Jump to content
LCVG

Carmack's Keynote - AI and Physics in today's gaming world. Let's discuss.


Romier S
 Share

Recommended Posts

Carmack comments on next gen consoles, current state of graphics development and a ton more:

 

http://www.gamespy.com/articles/641/641662p1.html

 

Multiple Processors for AI or Physics

 

Proponents of faster and faster processors sometimes argue that now that graphics are reaching their 'peak,' extra processing power can be dedicated to calculation-intensive physics or Artificial Intelligence. (Carmack relates how an Engineer at IBM told him that graphics were basically "done.") Carmack disagrees, seeing that graphics still have a long way to go. "We'd like to be doing Lord-of-the-Rings type rendering in real-time," he states. That's still an order of magnitude more than what's possible with current machines, and Carmack is looking forward to it.

 

That aside, Carmack spent a few minutes talking about Artificial Intelligence as something that can be offloaded to another processor for a cutting-edge game. Carmack is skeptical. AI is a very bleeding-edge science, and it can often be processor intensive, but when applied to games AI is usually a matter of scripting. What game designers want is a way to act as the 'director,' telling enemy and friendly characters where to stand and what to do. This doesn't take a ton of processing power.

 

Moreover, even if you did throw tons of resources toward the AI, it might not be the best thing for gameplay. For instance, writing tons and tons of code to enable monsters to hide in the shadows and sneak around behind the player would be interesting, but often these types of things could be scripted for a fraction of the effort and - for most players - the experience would be just as cool if not cooler. Carmack recounts how players of the original DOOM would think that the monsters were doing all sorts of scheming and plotting and ambushing when, in truth, they were just using the equivalent of one page of C code and running the most basic of scripts.

 

Next: Physics Simulation...

 

So, if we're not going to need a parallel processor to run reams of AI calculations, what about physics simulation? Carmack thinks we're a long way off from seeing any major movement in this area. If you created a huge world where objects all around you were simulated in real physics, it would definitely be processor-intensive. But there are a number of problems.

 

"The problem with physics is that it's not scalable in orders of magnitude like graphics," he explains. There's no equivalent of 'level of detail' in physics - objects far away have to be simulated just the same as objects that are close. Furthermore, you can't decide not to simulate something just because the player doesn't see it on the screen; if you knock over a pile of crates, they have to fall over even if your back is turned. And of course you don't want the physics to work differently based on whether a player is looking at the object being simulated or not.

 

 

Doom 3 had basic physics (you could knock those boxes around) but nothing that substantially impacted gameplay.

 

 

For this reason Carmack thinks we're still a ways away from seeing deep physics simulations being built into a game in such a way that they have a major impact on the gameplay. We'll see a lot of 'non-interactive' physics in the near future - such as realistic smoke pouring across the environment, or cooler simulations of liquid - but nothing major in the way of gameplay.

 

Carmack notes with some irony that in the same way id Software raises the graphical bar for games, other companies are raising the standards for physics, so that id has to integrate simulations into their games or they look dated.

 

He goes on to say that simulating a complex world is a hard problem. "Your game will either be fragile or slow," he says, and it had better do "something amazing" with regards to gameplay in order to make up for that. Carmack says that, in the next generation, he'd rather see processing power go towards making console games run at 60 frames per second rather than doing in-depth physics.

 

Open vs. Closed Console Platforms

 

Carmack's talk changed gears at this point, starting with an aside about Sony. Although he raves about the Xbox 360 development tools, Carmack noted that Sony is making noise about making the PS3 a more open platform. As a big proponent of open source and - well, open anything - the programmer is excited to see if this goes anywhere. His biggest pet peeve with the console market is how closed off it is: you have to apply to be a developer and get a special development kit and get product approval from the hardware manufacturer, etc. (As opposed to the PC market, where anyone can develop and game creators can release content updates, point releases, and so on.) That's just the nature of the market. If the PS3 opens up, it'll be more like the old Commodore Amiga, a platform anyone can use for a variety of applications. Certainly Microsoft will never do this with the Xbox product line, but Carmack is holding out hope that Sony could experiment.

