AGEIA PhysX Demonstration Video- WOW!

[Operandi]

I don't see any of this happening at all, the whole idea is fundamentally flawed.

In order for any physics processor to be of any use it will have to accounted for in game development. In other words the developer will have to assume the user is going to have one in their system in order to make any of these high physics content scenes practical. The problem if its not there (and they won't be) the game will be reduced to a slide show.

The bottom line is there is no easy way to implement this, its not like a new high end GPU where you can crank up the AA and AF and see a benefit.

 

[Dman877]

Correct me if I'm wrong but to show all that damage, game devs are going to have to render every nut and bolt in the airplane (for example). That's simply not going to happen.

 

[CraKaJaX]

holy sh*t.

 

[ribbon13]

Originally posted by: Googer

Originally posted by: Soccerman06
They are making different price cards. Aegis rep said on G4 that there will be $100-$300 versions for people to buy, not just one expensive card.

Thats just great.

I was hoping they would produce just one version of this card and avoid and save us all a ton of headache and confusion. Now that they are taking the Creative Labs Approach to selling, I say: "Die in a Fire Ageia"!

The moment Creative releases a PCIe card with Dolby Digital Live and doesn't resample behind the users back, I'll recant my negativity towards them. But not until then.

 

[Jeff7181]

I didn't think that was very impressive at all... was it supposed to be simulating all that under water?

 

[MobiusPizza]

Originally posted by: Dman877
Correct me if I'm wrong but to show all that damage, game devs are going to have to render every nut and bolt in the airplane (for example). That's simply not going to happen.

I think you are over compliticating it a bit.
Nuts and bolts are not necessarily needed to be emulated for realistic damage effect, although it makes it even more realstic. It's simply too much work
I don't think the pieces in the demo falling off had bolts rendered.

People are bashing how unrealistic it is saying that the metal pieces won't fall off like that with a hand gun, and saying that it's got no recoil is bullsh1t. The purpose of this demo is to show how things can be done and it doesn't mean in future game developers will all make a game with no hand gun recoil and a plane that can be damaged with a handgun. It is just there to show what can be done putting emphasis on the spot light not the gun or whether the object is a plane or not.

I do agree that processor is very overpriced. But what can you expect when new stuff come out.
If you are a company which paid billions on product research would charge premium on a monopoly product. Anyone who did basic business study would know already

Otherwise it might take them 5 years to recoup the expense in R&D

 

[drsafety]

I would pay maybe $100 for that card when most new games have support for it.

 

[Googer]

Originally posted by: ribbon13

Originally posted by: Googer

Originally posted by: Soccerman06
They are making different price cards. Aegis rep said on G4 that there will be $100-$300 versions for people to buy, not just one expensive card.

Thats just great.

I was hoping they would produce just one version of this card and avoid and save us all a ton of headache and confusion. Now that they are taking the Creative Labs Approach to selling, I say: "Die in a Fire Ageia"!

The moment Creative releases a PCIe card with Dolby Digital Live and doesn't resample behind the users back, I'll recant my negativity towards them. But not until then.

What I meant was they are going to be selling one card in six flavors and charge eight differant prices. That really annoys me because the most expensive card costs only $2 more to make and yet they charge $200 more for it. RIP OFF!:evil: Just make one version and charge the same amout so that most people can afford one. Profit through volume.

 

[razor2025]

The comments about dev team having nightmare about their "target rig" is quite valid. I just don't see how the developers can anticipate how many "gamers" have the PPU card. In graphics area, it's very easy to scale down the details to accomidate the user hardware. Slower GPU? No problem, you get crappier texture (no filter) and tons of jaggies @ crappy res (800x600). Super GPU? Jack up those AA and AF along with uber-res! Physics, on the other hand, affects not only in graphics, but also game mechanics. Unless the developer can somehow magically create similar gameplay from reduced amount of physics calculations required at that particular level, the gameplay will definitely be different. Got your PPU? Alright, you get thousands of physically accurate objects to interact with. No PPU? Too bad, you can have thousands of interactive objects @ super slow speed or have only hunreds of objects instead. As you can see, the developer team has a nightmarish dilemma. And unlike writing multiple graphics rendering path (like how Half-Life 2 was rumored to be), writing games with huge difference in amount of "physical" realism is insanely difficult, if not impossible, since every aspect of gameplay will be affected.

