Are you ready for Physics Chipsets??

JSL · Post by **JSL** » Thu May 12, 2005 7:09 pm

Ageia thinks we are and Rise of Legends is already going to be supporting it. Here is a link to the article:

Ageia PhysX Chip

I have to wonder how many bells and whistles will be disabled if you play a game without the chip in the PC. I also have to wonder how many other game companies are going to make use of this chip. Also, the possible price for it seems pretty high at $100 - $400.

What do you think? Will you buy one?

TheMix · Post by **TheMix** » Thu May 12, 2005 7:15 pm

Absolutely. Just probably not right away.

CeeKay · Post by **CeeKay** » Thu May 12, 2005 7:42 pm

didn't they already do this with the 'math coprocessor'?

Orpheo · Post by **Orpheo** » Thu May 12, 2005 8:06 pm

If it catches on and a lot of games use it, then sure. If it doesn't, then no. This could be just as important as a 3d card in the future, or it might just turn out to be like...like...some device that nobody buys anymore. Like force feedback mice?

knob · Post by **knob** » Thu May 12, 2005 8:36 pm

It's a good idea, but the price is a real turn off.

Sometimes I wish I could go back to the sprite days of Doom and Duke 3D.

noun · Post by **noun** » Thu May 12, 2005 8:56 pm

This nonsense will make me quit gaming altogether. I want the games I play to be fun, engaging, have an interesting storyline and have a decent ending. Physics accuracy is somewhere around #325 on the list.

On the other hand, I find it highly amusing that the mere existence of this chip is proof that the main problem with game quality today are their developers trying to get their 3D engines to behave properly...

Greg Wak · Post by **Greg Wak** » Thu May 12, 2005 9:25 pm

Only if devs make me. I was hoping to get a upgrade respite now that cpu freq has hit a wall. Silly me. First this and today nvidia claims their next gpu will double the power of the 6800 family. As for the physics chip I won't get one for the first big game that uses it, but when most start using it I know I will be forced to. Sigh.... At least it will probably take a while.

Kraken · Post by **Kraken** » Thu May 12, 2005 9:31 pm

I just sprang for a new monitor and vidgie card, and I'd like to replace my smaller hard drive, too. A gee-whiz chip is #326 on my list. Let me know when you buy yours, noun.

Blackhawk · Post by **Blackhawk** » Thu May 12, 2005 9:51 pm

I like the concept, but I'd just as see it built in to the next-gen gaming motherboard or video card than as a standalone card. Still, as someone said, if enough games out there support it, and if it isn't something that needs upgraded annually, then I'm all for it. Maybe it'll free up some CPU cycles for some better AI.

hitbyambulance · Post by **hitbyambulance** » Fri May 13, 2005 4:10 am

definitely will wait for second-gen, at least.

i remember back in 8th grade (1989 or so) thinking that what would be great for PCs would be a "SpriteBlaster" - an add-in card that allows actual hardware-generated sprites for games, just like multimedia computers, arcade machines, consoles....

Sterling · Post by **Sterling** » Fri May 13, 2005 9:43 am

noun wrote:This nonsense will make me quit gaming altogether. I want the games I play to be fun, engaging, have an interesting storyline and have a decent ending. Physics accuracy is somewhere around #325 on the list.

On the other hand, I find it highly amusing that the mere existence of this chip is proof that the main problem with game quality today are their developers trying to get their 3D engines to behave properly...

It isn't just physics accuracy but how many objects can be treated as individual physics entities.

I can think of a ton of fun things games could do if they knew that every brick a wall could be pulled out, or that if you cut a tree just right it'll fall a certain way, onto a player or over a river.

I'm sure there will be plenty of games that use it just to say they did, without actually considering gameplay, but I do see some potential.

Gizah · Post by **Gizah** » Fri May 13, 2005 10:11 am

The prospects of truly totally deformable terrain are quite cool, especially for shooters. The days of scorch mark decals on the picket fence where your rocket launcher tagged it will be all over.

