nVidia turning it's GPU's into 'PhysX Physics Processors'

3D card manufacturers shouldn't take this the wrong way, but ittakes a lot to make us crawl out of the communal Eurogamer bed (yes,all the Eurogamer writers share a single large bed - we do it forfrugality and communality, which remain our watchwords) and go to ahardware presentation. There's a nagging fear someone may talk maths atus and we'd come home clutching the local equivalent of magic beans.And then we'll be laughed at by our fellow writers and made to sleep inthe chilly end where the covers are thin and Tom left dubious stains.That's no fun at all.
Then again, there's some things you can'thelp but go and have a gawk at. So when an invite claims, "All toooften new hardware brings with it a small performance increase - maybea 5-10 percent over the previous fastest thing. Wouldn't it be far moreexciting to see a speed increase of x20 or even x100... well, we'll behappy to show just that on Friday," you have wander along. Even thoughyou suspect it may be a trap and they're going to attack you withill-shaped blades, you have to find out what on earth they're talkingabout.
As we suspected, it wasn't quite what we were hoping for.Sure, there are programs which gain a x100 increase via the methodsNVIDIA talks about on this particular Friday, but unless you're workingin economics or astrophysics modelling, it's not exactly that relevant.However, something more quietly astounding was explained. Mainly, thatdespite the fact that no-one you know bought a PhysX card, if you're aPC gamer with a relatively recent NVIDIA card, you've already got one.Or, at least, you will soon. Spooks.

Get him!

The primary idea NVIDIA was trying to push was Optimised PC - the approach discussed in Rob Fahey's interview with Roy Taylorthe other day. The idea being that the traditional PC approach whereyou buy the fastest PC processor you can doesn't actually lend the bestresults, at least in most situations. If you spent more on -predictably - a GPU-driven 3D card, for an increasing number of areas,you're going to get much higher performance. If the program is usingthe GPU in a meaningful way, anyway. NVIDIA highlights areas likeimage-processing and HD video-encoding, as well as - natch! - games.You lose in single-threaded activities - like, say, just booting up aprogram - but they argue a small loss in opening a Word Document isless noticeable than frames in games or similar.
Where it startsgetting interesting is NVIDIA's development language, CUDA. The problemwith all the threading programming methods is that it's radicallydifferent to single-threading (and, yes, we're getting into, "Why wouldanyone care about this but a programmer?" territory, but its backgroundfor the key point later). It's hard to do, and CUDA is basically a wayto make things more accessible.
NVIDIA claims anyone experiencedin C or C++ will be able to get a grip on it (i.e. not us, but theaforementioned programmers). This means that anyone who codes in CUDAcan program the GPU to do pretty much whatever they like; it's byturning the 3D card into a bank of processors that the financialanalysts and the astrophysics guys are getting such impressive results.And impressive savings, as it's a lot cheaper to do it this way.

Now, NVIDIA claims that the fact GPU solutions are cheaper is goingto push better GPUs into more business machines. This will help pushthe idea that an okay CPU/good GPU machine gives better performancethan a good CPU/okay GPU, leading to more machines with better GPUs...and so, making more PCs abstractly available for gaming. Or, at least,raising the bottom level of hardware that you can expect people to have.
Interms of a more general use, transcoding video can take hours. Later inJuly, all GeForce 8000+ cards will ship with Elemental HD, a programwhich manages to perform the odious task - in the words of NVIDIA - "ina matter of minutes". The software will also be available for people todownload online, probably with a small fee ala Quicksave if theyalready have a GeForce card.
Point being: this CUDA malarkeyisn't something that's just for future NVIDIA technology. It'ssomething that allows the hardware many PC gamers already have to berepurposed.
For example, PhysX. NVIDIA's Physics 3D Card systemwas only supported in a minor fashion, as no-one would buy a card justto make explosions fancier, but with CUDA it can run on one of theother GPUs. A proportion of the 3D card's power can be given over torunning physics, giving those fancy PhysX-style interactions withoutactually having a specific card for it. CUDA's porting to PhysX willbecome available to the public in July, but developers already have thetools.

The Euphoria engine of Natual Motion. It's hard to illustrate this sort of thing.

You'llbe able to - for example - manually, up front, decide to devote aproportion of your 3D card's power to PhysX. Alternatively, developerscan commandeer it and do exactly the same thing. The new generation ofcards which are about to be announced are able to deal with pretty muchanything that exists on the highest setting with power left over, sothat power can be given over to acting like a 3D card would.
Andit goes further. Where previously you'd have just thrown out your old3D card when you upgraded your PC to a new one, if you have a G8000+ 3Dcard already, you can keep it, and just set it to concentrate solely ondoing PhysX tasks. This isn't a SLI situation where you need two of thesame cards working in tandem - any post-8880 card, rather than beingput out to digital pasture, can be given a job of deciding how bits ofglass bounce off a skyscraper, or similar. NVIDIA claims it's talkingto ATI to try and get them to use CUDA too, which.... well, we'll seethere, eh?
The potential is interesting. Demos shown include Natural Motion, whose Euphoria engineis heavily physics-dependent, allowing unique, convincing moments ingames. A straight collision isn't enough, as straight ragdolls areludicrous - the system involving AI (so the hit object will try andmove limbs to protect self and similar) leads to impressivelynaturalistic results. The first sign of this publicly was in GrandTheft Auto IV, but Natural Motion's own American football game, Backbreaker,is a fascinating example of what a physics-heavier approach tocollisions can give games. And, with CUDA-esque use of GPUs to do thisstuff, the PhysX related boon is accessible to even more of us.
So they did talk some maths, then, but we survived.