Thus, it comes down to software. If Aegia can get some developers to create KILLER games that takes advantage of PPU (Killer = Game of Year caliber) AND gurantee them some type of financial benefit (since the game will probably have MUCH smaller audience at launch of PPU and game), they might be able to pull a success.

I think Aegia will find warm welcome from hardcore sim fans (those that spends $300+ on joystick setups alone) or some type of military/commercial simulator. For average consumer-level games, I just don't see how they can manage to convince the developers to write games for their PPU.

 

[razor2025]

As for comments about using PPU for science/engineering. This level of hardware is too low power for such use. Computational physics require super-computer level power, which by raw power from the super-comp, will blow the PPU out of the water. This goes same for engineering also. Remember, the PPU is meant as a consumer level product. It's not meant to be replacing super-comps doing physics # crunch. If Aegia had made the PPU useful for professional/research sector, they would be charging the card in tens of thousands of dollars. $299 would be super steal bargain.

 

[RaiderJ]

I almost imagine physics in games would be best when it goes unnoticed - kinda like really good music & sound effects. Keeps the realism and fantasy going - like a good book!

Cool video!

 

[Jeff7181]

Originally posted by: razor2025
The comments about dev team having nightmare about their "target rig" is quite valid. I just don't see how the developers can anticipate how many "gamers" have the PPU card. In graphics area, it's very easy to scale down the details to accomidate the user hardware. Slower GPU? No problem, you get crappier texture (no filter) and tons of jaggies @ crappy res (800x600). Super GPU? Jack up those AA and AF along with uber-res! Physics, on the other hand, affects not only in graphics, but also game mechanics. Unless the developer can somehow magically create similar gameplay from reduced amount of physics calculations required at that particular level, the gameplay will definitely be different. Got your PPU? Alright, you get thousands of physically accurate objects to interact with. No PPU? Too bad, you can have thousands of interactive objects @ super slow speed or have only hunreds of objects instead. As you can see, the developer team has a nightmarish dilemma. And unlike writing multiple graphics rendering path (like how Half-Life 2 was rumored to be), writing games with huge difference in amount of "physical" realism is insanely difficult, if not impossible, since every aspect of gameplay will be affected.

Thus, it comes down to software. If Aegia can get some developers to create KILLER games that takes advantage of PPU (Killer = Game of Year caliber) AND gurantee them some type of financial benefit (since the game will probably have MUCH smaller audience at launch of PPU and game), they might be able to pull a success.

I think Aegia will find warm welcome from hardcore sim fans (those that spends $300+ on joystick setups alone) or some type of military/commercial simulator. For average consumer-level games, I just don't see how they can manage to convince the developers to write games for their PPU.

I don't think it's that complicated. With a PPU you get more particles... that's it. A game is not going to be designed around a brick wall made of thousands of individual bricks. There may be a brick wall in the game made of thousands of individual bricks that you're able to destroy if you have a PPU... or there may be a regular flat model with a brick texture and bump mapping.

AFAIK, a PPU doesn't increase the accuracy of the physics or effect the physics in any way. It just allows physics for more objects to be calculated. So another example... if you have a PPU you might see more objects that you can manipule... instead of a pile of 6 barrels in Far Cry that you can knock over and roll around, it might be a pile of 20. High quality smoke effects (fluid dynamics) are another advantage to having a PPU. Without one you'll just have the same sorta real looking smoke you see in HL2. Maybe there's a book case you can knock over and each book is rendered individually with a PPU... without a PPU it might just be book case that you knock over, and a few books "spawn" when it falls.

 

[modedepe]

Originally posted by: AnnihilatorX
I do agree that processor is very overpriced. But what can you expect when new stuff come out.
If you are a company which paid billions on product research would charge premium on a monopoly product. Anyone who did basic business study would know already

Otherwise it might take them 5 years to recoup the expense in R&D

True, but they are also trying to create a market where one doesn't exist currently. Getting people to adopt early is going to take some doing.