 

Following that train of thought, Carmack asked how many people in the room had HDTVs at home. (I was surprised at how few people raised their hands - from where I sat it looked like less than a quarter of the audience. And these guys are hardcore!) Carmack pointed out that any sort of plan for using a console or set-top box as a computing device really requires the high resolution of an HDTV or computer monitor to be effective. Bringing the topic around full circle, he pointed out that Microsoft may enact a policy of requiring all Xbox 360 games to be rendered at HDTV resolutions, regardless of whether or not that decision helps with the design of the game. He says that some marketing person probably made that decision, another thing that bothers him about the console industry.

Link to comment
Share on other sites

For this reason Carmack thinks we're still a ways away from seeing deep physics simulations being built into a game in such a way that they have a major impact on the gameplay.

 

 

Ever heard of a game called Half-Life 2 Mr. Carmack? That game was built around its great physics engine. Physics effected gameplay a TON in that game. I'd like to know what Carmack means by "major impact on the gameplay" because IMO physics DOES have a MAJOR IMPACT on GAMEPLAY in HALF-LIFE 2. Not to mention HL2 is a gazillion times more fun and rewarding than anything Carmack and company have been responsible for but I won't get into that. That is just a personal opinion and we all know the rule regarding those... ;)

 

With alot of next gen engines touting realistic physics as being as important as graphics how can Carmack think we are still a ways off? I must be missing something here. Apparently he either thinks current state of the art physics (like HL2 and upcoming next gen titles) are not good enough or not worth implimenting in games? I mean he can't be that detached from the gaming industry that he hasn't played Half-Life 2 or seen the promising upcoming titles using realistic physics....is he?

 

*Edit: I just realized Carmack may have been referring to ID's future games with regards to having realistic physics. I took that quote as commenting on the industry as a whole. He does say right below the quote I posted that other companies raise the bar on physics when ID raises the bar on graphics so I guess having re-read that the above quote does sound like he is speaking about ID and not the industry as a whole. *shrug* He should have specified one way or the other. :)

 

Moreover, even if you did throw tons of resources toward the AI, it might not be the best thing for gameplay. For instance, writing tons and tons of code to enable monsters to hide in the shadows and sneak around behind the player would be interesting, but often these types of things could be scripted for a fraction of the effort and - for most players - the experience would be just as cool if not cooler.

 

 

I'm not even going to touch that quote.

Link to comment
Share on other sites

I'm not even going to touch that quote.

 

Why not?

As Carmack mentions, most of what gets called "AI" is scripted. At this point in gaming if scripted events are well executed in-game our suspension of disbelief in the game will cause us to think that NPC really acted on his own.

 

As for his physics comments...I think he was talking about simulating real physics. The Havok physics engine in Half-Life is just baby steps. Look at all the items in the game you can't manipulate. It doesn't make sense that I can lift barrels and boxes and throw them around but I can't break all the windows? I can beat down wooden crates and obstructions with my crow bar but not locked wooden doors?

Link to comment
Share on other sites

9.999999 times out of 10 when something cools happen in a game its scripted. Carmack is right on the dot that physics in current games are nothing but a fancy frill, a gimmick if you want. You can take the physics out of your example, Half-Life 2, and it is still a great game. The only game I can think of that requires physics is The Incredible Machine.

Link to comment
Share on other sites

That is just a personal opinion and we all know the rule regarding those...

There is no rule regarding stating your opinion here. If you have something to say, say it. This is the thread to do it in. I'm sure any Id fan would be happy to discuss any point you have to make Cyberwoo.

 

EDIT:

 

You can take the physics out of your example, Half-Life 2, and it is still a great game.

Sure, but I think Cyberwoo's point is that the addition of the physics makes HL2 that much better of a game. There are a few puzzles in HL2 that make great use of the games physics engine for instance. Also, the use of physics greatly enhances the combat in the game. The ability to use the environment as a weapon (say by running the ski-boat under a ramp therefore tearing out the support beams and causing the enemies to fall to thier death) just makes for one hell of a good time.