Arcanis · Post by **Arcanis** » Fri May 13, 2005 2:30 pm

looks like i've got to wait on building new tower or build a new one in a short time. I love games with realistic physics but i'm biased i loved physics in school. I even wrote a couple of programs to do some of my calculations for me, the teachers said it was good because if i understood enough to make the program i had to understand enough for their class. Now i have a chip devoted to my quanti and bolistic progectiles muahhahahaha!!!!

LawBeefaroni · Post by **LawBeefaroni** » Fri May 13, 2005 3:09 pm

noun wrote:This nonsense will make me quit gaming altogether. I want the games I play to be fun, engaging, have an interesting storyline and have a decent ending. Physics accuracy is somewhere around #325 on the list.

Physics didn't contribute in a huge way to the fun of HL2, FarCry, MP2, etc? Some of gaming's best moments for me have come from unscripted physics events.

I'll agree with you that forcing gamers to buy a whole new card just for the physics is too much. The games I listed above have excellent physics despite computing power limitations.

Blackhawk · Post by **Blackhawk** » Fri May 13, 2005 3:39 pm

LawBeefaroni wrote:Physics didn't contribute in a huge way to the fun of HL2, FarCry, MP2, etc? Some of gaming's best moments for me have come from unscripted physics events.

It isn't just that, either. Physics haven't been a major gameplay component with the exception of gimicks like the gravity gun simply because they're so limited by what's out there. They're mostly used for immersion - things falling realistically. A half-hundred interacting objects is about the limit. That lets you knock over crates, but cover a table with food and knock that table over, and you're pushing it. I've read that a physics card will raise the limit into the tens of thousands of objects interacting with each other at once. Imagine blowing a car into the air and having it come crashing down on a house in which every piece of furniture, every brick, and every piece of wood is able to interact with each other, all at once. You could create the kind of game that the original Red Faction Geo-Mod system wasn't able to pull off.

Peacedog · Post by **Peacedog** » Fri May 13, 2005 3:42 pm

I've read that a physics card will raise the limit into the tens of thousands of objects interacting with each other at once. Imagine blowing a car into the air and having it come crashing down on a house in which every piece of furniture, every brick, and every piece of wood is able to interact with each other, all at once

Hopefully, this will be accompanied with that wonderful Silent Storm "calculating path. . ." message that happens every other time I shot someone near something breakable, miss, and tag the object. And if we're *really* lucky, we'll get double our waiting pleasure.

LawBeefaroni · Post by **LawBeefaroni** » Fri May 13, 2005 4:02 pm

Peacedog wrote:
I've read that a physics card will raise the limit into the tens of thousands of objects interacting with each other at once. Imagine blowing a car into the air and having it come crashing down on a house in which every piece of furniture, every brick, and every piece of wood is able to interact with each other, all at once
Hopefully, this will be accompanied with that wonderful Silent Storm "calculating path. . ." message that happens every other time I shot someone near something breakable, miss, and tag the object. And if we're *really* lucky, we'll get double our waiting pleasure.

If you don't have the physics card, I suppose could happen. You don't want that to happen. Preorder the card now.

For some reason, I'm thinking about that Hercules graphics card I have mounted on my wall...

Caine · Post by **Caine** » Fri May 13, 2005 4:18 pm

this could be very cool for gaming in general, but it would only amount to much if the games used physics in believable and useful ways. deus ex 2 had some pretty decent physics, but very unrealistic in action. i would love for this to take off and give us all some incredible games to play in. this has more potential than the old vibrator mouse, but that's not saying much. iir, the only game that used it was black and white, and that was only for one small quest. hardly worth the expense or the effort.

Peacedog · Post by **Peacedog** » Sat May 14, 2005 11:18 am

If you don't have the physics card, I suppose could happen.

It actually does happen in Silent Storm.

You don't want that to happen. Preorder the card now.

I doubt that would help in silent storm.

baron calamity · Post by **baron calamity** » Sat May 14, 2005 11:37 am

I'm reead. Anything new to add something. I was an early adopter of 3d cards as well. Sure the first few were clunkers but it was still fun.