Originally posted by: Jeff7181

Originally posted by: razor2025
The comments about dev team having nightmare about their "target rig" is quite valid. I just don't see how the developers can anticipate how many "gamers" have the PPU card. In graphics area, it's very easy to scale down the details to accomidate the user hardware. Slower GPU? No problem, you get crappier texture (no filter) and tons of jaggies @ crappy res (800x600). Super GPU? Jack up those AA and AF along with uber-res! Physics, on the other hand, affects not only in graphics, but also game mechanics. Unless the developer can somehow magically create similar gameplay from reduced amount of physics calculations required at that particular level, the gameplay will definitely be different. Got your PPU? Alright, you get thousands of physically accurate objects to interact with. No PPU? Too bad, you can have thousands of interactive objects @ super slow speed or have only hunreds of objects instead. As you can see, the developer team has a nightmarish dilemma. And unlike writing multiple graphics rendering path (like how Half-Life 2 was rumored to be), writing games with huge difference in amount of "physical" realism is insanely difficult, if not impossible, since every aspect of gameplay will be affected.

Thus, it comes down to software. If Aegia can get some developers to create KILLER games that takes advantage of PPU (Killer = Game of Year caliber) AND gurantee them some type of financial benefit (since the game will probably have MUCH smaller audience at launch of PPU and game), they might be able to pull a success.

I think Aegia will find warm welcome from hardcore sim fans (those that spends $300+ on joystick setups alone) or some type of military/commercial simulator. For average consumer-level games, I just don't see how they can manage to convince the developers to write games for their PPU.

I don't think it's that complicated. With a PPU you get more particles... that's it. A game is not going to be designed around a brick wall made of thousands of individual bricks. There may be a brick wall in the game made of thousands of individual bricks that you're able to destroy if you have a PPU... or there may be a regular flat model with a brick texture and bump mapping.

AFAIK, a PPU doesn't increase the accuracy of the physics or effect the physics in any way. It just allows physics for more objects to be calculated. So another example... if you have a PPU you might see more objects that you can manipule... instead of a pile of 6 barrels in Far Cry that you can knock over and roll around, it might be a pile of 20. High quality smoke effects (fluid dynamics) are another advantage to having a PPU. Without one you'll just have the same sorta real looking smoke you see in HL2. Maybe there's a book case you can knock over and each book is rendered individually with a PPU... without a PPU it might just be book case that you knock over, and a few books "spawn" when it falls.

That was my thinking also, that the amount of objects for which physics would be calculated could be adjusted. I think it would definitely be manageable.

 

[ribbon13]

Originally posted by: modedepe
That was my thinking also, that the amount of objects for which physics would be calculated could be adjusted. I think it would definitely be manageable.

I.e. it will be really simple for the game engine to use the regular splat w/o PPU and w/PPU you get a beautiful intricate marvel of bloodspray with a direct hit from the rocket launcher?

 

[Elcs]

Originally posted by: Jeff7181

Originally posted by: razor2025
The comments about dev team having nightmare about their "target rig" is quite valid. I just don't see how the developers can anticipate how many "gamers" have the PPU card. In graphics area, it's very easy to scale down the details to accomidate the user hardware. Slower GPU? No problem, you get crappier texture (no filter) and tons of jaggies @ crappy res (800x600). Super GPU? Jack up those AA and AF along with uber-res! Physics, on the other hand, affects not only in graphics, but also game mechanics. Unless the developer can somehow magically create similar gameplay from reduced amount of physics calculations required at that particular level, the gameplay will definitely be different. Got your PPU? Alright, you get thousands of physically accurate objects to interact with. No PPU? Too bad, you can have thousands of interactive objects @ super slow speed or have only hunreds of objects instead. As you can see, the developer team has a nightmarish dilemma. And unlike writing multiple graphics rendering path (like how Half-Life 2 was rumored to be), writing games with huge difference in amount of "physical" realism is insanely difficult, if not impossible, since every aspect of gameplay will be affected.

Thus, it comes down to software. If Aegia can get some developers to create KILLER games that takes advantage of PPU (Killer = Game of Year caliber) AND gurantee them some type of financial benefit (since the game will probably have MUCH smaller audience at launch of PPU and game), they might be able to pull a success.

