Do you feel that game AI has been neglected because of aims to improve Graphics and realism? whats your opinion...
0
Game AI been neglected because of graphics?
Started by Metahawk, Mar 23 2005 10:00 PM
55 replies to this topic
Sponsor:
#2 Anonymous Poster_Anonymous Poster_* Guests - Reputation:
Posted 24 March 2005 - 02:33 AM
Yes, to a small extent. It has more to do with marketing though. It is hard to sell AI (and therefore to get the funding and time to build it).
#3 Members - Reputation: 220
Posted 24 March 2005 - 03:22 AM
Yes totally. Its the next bing thing tough. Because graphics are almost photorealistics now, and that they dont add much to gameplay. See the number of people playing SNES games on emulators... So now developpers will have to focus on AI.
Graphics still sells the game, but AI is what makes it great or not.
Graphics still sells the game, but AI is what makes it great or not.
#4 Members - Reputation: 463
Posted 24 March 2005 - 03:27 AM
IMO physics is going to be the next big thing. Us AI guys will have to wait a little bit longer...
My Website: ai-junkie.com | My Books: 'Programming Game AI by Example' & 'AI Techniques for Game Programming'
#5 Members - Reputation: 122
Posted 24 March 2005 - 05:53 AM
I agree Physics has become an important part of certain genres of gaming. However, I don't think it has done so at the cost of AI. Every project I know of has dedicated AI programmers, and indeed more AI programmers than Physics, and if you look around at who is hiring it is clear how in-demand AI programmers currently are.
The Graphics vs Gameplay balancing-act has eased somewhat thanks the the GPU. Now the graphics has its own processor it freed up a lot of CPU time to Gameplay, Physics, AI, etc.
The Graphics vs Gameplay balancing-act has eased somewhat thanks the the GPU. Now the graphics has its own processor it freed up a lot of CPU time to Gameplay, Physics, AI, etc.
#7 Members - Reputation: 157
Posted 24 March 2005 - 07:31 AM
Quote:
Original post by fup
IMO physics is going to be the next big thing. Us AI guys will have to wait a little bit longer...
I agree. It takes a lot more work to make a noticeable difference in AI compared with the difference that you can see from better physics.
#8 Members - Reputation: 1518
Posted 24 March 2005 - 08:51 AM
Just look at HL2 for "physics candy". I honestly got bored of how many times they tried to show off the physics engine as a form of gameplay mechanism [rolleyes].
As others have said, it's going to be a while before AI takes center stage - but with the advent of these multi-processor/multi-core systems, and the inevitable switch to multi-programming, I reckon AI will get a chance at the big time.... [smile]. Fingers crossed the argument that processing resources are limited will become null and void.
If a couple (or more!) big name games take the time to put some truly powerful AI into things, the sort of stuff that gets people going [oh] at how damn impressive it is, then they'll raise the bar. Raise the bar and others will almost certainly follow...
Jack
As others have said, it's going to be a while before AI takes center stage - but with the advent of these multi-processor/multi-core systems, and the inevitable switch to multi-programming, I reckon AI will get a chance at the big time.... [smile]. Fingers crossed the argument that processing resources are limited will become null and void.
If a couple (or more!) big name games take the time to put some truly powerful AI into things, the sort of stuff that gets people going [oh] at how damn impressive it is, then they'll raise the bar. Raise the bar and others will almost certainly follow...
Jack
<hr align="left" width="25%" />
Jack Hoxley <small>[</small><small> Forum FAQ | Revised FAQ | MVP Profile | Developer Journal ]</small>
Jack Hoxley <small>[</small><small> Forum FAQ | Revised FAQ | MVP Profile | Developer Journal ]</small>
#9 Members - Reputation: 157
Posted 25 March 2005 - 03:50 AM
I don't think its the processing power. I think its a lack of people who are knowledgeable about integrating the state of the art in AI in games. It takes a lot of processing power to create models. This is not a problem because you can do this during the development process where is is available and doesn't have to be done real-time.
Actually running models in game is not really that expensive. Some games like chess do require lots of processing power in game. That is because they are searching a known state space. The state space is not typically known in most modern games so these computationally intensive techniques are not applicable.
I would compare AI in most games more to AI for robots than for puzzle games. There are a set of percepts and a set of actions. The goal is to map a percept vector to an action at each decision point. This mapping is very expensive to develop, but very cheap to implement in game.
I think its more about deciding which AI algorithms are appropriate for various applications in games. A mob in a game could use many differnt AI paradigms:
- Stimulus Response
- Reinforcment Learning
- Emergent Swarm
- ANN
- GA
- Decision Tree
- Expert System
... etc
It really the knowledge about which methods are appropriate for individual application that is lacking.
Actually running models in game is not really that expensive. Some games like chess do require lots of processing power in game. That is because they are searching a known state space. The state space is not typically known in most modern games so these computationally intensive techniques are not applicable.
