A Gamer's Manifesto

Krimszon writes "The top 20 things you always knew were wrong about games, but were afraid to talk about, since you thought that was just the way is was."

Re:Ahh.. jumping puzzles...

[0x461FAB0BD7D2]

(Good) AI and graphics require huge amounts processing power. The fact is that good graphics attract gamers more than good AI. Look at E3 2005. The demos were all about graphics and how realistic they were.

Games use finite-state machines for AI simply because the range or variety of moves in each game is limited. And for each move or state, there is a logical reaction, not unlike rock-paper-scissors. It's hard to move forward on intelligence without expanding the variety of plays. Black and White worked because the range of abilities was far greater than any FPS.

However, for people like myself that prefer strategy and thinking over gfx, we still have the time-tested games of chess, go and sudoku.

Re:Better AI: do you really want it?

[MoonFog]

The AI in OF isn't what I would call genious. Rather, it "cheats". I've experienced sneaking through the woods wearing nothing but black. Then lay down and try to snipe someone from 300 feet away. If I miss, he turns and shoots me with a goddamnt AK47 in pitch black without knowing where the shot came from.

OF is a great game, but as you say, gets boring real quick.

Re:All your base are belong to me

[kyrcant]

the bulk of the gaming community is NOT teens, its young men, 18-30.

Re:Ahh.. jumping puzzles...

[angle_slam]

If you never finished Half Life, you don't understand the complaint: after finishing a bunch of levels of the "Greatest FPS Ever" you get to a level that is merely a jumping puzzle where you have to jump from one thing to another. You'd swear you were transported to a Mario game. Half-Life 2 doesn't have that flaw. There is a tiny jumping part, but it is trivially easy, unlike HL 1, which was surprisingly hard.

Re:Ahh.. jumping puzzles...

[Various Assortments]

Alice had a jump reticle that looked like a pair of ghostly shoeprints. You always knew if you were going to make a jump or not.

Re:No, really.

[Rirath.com]

No, really. I want to know which company patented the loading-screen game.

Last I heard, Namco owns loading screen games on the PSP. Not sure about in general though, but it's a good guess. Here's a quote from Game Developer magazine, interviewing EA's Dave McCarthy:

BS: Do you forsee anything like minigames during the loading screens?

DM: Minigames during loading screens is actually patented by Namco, so they're doing it!

Re:best games are often the cheapest

[Koiu Lpoi]

Roguelikes in general are really quite amazing. I prefer Angband, but we've all got our tastes.

For an absolutely amazing game series, try Geneforge [spidweb.com]. Low on graphics, but includes almost non-linear gameplay in an incredible RPG. Best of all (I think) they give you a HUGE demo to play around in - the first quarter of the game or so, maybe a bit less. And, there's 13+ endings in each of the games, so you keep coming back for more. Completly awesome games.

Your Math is Wrong

[barureddy]

1x cdrom is around 150k/s while 1x dvd rom is something like 1350k/s. Therefore once the math is done, the dvd drive is about twice as fast as the cdrom drive at their respective theoretical max speeds, which are rarely reached.

Re:A game developer's response...

[SilverJets]

daVinci1980

3. Don't bullshit me about your graphics
We wouldn't have to, except that by the logic in argument 2 this seems to be the #1 thing that people care about. You vote with your dollars. Your mouth is saying "graphics don't matter" but your wallet says "grapihcs are all that I care about. Shit in the box as long as the graphics are top notch." Doom 3, Unreal 3, Half-life 2... All top sellers because of their stellar unrelated gameplay?

Reread the article. You are missing the author's point. He is not arguing on this point that there should be more to a game than top-notch graphics. He's arguing that game publishers/developers should stop showing us consumers just the cut scene graphics and telling us this is what the game looks. If its a cut scene graphic, tell us its a cut scene graphic and then show us actually game-play graphics as well. I have lost count the number of times I have been severely disappointed by a game because all the box had (or even magazine advertisement) was the cut scene graphics. Then when I load it, I find that the actual game-play graphics (you know those graphics that I will be staring at for 99.9% of the time I am playing the game) look like they were drawn by a bunch of retarded monkeys. Its 2005 for crying out loud. Make the actual game graphics look like the cut scene graphics!!!

Re:Gamers never know what's good for them

[fbjon]

I always start out a game on easy. Good example: Half Life 2. So much nicer to have a relaxed game of shooting the crap out of hapless minions, while looking at the graphics and surroundings, checking out the details, wondering if it's possible to reach that ledge over there somehow. I don't want a game to stress me out constantly.

And if the game is good enough to play again, I'll play it again on a higher difficulty.

