not the way to create a game, it's the way you create a technology demo or benchmarking software. At one point in time games had plots, scripts, characters, and progression laid out before the engine was written (or incorporated in the case of licensed code). At that point in time it was simply unrealistic to try and write a game completely for the "wow" factor because graphics technology was simply to primitive to impress anyone enough in that regard to buy the game.
I totally agree with John in some of what he says. Games these days are too complex at times. The average MMORPG takes a few weeks, at the least, to really get the hang of. Some RPGs are even rather complex in terms of play mechanics, character advancement, UI manipulation, etc. However, while games may have a steep learning curve, they REALLY are lacking in the depth and difficulty of the games of yester-year.
I remember playing a game called Star Tropics back on my NES when I was 5-6. That game made me absolutely stretch the limits of my fresh-out-of-the-oven mind. Some of the puzzles in the game were so difficult that, at times, the game became a family affair, with both of my parents trying to help me figure out the puzzles necessary to advance in the game. Speed ahead a couple years to Land Stalker on the Genesis. A game in a very similar vein to the previously mentioned Star Tropics. Only 3 buttons were required to play, the menus were, at most, 1 level deep, and the gameplay was fueled by a sword, a jump button, and a special item. There were some puzzles in that game that, literally, took me WEEKS to figure out.
These games weren't difficult in the "cheap" sense that a lot of today's games are. Land Stalker and Star Tropics both presented the answer to a puzzle, but it really took some brainpower. Recent RPGs (final fantasy, Baldur's Gate, NWN, etc.) just don't give that complexity. THey give you hard enemies that take a high level to beat. Whoop-dee-fucking-doo. I don't want to spend hours leveling up in mind-numbingly simple battles! I WANT TO USE MY BRAIN!
Every now and then (maybe twice a year, if we're lucky), a game is released that really dwells in the roots of gaming. My recent favorite games that are hard in the sense that they require brainpower are Big Huge Games' Rise of Nations (which is complex in that it has a HELL of a lot of stuff to do) and the recently released Star Wars: Knights of the Old Republic (Bioware), both of which greatly surprised me with their depth and challenge. KOTOR really surprised me, in that it was an RPG... And it was based on the D&D ruleset... But the game was totally open-ended (left things up to the player), had some tough little puzzles, was action-packed while still staying true to RPG roots, didn't take weeks to get over the learning curve (it didn't even take a day, just a mere hour or two until you really knew what was going on) and didn't try to take up 100+ hours of the player's life.
Games designers really need to quit trying to make "sure bets", and try to innovate genres (like KOTOR and Rise of Nations)! I've had my fill of games like Unreal 2k3, Tomb Raider, Final Fantasy, and other cookie-cutter games. Let's see some INNOVATION AND CHALLENGE! Challenge and depth can, very easily in fact, be presented in a simple and easy-to-pick up manner. If an 8-bit NES game, that had a two-button controller, can make a game that stretches the minds of its players, then why can't a PC or an XBox game?!
Trent Polack
www.polycat.net
Typical for Carmack to agree; it's not as if depth of experience is something in which id has ever excelled. Just point and shoot, people.
System Shock 2, Thief, Deus Ex-- these are the games that are consistently lauded as the masterpieces of the genre, and are as consistently re-played as id's mindless mousekillers. Yes, they were complex. Heretic? Please-- when did you last launch that dog?
And yes, I do *still* play System Shock 2 and Thief.
Game complexity, when done right, enriches and intensifies the experience without making you feel guilty for playing. id never really figured that out, even when handed to them (witness their internal schism over Doom 3 -vs- their first complex RPG). They'd rather someone else do it with their tools. id isn't gunning for the literate gamer. They're looking for the quick buck: they're the Spielberg of game design.
Whatever makes you money, John.
Neither Japanese (Final Fantasy) nor Western (Baldur's Gate) RPGs are true RPGs. In a true RPG, you could generate a character, write your own background, description, etc, and the plot of the game would integrate and work with the character. Unfortunately, we don't have any AIs capable of doing this.
