PDA

View Full Version : Cellfactor Real-Time ClothTear Video


jAkUp
05-04-06, 12:56 AM
Amazing!

Click here (http://media.putfile.com/Movie-97-89)

einstein_314
05-04-06, 01:07 AM
Wow!! Sweet.

stncttr908
05-04-06, 01:08 AM
See, this is the kind of stuff that will put AGEIA cards in our systems.

knghtwhosaysni
05-04-06, 01:29 AM
See, this is the kind of stuff that will put AGEIA cards in our systems.Maybe if someone else will pay for mine...

But on a slightly different subject, does anyone know how this is gonna work online? Is it just gonna be the host's computer who does the calculation? Cuz I think that's how it works now. But then would each player need ageia if they are not each doing the physics calculations? Because if each player did the physics calculations for their own systems, I would think eventually there would be packet loss, lag, or somehow incomplete information would be sent to the player's Ageia cards that might result in objects landing in different spots after the misinformation.

Would anyone like to clear this up for me?

Zelda_fan
05-04-06, 01:51 AM
If something like this was incorporated into an RPG like Oblivion, I'd have no problem shelling out $300 for one of these cards.

CaptNKILL
05-04-06, 01:51 AM
Maybe if someone else will pay for mine...

But on a slightly different subject, does anyone know how this is gonna work online? Is it just gonna be the host's computer who does the calculation? Cuz I think that's how it works now. But then would each player need ageia if they are not each doing the physics calculations? Because if each player did the physics calculations for their own systems, I would think eventually there would be packet loss, lag, or somehow incomplete information would be sent to the player's Ageia cards that might result in objects landing in different spots after the misinformation.

Would anyone like to clear this up for me?
There are already slight issues with this in current games with advanced physics. For example, in Half-Life 2, someone will grab something with the gravity gun and it may appear to teleport to them form the other side of the room because the physics worked out differently for you for some reason.

All I can think is that its either going to take a huge amount of constant synchronizing (which I dont really know anything about), super super accurate physics calculations (that are perfect nearly every time) or the server is just going to have to do it all for the clients... but I think thatd create an enormous amount of crap that needs to be sent, so itd probably lag like crazy on all but the fastest internet connections.

As to ageia vs non ageia physics... they would probably both have to have the same number of objects that actually have an effect on gameplay, but the ageia system would have smoother and more responsive physics(but still the same things going on) and more "fluff" type stuff like accurate cloth physics and lots of tiny bouncing particles that dont really effect the player at all.

msxyz
05-04-06, 03:12 AM
Honestly, the impact of "credible" physics is nothing compared to what 3D accelerators brought to the table. The physics we have now are enough redundant for the task. I can go fart the point of saying that Karma in the Unreal Engine 2 (a very modest physics engine but that will run on a modest P3 500MHz as well) is more than enough for a gaming environment.

All the rest is just a gimmick. What does it add in terms of gameplay? Did someone actually like the stupid "monkey-push-the box" puzzles of half life ? Or clearing the road mindlessy tossing the debris aside?

I'd rather have the resources invested somewhere else. Give me some more credible AI that does not need pathnodes and that can figure for themselves the best way to navigate in a 3D environment. Give me an AI that's not so predictable or that "cheats" by knowing the player position and moves.

The fun part of this is that it's much less computation intensive to code a proper AI algorithm for a bot that investing resources in a high level physiscs simulation updated at every frame.

walterman
05-04-06, 08:20 AM
Is the SDK available ? I've interest in virtual clothing simulation. I would like to know if they're using linked particle systems for the cloths, or if they can use other more advanced methods like high order surfaces.

Edge
05-04-06, 08:58 AM
Honestly, the impact of "credible" physics is nothing compared to what 3D accelerators brought to the table. The physics we have now are enough redundant for the task. I can go fart the point of saying that Karma in the Unreal Engine 2 (a very modest physics engine but that will run on a modest P3 500MHz as well) is more than enough for a gaming environment.

All the rest is just a gimmick. What does it add in terms of gameplay? Did someone actually like the stupid "monkey-push-the box" puzzles of half life ? Or clearing the road mindlessy tossing the debris aside?

I don't think you have any idea what physics can actually add to games. When we start seeing prescripted animations replaced with life-like real-time physics driven animations, we'll start to see a whole new world of gameplay open up. It's pretty sad that a 7-year-old failed game like Trespasser brought more innovative uses of physics to the table than new games do just because no developer is willing to bother putting any time and effort into something when people are more wowed by flashy lights than they are by actual advancements in gameplay.