Planescape: Torment

[why-is-it]

Grim Fandango (1998) was the last *GREAT* Adventure game.

I really liked the art deco style of Grim Fandango, but it was not the last great RPG. That title belongs to Planescape: Torment (1999). It has storytelling like nothing else before or since. What a shame it never sold well...

Re:Why I bought HL2...

[daVinci1980]

Your post is very good...

However, there are a few things you should consider about raytracing.. While it is easy to do in parallel, it is also very limiting from a hardware standpoint. The problem is that those pesky rays can go anywhere once they hit a surface, including directions that you didn't intend for them to go in.

This means that more data has to be loaded in the scene (because the rays can even go back behind you or off perpendicular to the scene). There's also the problem of what the data storage is going to look like. IE, in order for this to be expressed in a way that is going to be extremely efficient to a GPU, you're going to have to come up with a common data format that works for all applications that intend on doing raytracing via a hardware solution.

Finally there's the problem that raytracing is O(resolution). That is to say that 800x600 is exactly twice as bad as 640x480. 1600x1200 is 4x worse than 800x600, and any sort of antialiasing on top of that is pure pain.

As for your other (implicit) question, "Why don't interesting games use interesting game engines?"

Because they cost an arm and a leg. Last time I checked, a license for the upcoming Unreal Engine 3 cost between 700K and 1 million dollars. Quake 3 was 1 million, without support. HL2 is in the same ballpark.

Considering that the "interesting" games often have a budget of less than a million bucks total, it's pretty easy to see why it's hard for the little shops that come out with good games to buy the big engines.

Re:CIVILIZATION

[patio11]

True, but its sort of a cop-out. Galactic Civilizations manages a MUCH better AI than CivIII by the conceptually simple, programatically difficult solution of multi-threading the AI code, which means that when you take five minutes and hit end the AI already knows what he's going to do on his turn. In Civ, the AI only starts thinking when the end button is pressed, and because its a real-time application and the user won't tolerate a 5 minute thinking period between turns, so instead it tries to cram all the calculations in to 5-15 seconds, which means you really have to skimp on your algorithms (like, for example, cruddy pathfinding so you can't efficiently use rail networks despite the fact that that should be a 100% solvable problem).

Re:A game developer's response...

[daVinci1980]

I think what you fail to realize is that the graphics are very much a part of the marketing campaign.

If a game looks like crap and people don't get to see any screenshots before release, they're not idiots. They'll realize that the game looks crappy so all they're getting to see is marketting materials and concept art (assuming that the team has a good concept guy).

As far as it goes though, this is simple supply and demand... There is a demand for crappy titles and sequels, because people keep buying them. As long as their is such a demand, supply will exist, or will rise to exist. Even if every single video game developer quit his job today, tomorrow you'd hear about all the new startups. And they wouldn't be any different.

The ball is in the hands of the consumer. To think that any single individual in the game development community has the power to affect the kind of change that gamers are calling for is silly.

Re:HALO

[jguthrie]

The problem with talking about how hard it is to do AI in games is the fact that game implementers don't need to do actual AI (like Douglas Hofstadter's group [indiana.edu] is trying to do) but to create NPC's that behave more realistically. I mean, if a cockroach, who has no more than six brain cells, can figure out how to hide if you're chasing it, then it should be no problem for a programmer to figure out how to make an NPC do it.

It occurs to me that, successful human armies don't let their soldiers wait until actual combat and rely on them to figure out what to do in any given circumstance. Instead, they attempt to enumerate the sorts of situations their soldiers are likely to find themselves in and attempt to train those soldiers to handle each of them. Emphasis is placed on recognizing the sort of situation they're in and reacting as they were taught in training. That sounds perfectly amenable to the sort of "rigged demo" approach commonly used in games.

Re:HALO

[MBraynard]

A friend of mine is working on AI for the military - he got a PhD from the MIT AI lab. Trust me, from talking to him and others, it's MUCH harder than you think.

The 'smartest guys in the room' cannot make an AI as smart as a roach - which has many more than six brain cells. See the recent competition to design a bot that can cross a small stretch of desert in CA.

Re:Ahh.. jumping puzzles...

[Eugene]

Sadly it's true.. I've been wanting a game with good AI for years (turn based strategy, or RTS), but so far I've seen there's nearly nothing has good AI. either the computer cheats by peeking at what you are doing, or it's simply have more resource then you do. but the modern processor do have a lot of processing power to implement a good AI. just that no coder willing to spend the resource/time/money to do it right.

but if you look at the AI topics/articles/publishing around the net and library.. most of them only center around a few topics (chess come to mind, and chess's AI isn't really true AI anyway, more of exhausted searches). Maybe it's time for programming to head to a new direction.....