Ageia's PPU's

[lexmark]

Not that I'm in a big rush to drop $200-$300 on more hardware (im not), but wasn't Ageia pushing for a ppu realese around x-mas? haven't seen any new press releases lately.

 

[BlacKJesuS]

They Stole It

 

[NoStateofMind]

I don't know but I found this is where you can see what it can do

 

[Wentelteefje]

Didn't I hear someone (was it Id?) claiming that those PPU's would be a waste, because they could implement the same into a graphics engine...? Not completely sure about it, but I recall having read this somewhere...

 

[hans030390]

Originally posted by: Wentelteefje
Didn't I hear someone (was it Id?) claiming that those PPU's would be a waste, because they could implement the same into a graphics engine...? Not completely sure about it, but I recall having read this somewhere...

They're doing something like that...I think. Still, a PPU would help a good bit with physic intensive games...

But, I wonder if it will only work if the games are programmed to take advantage of it...if thats the case, then it would only be useful to select upcoming games that use it.

 

[SonicIce]

so it got shot down? too bad, it could have been cool.

 

[Rage187]

Glad it died.

 

[Parkre]

I don't think it's dead.

They are just waiting for the right time... Unreal 3 has the technology, which may be what theyare waiting for.

I don't think they would waste their grand opening on a "Bet on Soldier"

I still think they are going to come out...

 

[NoStateofMind]

If you check out that link I posted and watched the Stoked Rider 3D Screenshot video, you would see why we DO want this to come out. I was impressed with the amount of workload it takes from the CPU.

 

[BlingBlingArsch]

Problem remains there are no games that can use the PPU right now, from what ive read it seems the Ageias card needs software implementation. Late rumours from CES say it will come within first half of 2006. Time will tell what games do benefit..

 

[Subwayeatbig]

the company Havock figured out a way to do it with the video card. Its the poors man way to get more physics processing.

 

[HamburgerBoy]

Originally posted by: Rage187
Glad it died.

Same here. Gaming computers are already expensive enough.

 

[BlingBlingArsch]

Originally posted by: Subwayeatbig
the company Havock figured out a way to do it with the video card. Its the poors man way to get more physics processing.

well they claim they have. i wouldnt be surprised if this was some kind of jumping on a train.

 

[apoppin]

Originally posted by: lexmark
Not that I'm in a big rush to drop $200-$300 on more hardware (im not), but wasn't Ageia pushing for a ppu realese around x-mas? haven't seen any new press releases lately.

AGEIA PhysX card demo pictured

the first question that we asked AGEIA was when the actual hardware will hit the shelves. We've been talking about this product since the Game Developers Conference/Cebit 2005 and we still haven?t seen it for sale.

We don?t have a date for you but it's now rescheduled for Q1 2006. The company high ranking executive explained to us that it is actually waiting for the content. Some content such as the demo we saw last night at AGEIA's booth is already available but those guys are waiting for some studios to finish their games.

Nvidia's well known graphic partner BFG will be exclusive for North America and those guys are very excited about the cards.

It's interesting that many of BFG guys are actually ex 3DFX and therefore are living through the birth of 3D graphic cards all over again. You first need hardware, and it's there now, and once you get it you need to convince developers and publishers that this is the way to go.

The demo had hundreds of barrels, boxes and other pieces that you could shoot with a great cannon and create some great physics experiences. We saw the demo of the PPU card with two Nvidia's cards in SLI versus the same machine with same SLI cards without Physics card. The guys ran the demo on both machines simultaneously and the machine without Physics Processing Unit was rendering only a few frames per second. The one with Physics card inside was rendering the same scene at more than 30 FPS constantly. It's 6 FPS without versus 30 with, not bad at all. It just looks great and we even have a picture that might show you a little bit what it's like.

 

[KeithP]

As I remember it, the idea put forth to use the GPU for physics calculations basically relied on the GPU having some excess capacity. Of course, the problem with that is any game that would really benefit from having a physics processor is probably not going to leave a lot of unused processing power on your GPU. The one exception might be SLI and Crossfire setups.

Is AGEIA's solution an open standard? Can anyone make hardware that will support it? It sounds to me like what they have is essentially like 3DFX's Glide except for physics. What we need for broad software support is a standard API so anyone can make hardware to accelerate physics. Something like OpenGL except it would be for physics calculations. I have a feeling AGEIA's product will end up like Glide. Some initial acceptance by software developers doom to fade away in the long run as better solutions emerge.

-KeithP

 

[sandorski]

Originally posted by: Parkre
I don't think it's dead.

They are just waiting for the right time... Unreal 3 has the technology, which may be what theyare waiting for.

I don't think they would waste their grand opening on a "Bet on Soldier"

I still think they are going to come out...

Wise choice. Bet On Soldier is a champ in the world of Suckage!

 

[Fox5]

Originally posted by: KeithP
As I remember it, the idea put forth to use the GPU for physics calculations basically relied on the GPU having some excess capacity. Of course, the problem with that is any game that would really benefit from having a physics processor is probably not going to leave a lot of unused processing power on your GPU. The one exception might be SLI and Crossfire setups.

Is AGEIA's solution an open standard? Can anyone make hardware that will support it? It sounds to me like what they have is essentially like 3DFX's Glide except for physics. What we need for broad software support is a standard API so anyone can make hardware to accelerate physics. Something like OpenGL except it would be for physics calculations. I have a feeling AGEIA's product will end up like Glide. Some initial acceptance by software developers doom to fade away in the long run as better solutions emerge.