FraGTastiK
05-04-06, 10:00 AM
... When we start seeing prescripted animations replaced with life-like real-time physics driven animations, we'll start to see a whole new world of gameplay open up. It's pretty sad that a 7-year-old failed game like Trespasser brought more innovative uses of physics to the table than new games do just because no developer is willing to bother putting any time and effort into something when people are more wowed by flashy lights than they are by actual advancements in gameplay

QFT.

Marvel_us
05-04-06, 10:03 AM
Nice find. This game should be very interesting.

jolle
05-04-06, 11:38 AM
Only release date Ive seen is Q4 2007, anyone seen anything else?
sounds like its going to be a bit late to the table, especially if the PPUs dont catch on..
I think CellFactor will require a PPU (read it somewhere) to run at all, and it makes sense since the physics seems to be key to the gameplay, and it LOOKS like a UT style online game.

msxyz
05-04-06, 11:40 AM
I don't think you have any idea what physics can actually add to games. When we start seeing prescripted animations replaced with life-like real-time physics driven animations, we'll start to see a whole new world of gameplay open up. It's pretty sad that a 7-year-old failed game like Trespasser brought more innovative uses of physics to the table than new games do just because no developer is willing to bother putting any time and effort into something when people are more wowed by flashy lights than they are by actual advancements in gameplay.
It's a matter of priority imho. I see credible AI as a much more needed aspect than having a realistically animated characters with flowing hairs and cloths still moving through the virtual worlds like one of pacman ghosts.

Honestly I believe all this buzz around physics is done purposefully to offer more of the same old soup and hide the fact that no substantial innovations have been made in the last years on the gameplay/game mechanics front.

jolle
05-04-06, 11:54 AM
Honestly I believe all this buzz around physics is done purposefully to offer more of the same old soup and hide the fact that no substantial innovations have been made in the last years on the gameplay/game mechanics front.
How X-File:ish of you..

I think its basicly the result of a moneydriven industry, whats hot and new and sells games like butter? well since HL2 its physics.
So you gotta have alot of physics to make a game attractive.
As you gotta have top notch graphics to make it attractive.
And what games make it to market or not is decided by the big publishers, who make all their decisions based on money, they "invest" in games they think will sell well, ie a good return on the investment, and they poke at developers about things they feel important to sell the games, sometimes forcing devs to compromise their original vision.

I think thats pretty much why games are becoming more and more like hollywood blockbusters, more and more games are "safe bets", things proven to sell well, and often that can get a bit gimmicky, you need this "one big new thing" to make it interesting enough.
Publishers rarely take intrest in inovation, cause there is always a risk in that.
Hopefully delivery systems as Steam can offer developers who are willing to take a risk with a new concept, a chance to try out their vision and then sell that game without conforming to a publisher.
But like Carmack said in a recent interview, the requirements and time to make gamecontent with the current level of detail possible is ALOT higher then it was when the modscene started to explode.
But still, maybe there is a "indie gaming" scene coming in on a larger front with systems like Steam, looking at Darwinia for example, who never did find a publisher outside of the UK, but is now available for purchase world wide on Steam.. I belive they won some indie gaming award aswell.

EDIT
Physics btw isnt all bad, it CAN add alot of gameplay value, I mean the PPU addition to GRAW obviously doesnt, but CellFactor pretty much DEPENDS on it.
HL2 used it in a pretty sensible way, and it could be used for totally new and innovative things, its basicly just another addition in the toolbox, it just needs someone to acctually do something useful with it.

jeffmd
05-04-06, 12:59 PM
the demo movie really impressed me.. how ever unless theres a hole to give me a blow job, there aint no way I am going to shell out 300 bucks for a card whose sole purpose is to accelerate physics in a half dozen games for the coming year.

Make it a 100 doller card and I'll think about it.

CaptNKILL
05-04-06, 01:02 PM
the demo movie really impressed me.. how ever unless theres a hole to give me a blow job, there aint no way I am going to shell out 300 bucks for a card whose sole purpose is to accelerate physics in a half dozen games for the coming year.

Make it a 100 doller card and I'll think about it.
:lol2:

Q
05-04-06, 02:08 PM
Too bad that was probably being run at 800x600 on a QuadSLI system with an FX-60 and 4 gigs of ram.

Edit: Whoops, forgot to say phasechange overclocked to 4ghz.