I think Aegia will find warm welcome from hardcore sim fans (those that spends $300+ on joystick setups alone) or some type of military/commercial simulator. For average consumer-level games, I just don't see how they can manage to convince the developers to write games for their PPU.

I don't think it's that complicated. With a PPU you get more particles... that's it. A game is not going to be designed around a brick wall made of thousands of individual bricks. There may be a brick wall in the game made of thousands of individual bricks that you're able to destroy if you have a PPU... or there may be a regular flat model with a brick texture and bump mapping.

AFAIK, a PPU doesn't increase the accuracy of the physics or effect the physics in any way. It just allows physics for more objects to be calculated. So another example... if you have a PPU you might see more objects that you can manipule... instead of a pile of 6 barrels in Far Cry that you can knock over and roll around, it might be a pile of 20. High quality smoke effects (fluid dynamics) are another advantage to having a PPU. Without one you'll just have the same sorta real looking smoke you see in HL2. Maybe there's a book case you can knock over and each book is rendered individually with a PPU... without a PPU it might just be book case that you knock over, and a few books "spawn" when it falls.

I thought one of the major points of the PPU was to reduce the number of scripted effects and turn them into real time.

Rather than walking past a certain point and having some predetermined explosion happen, I thought the PPU allowed Game Devs to do the exact same type of thing (massive explosions, heavy physics laden stuff) on the fly so to speak.

You walk along, a wall blows up and a previously unmovable barrel flys up in the air and lands. Scripted event.

If by some random occurance, 2 enemies walks past this scripted event, they may be killed and fall over since they werent coded into the script or expected by the game to walk past that area at that time.

But if that same event had happened with a PPU, it could be more realistic as their physics could be calculated as it happens so they could fly up, up and away and land with a couple of bounces.

I thought that was one of the biggest ideas for a PPU. Like the Water videos show, this could be great for fog and water spray (broken pipes, water splashes as you walk through water, rain splashing against a car windscreen with varying droplet size/angle of approach encorporating wind effects... would be great for atmosphere in a racing game)

 

[Jeff7181]

Originally posted by: Elcs

Originally posted by: Jeff7181

Originally posted by: razor2025
The comments about dev team having nightmare about their "target rig" is quite valid. I just don't see how the developers can anticipate how many "gamers" have the PPU card. In graphics area, it's very easy to scale down the details to accomidate the user hardware. Slower GPU? No problem, you get crappier texture (no filter) and tons of jaggies @ crappy res (800x600). Super GPU? Jack up those AA and AF along with uber-res! Physics, on the other hand, affects not only in graphics, but also game mechanics. Unless the developer can somehow magically create similar gameplay from reduced amount of physics calculations required at that particular level, the gameplay will definitely be different. Got your PPU? Alright, you get thousands of physically accurate objects to interact with. No PPU? Too bad, you can have thousands of interactive objects @ super slow speed or have only hunreds of objects instead. As you can see, the developer team has a nightmarish dilemma. And unlike writing multiple graphics rendering path (like how Half-Life 2 was rumored to be), writing games with huge difference in amount of "physical" realism is insanely difficult, if not impossible, since every aspect of gameplay will be affected.

Thus, it comes down to software. If Aegia can get some developers to create KILLER games that takes advantage of PPU (Killer = Game of Year caliber) AND gurantee them some type of financial benefit (since the game will probably have MUCH smaller audience at launch of PPU and game), they might be able to pull a success.

I think Aegia will find warm welcome from hardcore sim fans (those that spends $300+ on joystick setups alone) or some type of military/commercial simulator. For average consumer-level games, I just don't see how they can manage to convince the developers to write games for their PPU.

I don't think it's that complicated. With a PPU you get more particles... that's it. A game is not going to be designed around a brick wall made of thousands of individual bricks. There may be a brick wall in the game made of thousands of individual bricks that you're able to destroy if you have a PPU... or there may be a regular flat model with a brick texture and bump mapping.