Sparhawk · Post by **Sparhawk** » Mon May 16, 2005 11:07 am

Physics plays a HUGE roll in the fun factor. Immersivness comes with believing that the environment you're in could exist. Driving a car that feels like it isn't on the ground vs. one that feels like it is on the ground are good examples of how physics affects our quality of gameplay. Imagine a Rally racing game with no physics. Or a shooter without physics (would you really be satisfied with the bodies in HL2 just going from a standing position to laying on the ground without any realistic movement?) What about when you enter a room in a game and shoot at a bunch of items laying on a counter and they don't move? You just blew up a tank with your rocket launcher but when you fire said rocket at a drinking glass sitting atop a bar said glass doesn't budge; is that immersive? Does that excite you?

I think many take for granted how much physics adds to gameplay and do not realise how large a part physics plays in our everyday gaming.

noun · Post by **noun** » Mon May 16, 2005 1:04 pm

I disagree - this new focus on the physics chipset, if it catches on, will just be another reason to move the focus away from game design and more towards photo realism. I maintain that this sudden importance on physics is particular with 3D games and the challenges in trying to code a realistic 3D environment. There has got to be a happy medium between one of Jeff Vogel's games and a Lord of the Rings computer generated battle scene, preferably the option that leans towards better gameplay.

Jeff V · Post by **Jeff V** » Mon May 16, 2005 1:12 pm

noun wrote:I disagree - this new focus on the physics chipset, if it catches on, will just be another reason to move the focus away from game design and more towards photo realism.

Why do you consider game design and photo realism to be diametrically opposed? Why can't a photo-realistic game be designed well too?

martindemon · Post by **martindemon** » Mon May 16, 2005 1:13 pm

They should use the dual core technology to do it! Far less expensive for those who want to compute advanced physics. You could select simple or advanced physics on the menu when you have a dual-core or dual pcu computer.

ChrisGrenard · Post by **ChrisGrenard** » Mon May 16, 2005 1:15 pm

Jeff V wrote:
noun wrote:I disagree - this new focus on the physics chipset, if it catches on, will just be another reason to move the focus away from game design and more towards photo realism.
Why do you consider game design and photo realism to be diametrically opposed? Why can't a photo-realistic game be designed well too?

I suspect that noun feels like I do on this (just assuming though).

My fear is that game design is spending far more of its focus time on realism than on innovation. If this physics chipset makes it *easier* to program physics into a game (thus taking less programming time) then I'm all for it. If it takes an increased amount of time, then I'm kinda against it.

noun · Post by **noun** » Mon May 16, 2005 1:28 pm

Jeff V wrote:
noun wrote:I disagree - this new focus on the physics chipset, if it catches on, will just be another reason to move the focus away from game design and more towards photo realism.
Why do you consider game design and photo realism to be diametrically opposed? Why can't a photo-realistic game be designed well too?

That's a great question. Why can't they? Why do so many games look great but suck to play? I'm not saying that it's impossible to do both, but in today's corporate environment, I'm concerned that it's far more likely that the crap will look even more beautiful, but stink worse. Lovely analogy, I know.

The Meal · Post by **The Meal** » Mon May 16, 2005 2:14 pm

noun wrote:
Jeff V wrote:
noun wrote:I disagree - this new focus on the physics chipset, if it catches on, will just be another reason to move the focus away from game design and more towards photo realism.
Why do you consider game design and photo realism to be diametrically opposed? Why can't a photo-realistic game be designed well too?
That's a great question. Why can't they? Why do so many games look great but suck to play? I'm not saying that it's impossible to do both, but in today's corporate environment, I'm concerned that it's far more likely that the crap will look even more beautiful, but stink worse. Lovely analogy, I know.

Yep. "We've got money for 60 people to make our game. Should we stick 45 of them into the art development, or into AI/scripting/enviornmental design?"

~Neal

Jeff V · Post by **Jeff V** » Mon May 16, 2005 7:54 pm

The Meal wrote:
noun wrote:
Jeff V wrote:
noun wrote:I disagree - this new focus on the physics chipset, if it catches on, will just be another reason to move the focus away from game design and more towards photo realism.
Why do you consider game design and photo realism to be diametrically opposed? Why can't a photo-realistic game be designed well too?
That's a great question. Why can't they? Why do so many games look great but suck to play? I'm not saying that it's impossible to do both, but in today's corporate environment, I'm concerned that it's far more likely that the crap will look even more beautiful, but stink worse. Lovely analogy, I know.
Yep. "We've got money for 60 people to make our game. Should we stick 45 of them into the art development, or into AI/scripting/enviornmental design?"

