Physicists on Game Physics

Seeing fluids act like fluids, and clothing moving and flowing like cloth will what will sell Physics to the common Joe....

When Joe goes to play his MMORPG.. and he goes over to his buddy John's house, and see's on John's computer the same game, but with water acting/looking much cooler, and when his toon runs around, the cape moves, collides and flows in the wind...

Then goes home, see's how his toon's clothing is rock hard, and the water just reflects, and only has the generic lame ripple effect... He'll go out and buy a Physics Processor...

I forsee physics-based games, the specialized hardware it runs on and the people that are attracted to it continue to form a growing community; it's happening right now. As more content becomes available and the technology improves, this niche sector of gaming will take on a larger presence. It's not going away.

In a similar way (but to a lesser degree) that the hardware-accellerated fps community branched away from it's software-rendered sibling, so too will this take on a fully supported life of it's own. That can only be a good thing for gaming as a whole, because computer gaming as an art or a hobby borrows from other aspects of itself to develop newer less mature iterations.

The physics components will continue to grow and become more powerful, and the games and simulations that are made will be there to take advantage of that. A persistant world and all that implies will take on a whole new meaning.

I don't think the whole Physics co-processor will really be anything except a niche tool until such time as it is integrated into motherboards or directly into a CPU (even less likely). It's not like a sound card where you can tell if you don't have sound or not, or like a video card where you can see that the graphics aren't any good. Joe Average hasn't taken high school physics, let alone be knowledgable about it to know if what he's seeing on the computer is realistic or not. I think CR is right - commercially acceptable is a whole other story from realism.

As long as the approximations are accurate enough, it's all good. Computers do loads of mathematical approximations. Even when it comes to a simple number like a 1/3, that number is typically just stored as an approximations (.333333..., the number of decimal places depends on various stuff).

To give a different example, light in realtime 3D graphics is usually just a very crude approximation to how light really behaves. Usually lighting in computer games isn't even computed in a proper physics-based way, rather some simplistic model / hackery is used to make lighting look somewhat decent. There's no way current computers can do accurate bi-directional ray-tracing (in this method, basically for each pixel in the graphics scene, rays are "shot" around and the colour of the pixel is thus calculated) for complex environments in real-time, and even ray-tracing which is pretty damn accurate, isn't an absolutely accurate way of calculating lighting, since light doesn't actually travel in perfectly straight lines (gravity affects light). But there's no real need for 100% accuracy and full, correct modelling of every single physical property of light (or any other physics stuff), when somewhat crude models do the job and give us smooth framerates.

why, as a gamer, would i care about the long term effects?

the only major situation i would see this would be with buildings - perhaps with RTSs at the forefront, but then again, "physics" and "RTS" are two words that are rarely found together (that is, what we think of gaming physics)

All that matters is plausible physics. For years tons of researchers were chasing the accurate physics dream and finally about 5 years ago they finally gave up and the meme switched to pursueing plausible physics instead.

The problem is that no matter how careful and precise you are you still will not be able to predict what will happen in the real world. There are just so many freak'n variables even dropping a single box is impossible to precisely predict what will happen in the real world.

yeah, but with about psychics on physicists on game phsysics?

I programmed advanced physics models as a job in the early 90's. If we turned approximations off, in 6 hours we could maybe simulate a whopping 1 "real-world" second !!! I'm jealous of the guys doing this today with the CPU speeds available.

Physics as a science is overrated. I'd rather see some in game chemistry, or maybe microbiology.

No link to the article in your news post :)