I would compare AI in most games more to AI for robots than for puzzle games. There are a set of percepts and a set of actions. The goal is to map a percept vector to an action at each decision point. This mapping is very expensive to develop, but very cheap to implement in game.
I think its more about deciding which AI algorithms are appropriate for various applications in games. A mob in a game could use many differnt AI paradigms:
- Stimulus Response
- Reinforcment Learning
- Emergent Swarm
- ANN
- GA
- Decision Tree
- Expert System
... etc
It really the knowledge about which methods are appropriate for individual application that is lacking.
#11 Members - Reputation: 134
Posted 25 March 2005 - 06:00 AM
I definitely think that AI has been neglected, but for reasons other than just graphics. It is true that with today's technology it is easier to create some nice eye candy, and with our fast paced marketing its a lot easier to show off a cool explosion or other graphical element than it is to showcase an AI system, but I think that AI is a fine art. If you make it too "smart" then the player will have a hard time competing against it and won't want to play. If you don't make it "smart" enough then the player will decide that it is stupid and not like it. You have to make it just right, or believable. You have to make it so natural that the player never even notices it. Thats a very hard thing to do.
As for physics, yea, physics are going to be big in the coming years.
Physics Processing Units
As for physics, yea, physics are going to be big in the coming years.
Physics Processing Units
#12 Anonymous Poster_Anonymous Poster_* Guests - Reputation:
Posted 25 March 2005 - 06:24 AM
I dont know but when I talk about intelligent/challenging AI, I'm not refering to the difficulty of a game, or if its too "smart". The real exciting AI is going to be when you can talk to a computer character, and have a "intelligent" conversation. When a NPC can react and learn things, without scriping.
#13 Members - Reputation: 432
Posted 25 March 2005 - 06:28 AM
Forgive me for not having read the entire thread.
Graphics are far from photorealistic. Real-time graphics in games, that is. AI is indeed an important part of a game, even though perhaps more subtle than graphics. What brings you to the conclusion that developers will now have to focus on AI? Even though alot of advancement has been made in the field of graphics in the recent years, it still remains a hot subject of research. It is hardly 'spent'.
In my opinion, this is an oxymoron.
The goal of AI is not necesarilly making the opponent of the player harder to defeat. That is up to the internal machinations of the game, (it's what the difficulty level is for).
Quote:
Original post by Steadtler
Yes totally. Its the next bing thing tough. Because graphics are almost photorealistics now, and that they dont add much to gameplay. See the number of people playing SNES games on emulators... So now developpers will have to focus on AI.
Graphics are far from photorealistic. Real-time graphics in games, that is. AI is indeed an important part of a game, even though perhaps more subtle than graphics. What brings you to the conclusion that developers will now have to focus on AI? Even though alot of advancement has been made in the field of graphics in the recent years, it still remains a hot subject of research. It is hardly 'spent'.
Quote:
Graphics still sells the game, but AI is what makes it great or not.
In my opinion, this is an oxymoron.
Quote:
Original post by tolleyc
If you make it too "smart" then the player will have a hard time competing against it and won't want to play. If you don't make it "smart" enough then the player will decide that it is stupid and not like it.
The goal of AI is not necesarilly making the opponent of the player harder to defeat. That is up to the internal machinations of the game, (it's what the difficulty level is for).
#15 Members - Reputation: 134
Posted 25 March 2005 - 07:48 AM
Diodor, James Trotter, and Anonymous Poster, your right. I had forgotten about things like chatting with the computer, which add an entirely new level of realism to a game, or just using the AI to control an ally or background pieces, like villagers. Also Diodor, I agree with you when you say that the problems need to be designed to be solved by the AI.
#16 Members - Reputation: 100
Posted 25 March 2005 - 09:15 AM
I think that graphics has nothing to do with AI
Some kind of games do not need a photorealistic graphics.
Also ,I do not agree that a too "smart " AI can frustate most of the game users ,as long as it is a "human" like AI (making errors, I mean)
In any case some "niches" of the market would , for sure, appreciate a top AI
The point is that not only commercial games do not use sophisticated AI techniques but even AI "experts" have never produced a game (maybe "Creatures") or at least fragments of game, as far as I know , using sophisticated AI techniques.
Even simple Demos included in AI game programming books are far away from being too "smart".
The conclusion is obviuos, in my opinion.
Some kind of games do not need a photorealistic graphics.
Also ,I do not agree that a too "smart " AI can frustate most of the game users ,as long as it is a "human" like AI (making errors, I mean)
In any case some "niches" of the market would , for sure, appreciate a top AI
The point is that not only commercial games do not use sophisticated AI techniques but even AI "experts" have never produced a game (maybe "Creatures") or at least fragments of game, as far as I know , using sophisticated AI techniques.
Even simple Demos included in AI game programming books are far away from being too "smart".