~Neal

That's because pretty graphics sells, tough AIs do not. This won't change until consumers stop flocking like a moth to flame to the best eye candy. Meanwhile, almost nobody refuses to buy a game solely based on a crappy AI. The converse is also true: a good AI doesn't sell a commensurate number of copies to justify development costs.

SkyLander · Post by **SkyLander** » Mon May 16, 2005 8:25 pm

What I see is that with the PPU you have set physics Engines that'll work with it. Basically if you are developing a game you build the 3d engine and then intigrate it into the Physics engine you bought to work with the chip. Like now it works with the NovodeX engine, and Ageai is trying to get rights to the Havoc engine. Now if your game supports the NovodeX or the Havoc engine it will work with the PPU. Thats what I gather from it atleast.

I think this is a step in the right direction although I don't really want to throw down 400 dollars for another card. I spend enough as it is buying Processors a vid card and every once in awhile a new sound card. Now if they integrate the PPU into the motherboard or pair it with the video card that would be different. Idealy you would put it with the vid card.

I think the PPU won't be essential in telling a story but it would make things look a lot cooler. Realistic clothes and hair, more realistic explosions. Being able to take out a house for the hell of it. It would just be more fun. See a sniper in a window and you have a rocket launcher? Just blow up the wall to get to the sniper.

The only problem I see is with multiplayer, how do you build a game for people that have the PPU and people that don't have the PPU? Take my sniper example, if you don't have the PPU you can't blow up the wall. Only thing I can think of is what they did in Jedi: Outcast. The single player game was vastly better than the multiplayer part. Single player was slower and more calculated with the lightsaber duals. But in Multiplayer is just became a mash fest, they totally dumbed down the blocking and you only had a set moves. The wo how did I do that move went away. It will most likely become the same thing in singleplayer you had an awesome moment that you want to see again in multiplayer but it becomes dumbed down to let everyone play.

Jeff V · Post by **Jeff V** » Mon May 16, 2005 9:25 pm

Like anything else, it will sink or swim based on the "killer apps" that accompany it out the door.

If there are some games that knock your socks off with the addition of this card, then it stands a good chance to survive. If the chicken comes before the egg, however, then the Shake-n-Bake is never gonna stick. A strategy game like the new RON won't be enough to create a demanding market: all of the major genres need to be hit (RPG, shooter, sim) to get the masses on board.

Has there been talk of any games besides the new RON supporting this?

DArtagnan · Post by **DArtagnan** » Tue May 17, 2005 7:41 am

I doubt that the introduction of this chip alone will cause a degradation in game design. Naturally, the corporate environment will want to sell games, rather than having them be respected or appreciated. That hasn’t changed and will probably not change for a good long while. However, there are those rare developers out there who care about their products, in ways other than commercial potential.

This chip will, I suspect, help augment our experience with those games. I don’t think we should hold back superior technology because of human nature, as long as it’s the entertainment industry we’re talking about. Besides, the force of evolution in any form is pretty irresistable and I consider it futile to stand against it. Physics is just another step towards realism, which is a factor the gaming industry will always be striving for.

Also, the physics systems in games today, though still quite limited, have helped immersion a very great deal in my opinion. Think of Silent Storm, or even the original X-Com. What would those games be without destructable terrain, or multiple height levels and what not. I think many gamers tend to forget what kinds of things make a difference, and only notice the lack of certain features, rather than the existence of them.

Gizah · Post by **Gizah** » Tue May 17, 2005 9:11 am

Think of Silent Storm, or even the original X-Com. What would those games be without destructable terrain, or multiple height levels and what not. I think many gamers tend to forget what kinds of things make a difference, and only notice the lack of certain features, rather than the existence of them.

Excellent point.

Peacedog · Post by **Peacedog** » Tue May 17, 2005 9:39 am

Excellent point.

And yet, X-com got by just fine without a physics chipset.

DArtagnan · Post by **DArtagnan** » Tue May 17, 2005 9:53 am

And yet, X-com got by just fine without a physics chipset.