Link to comment
Share on other sites

Why not?

As for his physics comments...I think he was talking about simulating real physics. The Havok physics engine in Half-Life is just baby steps. Look at all the items in the game you can't manipulate. It doesn't make sense that I can lift barrels and boxes and throw them around but I can't break all the windows? I can beat down wooden crates and obstructions with my crow bar but not locked wooden doors?

 

:tu

 

It's like in RPGs when you can't just hop over a little drop or rise in the terrain, or you can summon Bahamut to drop a giant fireball attack on your enemies, but you can't burn down a wooden door with it. Kinda silly.

Link to comment
Share on other sites

There is no rule regarding stating your opinion here. If you have something to say, say it. This is the thread to do it in. I'm sure any Id fan would be happy to discuss any point you have to make Cyberwoo.

 

After re-reading my quote I can see how you could get the impression I was referring to LCVG. Not at all. I guess I shouldn't have used the word "rule". I was referring to the well known phrase "opinions are like assholes....everyone has them". I'm never afraid to give my opinion on here. No need to worry about that. ;)

Link to comment
Share on other sites

Sure, but I think Cyberwoo's point is that the addition of the physics makes HL2 that much better of a game.

 

Absolutely. The kickass physics in HL2 is just one of the features that seperates HL2 from the rest of the first person shooter crowd. You take those physics out and replace them with *insert your standard first person shooter here* type physics and it isn't the same. A good example of this is the gravity gun ripoff in the add-on for Doom III. It isn't nearly as fun as the one in HL2. The reason is quite simple. The pysics in HL2 react alot more realisticly then in Doom III. So do the physics in HL2 make the game? Not at all. But they surely help put it in another class. Combine that with the amazing sound, level design, pacing, storyline, characters, etc and you get one kickass game with few if any flaws.

 

You can call the physics in HL2 gimmicky all you want but to me they add ALOT to the game and just help make the game FEEL better. I just have a big problem with things like physics, AI, storyline, etc taking a backseat to eye candy so these type of discussions always fire me up. :)

Link to comment
Share on other sites

Look at all the items in the game you can't manipulate. It doesn't make sense that I can lift barrels and boxes and throw them around but I can't break all the windows? I can beat down wooden crates and obstructions with my crow bar but not locked wooden doors?

 

The biggest problem with this line of thinking is that once you make the physics of aa game completely realistic -- in other words, you can realistically destroy anything you want in a game given you have the right tool to do it -- the game designers have to account for everything on the other side of that object. For instance, if you could destroy every wooden door in Half-Life 2 with your crowbar, the designers have to create that space on the other side. That means you'd probably have another room with another wooden door and windows that you would need to break, even if the structure of the level doesn't benefit from creating those places at all.

 

Or better yet, let's say that you could use explosive barrels and the Gravity Gun to destroy buildings in Ravenholm. What happens when you destroy a building you needed to get on top of to progress? Or better yet, what happens if you destroy the buildings that act as a wall to keep you within the game world? You don't really expect Valve to create an entire world beyond that perimeter, do you?

 

My point is that realistic physics can be a great addition to a game, but limits have to be placed on what you can do simply because you need to serve the needs of making a fun game. Games are not realistic representations of our world, and don?t I really want them to be. I expect rules and boundaries to what can be done, which is why I don't necessarily get upset in a game like Haunting Ground, where the main character cannot step over a velvet rope. I know it's not realistic, but I'm willing to accept that in the rules of the game, I have to go around rope.

 

As for Carmack?s comments regarding A.I., isn?t most of what we consider to be A.I. anyway just a list of commands that tells a character, ?When X happens, perform Y??

Link to comment
Share on other sites

The biggest problem with this line of thinking is that once you make the physics of aa game completely realistic -- in other words, you can realistically destroy anything you want in a game given you have the right tool to do it -- the game designers have to account for everything on the other side of that object. For instance, if you could destroy every wooden door in Half-Life 2 with your crowbar, the designers have to create that space on the other side. That means you'd probably have another room with another wooden door and windows that you would need to break, even if the structure of the level doesn't benefit from creating those places at all.