-KeithP

I'd say Ageia is more like Aureal.

Anyhow, the graphics card idea sounds better. I'd rather spend $300 to $400 on a 2nd video card than $300 to $400 on a physics processor.....though if the physics processor is much more capable (from what I've heard about it, it's not) then it would be better.

 

[LungExpansion]

The PPU will make a great addition and it does what the GPU's currently werent designed to do. A GPU can render a scene but its not the ideal component for determining the movement, collision detection and other various items which are physics based.

Dont confuse physics calculations in real time to rendering a scene that contains minimal physics. If the GPU had to determine the flow of a projectile while doing collison detection and rendering a GPU would be crushed. It cant do both without a major sacrafice in performance and adding a second GPU wont solve that problem. Rendering a frame of points in time is not physics.

Cpu's are poor at Physics plain and simple. There is a demo where they put the PPU against a dual core top of the line Intel cpu and the PPU makes a joke out of the dual core and it does it with only 28watts of power.

The PPU will make games a lot more interesting and free up the CPU for game AI and user AI. If the GPU was so great at physics we would be using it rather than having the CPU calculate all the physics currently.

 

[Fox5]

Originally posted by: LungExpansion
The PPU will make a great addition and it does what the GPU's currently werent designed to do. A GPU can render a scene but its not the ideal component for determining the movement, collision detection and other various items which are physics based.

Dont confuse physics calculations in real time to rendering a scene that contains minimal physics. If the GPU had to determine the flow of a projectile while doing collison detection and rendering a GPU would be crushed. It cant do both without a major sacrafice in performance and adding a second GPU wont solve that problem. Rendering a frame of points in time is not physics.

Cpu's are poor at Physics plain and simple. There is a demo where they put the PPU against a dual core top of the line Intel cpu and the PPU makes a joke out of the dual core and it does it with only 28watts of power.

The PPU will make games a lot more interesting and free up the CPU for game AI and user AI. If the GPU was so great at physics we would be using it rather than having the CPU calculate all the physics currently.

Do we really need a seperate processor though? CPUs are excellent at collision detection, and GPUs are flops monsters. PCI-Express provides a fast bi-directional bus for the cpu and the gpu to communicate, why can't the cpu handle the collision detection and whatever the gpu is poor at, while the gpu handles the heavy calculations, such as animation, trajectory, deformation?

 

[Mocib]

Here's the latest article about AGEIA and the PhysX PPU: http://www.gdhardware.com/hardware/ppu/ageia/dec04/001.htm

The PPU is not dead. It was intented to be released in December but the date has been pushed to Q1 2006 because the games are not ready yet.

 

[Munky]

Of course it needs developers to explicitly code for it. Otherwise how would you know what code to unload to the ppu and what not to unload. It will probably need it's own API, or possibly become incorporated into future versions of DX if it gains popularity. The problem is there's not much demand for the thing, and rightfully so. It's a game we're talking about, not some mathematical modeling of a chain reaction. You can just as easily unload a lot of the calculations on the gpu because in current games the vertex processors are just sitting twiddling their thumbs half the time, waiting for the pixel processors to finish rendering the frame. Add multi-core cpu's to the equation, and I hardly see a reason to spend $200+ on additional physics HW.

 

[John Reynolds]

I've said this before, but what AGEIA needs at the time of hardware launch is a AAA title that really shows an appreciable benefit from supporting the PPU, something that goes way beyond increased frame rate from the physics code being hardware rather than software accelerated. But what that title could be is a mystery to me: UT2007? Warhammer Online? It's gotta be a high profile title and there needs to be a really well thought out marketing push behind it.

 

[nitromullet]

I think that Ageia needs to figure out a way to make the PPU useful. Obvioulsy, having games that take advantage of it will help, but what about something actually useful when you're not gaming so it doesn't just sit there most of the time? Maybe something in the driver that makes it grab your AV process or other background tasks to offload them from your CPU. Of course, with multi-core CPU's this may not be necessary, but it never hurts to have extra power at your disposal. IMO, if the PPU fails it will be because it is a one-trick pony that is useful only for entertainment. Look at every other component in your PC... They all can do more than one function, have many uses, or they are in constant use. Plus, a card that does stuff that can easily explained to the wife doesn't hurt either. It will be a hard sell for most guys of all they have is, "yeah, I know it's $300, but look at how much better it makes the barrels fly around the hanger in this demo."

Another option is that the chips come integrated into the motherboard. ...maybe a losing propositon for Ageia initially, but possibly the way to get them into consumers' machines to get them exposed and hooked to the PPU.

 

[John Reynolds]

Yea, but with that perspective how would you explain the success of the Voodoo 1 back in '96? It too was essentially a one-trick pony that was useful only for entertainment purposes (games).

I agree, though, that AGEIA has a tough hill to climb.

 

[the Chase]

Wow great posts on this(especially last four)- not just to say good its dead(as in many other threads on this)- I'm as cheap as they come but this IS innovation. The problem I see(and what Ageia will have to try to conquer) is that most everyone(including me) have no idea what playing a game with all the new physics would be like. You can't make a value judgement on whether a new piece of hardware is worth X amount of $ or not until you experience it or your peers say- o mee god this thing will blow you away like I've never had an experience so intense!! I DO beleive though that this card will provide physics on a level that no other piece of hardware can match for at least a couple of years and probably more. But will it be worth $200-$300 or more for those physics? It takes an awful lot to pry my creaky old wallet up....