Grand Theft Auto got by fine with a 2D top-down perspective with bland graphics. Somehow I suspect we would have a number of disappointed fans if Rockstar went back to that for their next GTA.

Wolfenstein 3D got by fine...

TES1: Arena got by fine...

Get it?

Peacedog · Post by **Peacedog** » Tue May 17, 2005 10:09 am

Get it?

Yes, though I'm not sure you understand what my point was (understandable given my lack of elaboration). Though to speak to your latest point, plenty of games got flashier graphics and didn't benefit. Silent Storm wasn't even close to better than X-com, nevermind whether or not the sci-fi angle worked. I don't think the number of crappy games coming out decreased with the advent of 3d. In fact, I think it went up at one point (though it's likely just flucutated with the overall number of games being made, proportionally).

The point about X-com was not about the "simpler days" nor about "more basci techonology being better" (also, I said nothing about the other games you mentioned nor will I; neither is a game I am anything other than indifferent too). My point was that something was pulled off without the physics chipset, and a wizbang 3d engine to boot. Doing something like that is *not* dependant on 3d nor on the physics chip. Taking it to the level where it does need those things brings other issues onto the table.

The physics chipset might be the bees knees. They said the same thing about 3d. Unfortunately, designers are having to spend more and more time working on their engines. Time spend on their engine is not time spent on the gameplay or AI or anyuthing else. Even when the engine has major gameplay ramifications (like a catapult simulator using the latest 3D game engine and some of that physics stuff).

Chris worries that designers won't focus enough on innovation and spend too much time on realism. I'd tweak his comment: they already spend too much time on the engine and not enough time on the game (never mind the innovation angle; not that I don't want to see mroe but hell I'd settle for some fun games at this point, and too many people are incapable of even that). How is this going to allow them to spend more time elsewhere? It won't, most likely. In fact, it could lead them to spend to more time with the things that matter the least.

Maybe it will, maybe it won't. It's hard to get exicted about this, however, given the current state of the industry. We do not need the ability to push more polygons. Or calculate more trajectories. We need the ability to create and market games for non-obscene costs and turn a profit on them.

edit: forgot to add the point in paragraph 2.

Samurai · Post by **Samurai** » Tue May 17, 2005 10:40 am

Not that I want to sound clueless about computers, but where exactly is this chip going to be placed? Are they planning on implementing this in new motherboards, put it in a PCI slot, duct taping it to the case... or what?

Peacedog · Post by **Peacedog** » Tue May 17, 2005 10:56 am

Not that I want to sound clueless about computers, but where exactly is this chip going to be placed? Are they planning on implementing this in new motherboards, put it in a PCI slot, duct taping it to the case... or what?

The link in this thread contains another article linked to at the bottom, about the chip. It mentions both add-in cards *and* chips that go on motherboards, interestingly enough.

Arcanis · Post by **Arcanis** » Tue May 17, 2005 12:09 pm

as was stated earlier in the post it will do well if some great games using it come out with or shortly after it does. It will undoubtedly create a whole genre of perfectly realistic beutiful crap for $90 a piece, that is life and unfortunately the game industry. We are greedy people we want beutiful games that are completely immersive and fun to play. what the industry has forgotten over the past couple of years is that games are ment to be fun first and entertain us. They instead focus on trying to sell us the most crap that they can.

ChrisGwinn · Post by **ChrisGwinn** » Tue May 17, 2005 12:51 pm

Unfortunately, designers are having to spend more and more time working on their engines. Time spend on their engine is not time spent on the gameplay or AI or anyuthing else.

I'm not convinced that this is actually true, for a couple of reasons. First, the pre-canned engine and middleware market has exploded in the last five or ten years. The number of developers that actually build a 3D rendering engine, do their own sound encoding, write their own accurate physics engine, etc. seems pretty low. Second, the job roles and skill sets are very different for gameplay vs/ engine vs/ AI. Programmer X usually isn't splitting time between the different types of work.