AFAIK, a PPU doesn't increase the accuracy of the physics or effect the physics in any way. It just allows physics for more objects to be calculated. So another example... if you have a PPU you might see more objects that you can manipule... instead of a pile of 6 barrels in Far Cry that you can knock over and roll around, it might be a pile of 20. High quality smoke effects (fluid dynamics) are another advantage to having a PPU. Without one you'll just have the same sorta real looking smoke you see in HL2. Maybe there's a book case you can knock over and each book is rendered individually with a PPU... without a PPU it might just be book case that you knock over, and a few books "spawn" when it falls.

I thought one of the major points of the PPU was to reduce the number of scripted effects and turn them into real time.

Rather than walking past a certain point and having some predetermined explosion happen, I thought the PPU allowed Game Devs to do the exact same type of thing (massive explosions, heavy physics laden stuff) on the fly so to speak.

You walk along, a wall blows up and a previously unmovable barrel flys up in the air and lands. Scripted event.

If by some random occurance, 2 enemies walks past this scripted event, they may be killed and fall over since they werent coded into the script or expected by the game to walk past that area at that time.

But if that same event had happened with a PPU, it could be more realistic as their physics could be calculated as it happens so they could fly up, up and away and land with a couple of bounces.

I thought that was one of the biggest ideas for a PPU. Like the Water videos show, this could be great for fog and water spray (broken pipes, water splashes as you walk through water, rain splashing against a car windscreen with varying droplet size/angle of approach encorporating wind effects... would be great for atmosphere in a racing game)

Right... but a PPU doesn't allow better physics... it allows MORE physics. I think you're saying the same thing I am, but thinking of the application, not the technology behind it. All that stuff you said will all be possible because a PPU can do more physics calculations than a CPU can, not because it's more accurate or something like that. All that is possible right now, but it would be too slow because of the extra physics calculations required for the physics on all the extra objects.

*EDIT* A PPU isn't doing something a CPU can't... it's doing more things faster than a CPU can. The birth of the PPU is essentially the same as the birth of the GPU when you think about it. It's specialized hardware created to do a specialized task faster than a CPU can do it, which allows more work to be done in the same amount of time.

 

[Mocib]

I like these two newer demo videos much better:

Video 2 - Metal bending
Video 3 - Fluid simulation

http://www.airtightgames.com/currentproject.html

Those are a few months old videos though... I really hope that AGEIA would release more demos soon.

 

[Elcs]

I agree Jeff7181. I think this is a step forward.

Fluid demo is good but none of the crates are damager or toppelled which may happen in real life.

Metal bending is awesome but the crane looks a little bit like jelly when its hit.

Large Scale Destruction is just the Plane crashing into the cylinder.

 

[Thor86]

As a separate card, no way it's going to happen. As a co-processor for cpus or gpus, maybe.

 

[lexmark]

the first time I watch it, the first half was boring and unrealistic to me. The aviation crash towards the end was impressive though. Try this:

FF to the the airplane crash scene, and slow down the video some more. You can get a better feeling of how each of box reacts to the plane as its moving into it. I missed alot of the movement the first couple of times I watched it, but I got a better idea of the physics in super slomo. just try it!

still doesnt warrant a price tag over $200 though

 

[Googer]

Originally posted by: The Pentium Guy

Originally posted by: Unkno
if nobody buys it, it would soon end up to nearly the same price for a sound card

Let's start the ATOT effect. Nobody buy this, it'll benifit us all....

.

I second that!

 

[Googer]

Originally posted by: C6FT7
LOL that is very poor accuracy.

The gravity is way off, bombs hitting the railing (at the end) would definitely cause damage and not be deflected.

Don't even get me started on the fact that the handgun was fired many times yet there was no recoil, the hammer didn't move and it never needed a clip change!

Fakopolis!

Changing clips has nothing to do with physics. Ageia says that the Physics Enviroment is completey cusomizeable. So it would be possible to make the bombs hit quicker and stronger gravitiy could be change by altering a few numerical values; the hammer is just an anamation deficit.

 

[lexmark]

why didn't they make the demo with up-to-date graphics?

 

[Googer]

WORD TO AGEIA: I hope you are listening to everything we are saying in this thread. If you choose to ignore us, then you shall pay the price.

 

[lexmark]

what effect will a PPU have on games that came before it? better framerates?