Japanese and Western RPGs have taken different routes, and neither is inherently better or worse. In Icewind Dale, I loved writing really long, descriptive histories for the characters. The thing that annoyed me was that, in the end, these histories meant bugger all throughout the game. Even if I made my character a morose, introspective type, the game would still popup conversation options totally counter to the characters personality. Even though my little fighter was raised by orcs from childhood, he is still forced to react to an encounter with orcs the same way any other character would.
In the opposite way, Japanese-style RPGs weave the character's background into the story very tightly. Because they do this, they limit the gamers choice. It means in Final Fantasy VI, I can't make the protagonist a 6-foot, muscled black guy. I'm stuck with Terra. On the other hand, it means that at all times, Terra acts like Terra, reacts in ways Terra would, and is generally consistent with her own character.
Personally, I prefer the tightly woven character-plots of the Final Fantasy series. But all of these type of games offer this trade off. Consider Baldur's Gate; All your NPCs were pre-generated, your own character had much of his background specified, and, as a result, the story of the Baldur's Gate series can be more tightly woven around the protagonist.
Until someone in AI solves the natural language problem, we're going to be stuck with this tradeoff.
Just because you're paranoid doesn't mean there isn't an invisible demon about to eat your face
The best games have a small set of simple rules from which complex behavior emerges. I think the most classic example of this is Boulder Dash which in it's original form features only a handful different blocks yet the variation between the levels was astounding. I've always thought that the best way to create a game is to start with the basic laws of physics which may or may not be modelled after our universe, then add a few different entities with some clearly defined rules of behavior including the interaction with other entities.
The beauty of this is that the game becomes predictable of the player. He/she will not be frustrated by seemingly arbitrary rules, like in the old Sierra On-Line adventure games where standing on the wrong pixel would get you killed, but instead will understand the action and reaction that lead to the players demise and will learn to avoid it. You want the gamer to go "aah, of course!" instead of "what the fsck?!".
Also, since the game's complexity springs forth from the interactions between the rules rather than the rules themselves, you get what's called "emergent gaming", where the game mechanics appear between the lines through the complex interactions of those rules. This means that although the rules are simple and predictable, you have created a breeding ground for complex behavior goes beyond what the game designer himself may have envisioned.
It's a sad fact that games were more like this before the 3D-card revolution.
I understand why the industry want simpler games as they are trying to expand their customer base which today consist of mostly hard core gamers. Especially on the PC. There are plenty of examples of mainstream hits, but a hardcore gamer will often spend 10 times or more on games than a "causal gamer".
Since games are usually created by gamers who invariably create games that they would like to play themselves I remain confident that there will still be games I'll want to play in 10 years from now.
Simple, instantly playable games is the domain of handheld devices. Complex games fit better on the PC-platform. Consoles are somewhere in the middle. This is linked not only to how we use handhelds/consoles/PCs differently, but also to the technical limitations of the device.
A witty
Keyboards were designed many many years ago as very low bandwidth devices. To save bandwith, many keys reuse the same codes, so that only one key signal can be sent to the computer at one time. The special keys (CTRL, ALT, SHIFT) are given special codes so they can be used in combination with the regular keys.
You can think of it as having 7 bits (allowing 128 keys) plus 3 bits for the special keys. So each time a key is pressed or released, a 10 bit signal is sent to the computer. The computer remembers the last signal, and assumes that if no signal is received, then the keys from the last signal are being held down.
This was important to game writers, because some combinations would not work. If "P" is "move left", and "O" is "shoot", then moving left and shooting would not be possible.
1. Hold "O". Computer sees that "O" was pressed.
2. Hit "P". Computer sees "P" was pressed. It assumes that the "O" must be released.
The special keys did not have this "feature", so they were used for actions, such as shooting, that might be done simultaneously with another action. Moving "shoot" to "CTRL":
1. Hold "O". Computer sees that "O" was pressed.
2. Hit "CTRL". Computer sees "CTRL+O" was pressed.
Keyboard technology may have advanced since the 80s, so these issues may have been solved.
I spend my life entertaining my brain.