The conclusion is obviuos, in my opinion.
#17 Members - Reputation: 432
Posted 25 March 2005 - 09:21 AM
Quote:
Original post by Diodor
The problem is AI programmers aren't designers too. Games will have great AI when the problems the AI is supposed to solve are designed to be solvable and are designed to make the AI look smart.
Which problems are you referring to?
The current discussion seems rather diffuse, I must enquire of the original poster: What was your intention with this thread?
Truth to tell, I'm not much of an AI guy, (I've got that phat AI Game Programming Wisdom 2 on my bookshelf, but haven't had time to read it...), but this is an interesting discussion. How does it work in AI programming, compared to, for instance, graphics? Is there a kind of SIGGRAPH where AI gurus present their new brilliant algorithms?
#18 Members - Reputation: 122
Posted 25 March 2005 - 09:29 AM
Quote:
Original post by James Trotter Quote:
Original post by Diodor
The problem is AI programmers aren't designers too. Games will have great AI when the problems the AI is supposed to solve are designed to be solvable and are designed to make the AI look smart.
Which problems are you referring to?
The current discussion seems rather diffuse, I must enquire of the original poster: What was your intention with this thread?
Truth to tell, I'm not much of an AI guy, (I've got that phat AI Game Programming Wisdom 2 on my bookshelf, but haven't had time to read it...), but this is an interesting discussion. How does it work in AI programming, compared to, for instance, graphics? Is there a kind of SIGGRAPH where AI gurus present their new brilliant algorithms?
I am writing a dissertation on games AI and am interested in frequent gamers opinions on the current state of game AI.
#19 Anonymous Poster_Anonymous Poster_* Guests - Reputation:
Posted 25 March 2005 - 07:06 PM
I think the "quest for better graphics" has been both a boon and bane to AI in videogames and its true victim was gameplay. In respect to AI, i'd say it was a boon in that better graphics will call for better AI, in order to mantain a consistent level of suspension of disbelief. It is a bane in that so far graphics have taken most of programmers attention during development, and allocated more resources during runtime. As more memory and multiprocessor systems become available it will become less of a hassle to implement better AI and better AI will become more commonplace. Especially in an era where shooting games are the top sellers, because in the case of shooters better AI pretty much directly translates to better gameplay.
The history of shooters is a fine example, graphics propelled sales and after being wowed by eyecandy gamers wanted more... ie gameplay, AI. More modern shooters where gameplay is still basically standard run'n'gun you see the advancement in gameplay coming directly from AI. Squad Based tactics, flanking, putting heavy armor units in the front line and other such methodologies from your computer opponents.
We are in an era of 'generic programming' where code reuse has become an end into itself, even if from a technical standpoint it would be better to custom code certain functionality. However this same 'generic programming' fits perfectly into the current economic and state (cross platform) climate of the game industry. It costs so much to produce a game, in combination with the probability of it selling enough units to make a profit, it becomes apparent that 'to the metal programming' that was common in the days of old in no longer economically feasible the way things currently are.
So in summary AI will see evolutionary growth/use up until either A. Graphics have become conquered and become trivial to implement. or B. Hardware becomes cheap enough so that programming using cost effective generic methods will allow a better level of AI with a minimal amount of effort to integrate it.
Think of a game like a woman, its not her brain you saw from across the room, that got you to walk over there and start a conversation, although it might be the reason you stay. Or you might stay because the 'graphics' but what was I talking about again... I forget... I'll let someone else finish off this last piece of wisdom.
The history of shooters is a fine example, graphics propelled sales and after being wowed by eyecandy gamers wanted more... ie gameplay, AI. More modern shooters where gameplay is still basically standard run'n'gun you see the advancement in gameplay coming directly from AI. Squad Based tactics, flanking, putting heavy armor units in the front line and other such methodologies from your computer opponents.
We are in an era of 'generic programming' where code reuse has become an end into itself, even if from a technical standpoint it would be better to custom code certain functionality. However this same 'generic programming' fits perfectly into the current economic and state (cross platform) climate of the game industry. It costs so much to produce a game, in combination with the probability of it selling enough units to make a profit, it becomes apparent that 'to the metal programming' that was common in the days of old in no longer economically feasible the way things currently are.
So in summary AI will see evolutionary growth/use up until either A. Graphics have become conquered and become trivial to implement. or B. Hardware becomes cheap enough so that programming using cost effective generic methods will allow a better level of AI with a minimal amount of effort to integrate it.
Think of a game like a woman, its not her brain you saw from across the room, that got you to walk over there and start a conversation, although it might be the reason you stay. Or you might stay because the 'graphics' but what was I talking about again... I forget... I'll let someone else finish off this last piece of wisdom.
#20 Anonymous Poster_Anonymous Poster_* Guests - Reputation:
Posted 25 March 2005 - 07:13 PM
If World of Warcraft is any indication, I'd say "absolutely"