 

I didn't intend to imply that the limits on physics in HL2 were bad or limiting. That's just the state of game design in the here and now. We could really derail this thread on a design tangent and get into the linearity of games but that's probably best saved for another thread (an interesting one to be sure).

 

My point was that the Havok engine is just the tip of the iceberg. Physics in HL2 are great an a lot of fun but because of those limitations you mention they are really stranded as being gameplay features at this point. Games will need to overcome design issues before physics can get really real -which I think is what Carmack was saying (at least in part).

Link to comment
Share on other sites

The biggest problem with this line of thinking is that once you make the physics of aa game completely realistic -- in other words, you can realistically destroy anything you want in a game given you have the right tool to do it -- the game designers have to account for everything on the other side of that object.

 

I think the biggest thing your point raises is once the physics keep going and going and going further, it adds significant amounts of processing that needs to be done to deal with the borders realistic physics breaks down. Carmack mentions that, and he is skeptical that all that extra processing has that much of a payoff. He's not just saying that graphics are better, and phsyics are shit. To me, he is questioning the payoff of keeping on adding more and more with the processing power developers have readily available to them right now.

 

I think the people throwing HL2 out so fast are overlooking his point that he just doesn't feel it is as easy to measure "how much" phsyics a game should have given the amount of processing power available to us. HL2 kept it fairly tightly bound with the gravity gun and what it could do, but if you start going beyond that and making fully realistic worlds at what point is that worth it for the possible limitations you have to place on other aspects of the game such as graphics, map sizes (adding more of the ever exciting "loading" screens as you run through an area - we all love that, right?), limiting the AI, scripted or on-the-fly processed enemies placement/movement, limited types of attacks as each one would need processing to do on the fly, etc.

 

I don't think graphics are the be all and the end all - I still run a Geforce 4 and my console of choice is a PS2 - but I think he is on to something by suggesting getting games running at 60 FPS before you start adding processor intensive AI and physics. You can have all the gadgets in a game in the world, but if I have a framerate issue (with good or bad graphics quality) when playing a game, that matters a lot more to me than if I can pick up a toilet and throw it at somebody. I hate video chop so much :mad:

Link to comment
Share on other sites

That's just the state of game design in the here and now. We could really derail this thread on a design tangent and get into the linearity of games but that's probably best saved for another thread (an interesting one to be sure).

A rather excellent idea and one request I'd be quite happy to oblige. I've gone ahead and split this from the original Id relevance thread. Feel free to continue this discussion without worry of derailment gents.

Link to comment
Share on other sites

Doom III is everything ID stands for (for better or for worse). A cutting edge graphics engine wrapped around a standard "run n shoot everything that moves" type of FPS. I look at the games ID releases as not much more than a "hey look what our latest engine can do" type of game. A playable tech demo if you will. I'm not dissing that at all but the fact is the gameplay has remained the same since Wolfenstein. Surely I am not the only one who is tired of that type of gameplay. That being said I still look forward to every new game ID releases for one reason....licensing. When ID releases a game using its latest engine you know other developers are chomping at the bit to use ID's new engine to create their latest and greatest game(s). Take Valve for example. They take technology from ID, apply their own little tweaks here and there and then focus on gameplay, gameplay and more gameplay. Half-Life and Half-Life 2 are two of the greatest FPS games of alltime because of this. So in that regard I will be forever grateful to Carmack and ID (and the Unreal boys for that matter) for the engines they continue to create because without them the HL's of the world would never see the light of day. That should clear up any "woo is always dissin on ID games" type of thoughts. Just thought I would mention that before I make anymore comments regarding the topic at hand. ;)

 

Anyway I think Carmack brings up alot of valid points. For example you always want a consistant solid framerate without any noticible slowdown. Things like good physics, realistic A.I., etc should never be implimented if it is going to cause the game to have framerate issues that hinder gameplay. A bad framerate ,especially during heavy action sequences, can just ruin a game. So when Carmack says he would rather focus on getting a solid 60fps out of a game rather than dumping a bunch of code into things like physics, AI, etc I can see his point. However I also think it is a bit of a copout. Idealy you want both. All the physics, all the A.I., and state of the art graphics al running at a solid 60fps. Obviously this doesn't happen for a variety of reasons. The mosy obvious limitation (and the easiest to blame it on) would be the hardware. Second would be the time it takes to impliment all those things and time is money....