"So what?", you say, "replace 'time' with 'resources'" and the point still holds. And to a certain extent, you're right. But I think you're wrong in assuming that games skimp on gameplay/AI and don't skip on engine costs. It's just that you can buy a robust modern graphics engine and you can't buy a drop-in AI or gameplay engine. You can skimp all around, it's just that there's enough of a market for 3D middleware that it's cheaper and better to buy vs/ build.

I'm also old and cranky enough to remember that the eye candy vs gameplay debate was going on long before 3D was around.

Honestly, does anybody really think that, say, XCom wouldn't be better with a real 3D engine, or that they wouldn't have used one if it had been available? Line of site and multiple elevations were really hard to work with, and would be dramatically simplified in a 3D environment, even with nothing else changed.

Peacedog · Post by **Peacedog** » Tue May 17, 2005 1:40 pm

I'm not convinced that this is actually true, for a couple of reasons. First, the pre-canned engine and middleware market has exploded in the last five or ten years. The number of developers that actually build a 3D rendering engine, do their own sound encoding, write their own accurate physics engine, etc. seems pretty low. Second, the job roles and skill sets are very different for gameplay vs/ engine vs/ AI. Programmer X usually isn't splitting time between the different types of work.

That's true, however it would appear that there still needs to be quite a bit done once the package is purchased (well, that's how I understand it). Are these packages really benefitting development?

Game development times look they they are continuing to balloon to me, despite the fact that we have unprecedented ability to develop complicated software projects at out fingertips. I'm sure part of it is because game programming is still Assembly/C/C++ - languages that don't have some of the advantages of the modern OO languages in terms of ease of development (of course, they have their own upsides). Costs are rising. I'm not so sure that pre-made engines have helped out on average (I'd love to see some numbers though). Or, perhaps they can't help enough.

But I think you're wrong in assuming that games skimp on gameplay/AI and don't skip on engine costs. It's just that you can buy a robust modern graphics engine and you can't buy a drop-in AI or gameplay engine. You can skimp all around, it's just that there's enough of a market for 3D middleware that it's cheaper and better to buy vs/ build.

I don't think skimp is the word I'd chose, though I'm sure I've used it in the past (and I wish I had used resources in the original comment, and not time). I don't think it's as simple as "we have 100 utils, we'll spend 60 on glitz, and the rest can go to the other stuff".

I'm also old and cranky enough to remember that the eye candy vs gameplay debate was going on long before 3D was around.

Me too. However, this is more than an eye candy issue IMO - though it's probably most easily reflected in that manner.

Honestly, does anybody really think that, say, XCom wouldn't be better with a real 3D engine, or that they wouldn't have used one if it had been available? Line of site and multiple elevations were really hard to work with, and would be dramatically simplified in a 3D environment, even with nothing else changed.

Silent Storm was a decent game I thought; not great. My computer certainly played a role in that. It started pretty good, but long before the Sci-Fi elements kicked in (honestly, I didn't have much of a problem with them) I was getting board with some of the processing times that surrounded missed shots and the like. I've gotten the "calculating path" message for up to 30 seconds at times. It's ridiculous. It wasn't enough to stop me from playing at first, and it wasn't the sole reason I quit playing by any stretch.

Yes, rending that door to splinters and then busting through holds a certain thrill. But the price I paid for it was alot of "ho hum" time, sitting there waiting for my poor beleaguered P4 and Geforce 3 to figure out what the hell had happened. I don't have the best system in the world, by a *long* shot, but neither is it wimpy (P4 1.8a; 512 megs of pc2100 RAM at the time, Geforce 3 ti 200).

Was the ability to render a door to shreds in itty bitty pieces, where each door would get holes in different places as opposed to the "broken door graphic" really worth it? I think it's debateable. Individual mileages will vary, of course, but what specifically was I provided with that I wasn't provided with in X-com? I'm not saying that a new X-com couldn't be 3d either (I don't think 3d is evil or anything)l, but I think it is telling that we haven't seen a game outdo X-Com (perhaps the canned fourth installment would have, too bad we'll never know).

Back on the chip, I'm going to be interested in hearing what the "on board" solution for this chip is going to be.

Octopus Overlords

Are you ready for Physics Chipsets??

Is Ageia's PhysX Chip A Great Idea?

Are you ready for Physics Chipsets??