Thanks to a forum member, the Cell Factor demo can runs either with or without the formerly \'required\' PhysX accelerator card. Our tests reveal astonishing performance differences.

Speak out in the Toms's Hardware reader survey!

had bought one from bfg the day they hit the market and was so disappointed with the performance. I think the effects had really slowed down the game play, good idea but this product was not ready for the market, never ever buy something from white paper.

Anyone else notice that the links for video 1 and video 3 seem switched in the article?

yes

Videos 1 and 3 are switched and I think 4 and 5 are as well.

good article and interesting. i have to say that ageia obviously doesn't have any respect for its users or market. the fact that they didn't anticipate someone would develop a workaround shows they think we are idiots.

also i am convinced that the new demo made sure the game would run badly without the card and not due to the system not being able to do it alone.

ageia are on a slippery slope. how long until someone manages to bypass they blocking of cloth and liquid simulation and then they will be shown as the fraudsters they are.

Based on personal experience using the Ageia physX libraries (code wise). I think Ageia actually has a very strong foundation in the industry. Their code is freely available (in my case, used in a student project), and is INCREDIBLY easy to get up and running in no time. Which is an advantage over a few other phsyics libraries that I had experimented with (Newton, ODE).

Albeit, when I was working with the ageia libraries, they still have yet to get triange collision working properly (if they are at all going to attempt to get it working). Which right now is a _major_ disadvantage of using their libraries as opposed to some other ones.

The major advantage, of course, is that you can offload the physX code over to the physics card itself.

Honestly, they have a very good software product going right now, so at the least, I think they could stay afloat by merely licensing out their physics library. (We ran all our code without their card ha ha, but we didn't use anything too advanced besides some height map collision and some of their own special collision body detections).

Honestly, I don't think I'd invest in a physics card that cost more than $50. But meh, I'm also a broke college student

And really, these guys are far from stupid. They're just desperately trying to get everyone onboard with offloaded physics and intense environments. And seriously, if a friggin dual core got bogged down on a cloth simulation (I hate to tell you, but that stuff is increidbly complex computationally, go try a cloth simulation in maya and see how long that takes to render in NON-REALTIME). You have _VERY_ little hope of getting your physics code to run off just the second core of a processor (which I thought was possible initially until seeing this benchmark).

Now, when we get up to quad cores, that will be something entirely different. Somehow I doubt you could destroy 2 whole cores on properly written (read: as inacurate as possible to allow proper real-time computation) physics code.

I guess the future holds that prospect in sight.

Also, physics code is most definately here to stay, so everyone get pumped for some serious stuff in future gaming! (which is what it's all about isn't it?)

-T

Hmmm, I notice that this is one again in the Memory section of the forum as wello as the Conroe and nVidia article. Any particular reason for that?

On topic, I look forward to when this stuff gets offloaded to CPU/GPU because I don't really think there will widespread adoption of yet another card that's $300. So far I'd say that's pretty accurate.

That cloth tearing video was pretty cool, I'll admit. As mentioned above, once quad cores come out and dual cores are completely main-stream (they're getting pretty close already I think) then offloading the physics to 1 or 2 CPU cores would be the best thing.

I don't claim to know how any of this stuff really works but there has to be a better solution than Ageia's. Or the price needs to drop. I could see gamers forking out like max $150 for this and only once it's well esablished.

I don't want them to succeed because that will mean my gaming will get even more expensive....and I already think the GPU card pricing is robbery!!!

nVidia/ATI should simply incorporate a chip on their cards (IF NECESSARY) for physics. But as their cards get faster and faster and I see these cards coming standard with dual GPUs (SLI on card) on them anyway...another chip may not be necessary.

Add in cards suck, are speed limiting, require more drivers, take up space, make more heat etc. Do it on the graphics card!

I think I speak for a lot of people when I say I haven't really found a need for a product like this.

I think I speak for a lot of people when I say I hope a lot more games come out supporting this card SOON, so we actually have some good information to base a decision on. Right now all we have is a bit of code slapped into GRAW at the last minute that needed patching due to initial low FPS, and the Cell Factor tech demo. And now this Cell Factor deconstruction experiment which may or may not be a valid test. I wish all game developers would figure out how Crytek is doing their amazing creations. And I wish ASUS would just buy them and mount the PhysX chip on every motherboard.

Until then, I'll hope they survive because we are in great need of more advanced physics in games. Besides, the ATI and nvidia solutions are also just paper announcements. Until I see game reviews of their physics acceleration solutions I'm not buying into needing 2 or 3 video cards either.

how much forums do you post this stuff on and what is you monthly wage from ageia. i might've be interested if i was the selling out kind but i aint.

I'm more than happy with the cloth simulation in other games without 'PPU Acelleration'.

To be honest, it really wouldnt surprise me if the whole Cell Factor thing was a fake, and the cloth simulation worked fine on a decent CPU, but that Aegia had deliberately hamstrung it when in software mode, to try to prove a point.

I think Ageia's product is a great idea. But yes $300 is a little high. $100 would be a better price point for mass adoption. I realize they have to sell it high now because they're not selling in mass yet. With the proper adoption by game developers, the product would be a great investment.

The issue is not whether we need physics acceleration in games, but how we're going to implement it. The article said it best that the beauty of GPU accelerated physics is that even if the game doesn't support it, you can still use both cards for rendering. But the nice thing about Ageia's card is that you can still use both GPUs for rendering, and have a dedicated physics processor thats designed to do only that.

The main things Ageia needs to do right now is get their costs lower and also switch to a PCI-E bus for the card. PCI is too slow. Another step would be to work with Nvidia and ATI at seeing if they can sell their chips directly to both companies, and put the chip right on the GPU. Then they could sell millions of chips which means they can offer the chip at low price to the big boys. It also eliminates the slow cross talk between the CPU and the PCI bus.

i notice this is degenerating into another well trodden physx debate. could we perhaps talk more of wehter or not games will actually need a PPU or is it a sham like the cellfactor demo.

so, how many people actually think ageia is a legitimate company and how many people think they rigged the demo?