 

I guess when its all said and done I would rather Carmack focus on creating engines that look great and run at a solid FPS because in the end other developers are going to wind up adding the physics, the AI, and other gameplay effecting content that I crave later on anyway. My thing is if others can do it (using his core technology) why does Carmack make it sound like its impossible/impractible to do at this time? Half-Life 2 didn't have any framerate issues for me. Granted it wasn't using the latest eye candy but as a whole it looked better (the outdoor environment, the textures, the animation, etc) and played a million times better than something like Doom III. In the end aren't better games what it is all about?

 

So in summary my "Jerry Springeresque" final thought should be this. I favor fun and innovative gameplay using slightly dated graphics technology over tired and "been there done that" gameplay using the latest graphics technology. This is why I have a problem when someone like a Carmack downplays things like physics and AI in favor of creating more eye candy. I want the whole package running at 60fps and if that means less eye candy/state of the art graphics technology than so be it.

Link to comment
Share on other sites

That about sums up ID's design philosophy for me as well. Great engine makers, mediocre game makers. Opinion of course but, an opinion shared by Carmack himself.

 

I remember back in the day, Carmack was mentioning how they come up with the game engine, then tack on a flimsy story. That was Doom/Quake days, and straight from John's mouth.

 

Didn't ID hire some sci fi writer to come up with the Doom 3 story? Could've swore I read that in my "making of" book. Even so, I still have to side with those that feel Doom III has a weak story and is basically just a playable tech demo fused to a "monster locker" game.

 

I'm holding out for Quake IV, in terms of actual gameplay being much improved. The graphics and even the graphics scheme damn near look identical to Doom III, so the only thing that has me excited is the team play aspects of Q4.

 

If Raven can nail it, I'll be a happy man. I don't think that anyone can seriously argue that the Doom 3 engine isn't attractive looking, so graphically speaking, I can't imagine being disappointed by Q4. Unless, having very similar graphics to Doom 3, could be used as a disappointing feature.

 

Anyhoo...while having much respect for ID, I have to say they've lost some of their luster, in my eyes, since there are other companies now that make as good as engines, or better (new Unreal engine?), with MUCH better stories, ala Half Life 2.

 

Interested in seeing what the future holds for ID, especially with Carmack hinting, here and there, that he's tired of tech design.

Link to comment
Share on other sites

I'm probably one of the few on this board that really likes the story behind Doom 3. Given the basic premise, I can't really think of any good reason to make it more complex so to speak. It's Doom Guy versus demons trying to kill him. I like the bits and pieces like the PDA's that flavor the game, and it doesn't complicate it beyond that. Otherwise, I wouldn't doubt other games have better stories, but I'm happy with Doom 3 and the ambience the game's setting provides.

 

Honestly, I really don't see the difference between AI and scripted actions/reactions. I really don't see the difference between what we do either. Far as I'm concerned, we're pretty scripted as well. Get hungry, eat. See hotness, :drool . etc. etc. For that matter, do we really need to make a program so complicated that the enemy really has to make a choice between fighting, fleeing, or shitting their pants? I mean that would be the only advantage to having "intelligence" in a program.

 

I agree with the posts about physics. It's fun seeing how far you can go, but there really needs to be limits.

Link to comment
Share on other sites

Actually, the more I think about it, the story isn't bad at all, I just feel that it was implemented in a way that I really didn't feel the impact from it. Maybe it was the overall BOO!!! scares, that wore my patience out. Nothing in there changed over the course of the game. Same thing, basically, over and over.

 

The thing that kept me going was that I just wanted to see what was around the next corner. :)

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...