It is easier to fight for one's principles than to live up to them - Alfred Adler
I received a phone call just pre-CeBIT/Games developers Conference time from a full-on PR manager telling me to fly to San Francisco to see something breathtaking.
I decided to go there totally not expecting anything much. Imagine my surprise when I actually saw something cool, something that I think has the potential to change the face of gaming. I saw the hardware up and running and I am still convinced that this is going to change to face of computer gaming. It won't be fast and easy but it will happen.
The Asus card in our test rig comes with 128MB of memory and is available in the shops today. It is very easy to install it, you just plug it in a free PCI slot, connect Molex connector to power the card. You boot the machine, install the drivers, reboot and you are ready to rock and role-play.

The biggest problem of this marchitecture is the lack of games. We got a single demo of a really cool game CellFactor: Combat Training something that can really show you what the game is made of and we tried Tom Clancy's Ghost Recon Advanced Warfighter.
A lot of sites talked about Tom Clancy's Ghost Recon Advanced Warfighter, also known as GRAW and the implementation of the hardware physics is far from impressive. You will see a little bit more particles and that's about it.
CellFactor is totally different story. It really looks breathtaking. It takes ages to load on 4000+ with 1GB of memory and Geforce 7800 GTX times two in SLI. Even when it loads you need to play it for a minute or so to pre-cache everything and then it goes rather smoothly. To be honest we expected more than average of 30 frames per second but I can say that the game felt really playable at all times.

We like the game. You can move an object and blow suff up and shoot almost anything on the scene. You can create massive explosions with hundreds of pipes and barrels flying around. You can even kill your opponent with a barrel or a pipe that you can send flying with your left hand force. It's fun and you can even play it online with all the happy people that have PhysX card. The bad thing is that with a single 7800 GTX card it can drop down to 13 frames per second and that is really low for 1280x1024 resolution. It is a demo so it has to be better when the final version comes out. You won't be able to play it without PhysX card. It looks and feels good but it definitely could use some performance polishing.
Conclusion
Do we think it's cool? Yes, of course we do, but it will take some time for PhysX to get estblished. We suggest
that Unreal Tournament 2007 is about to change to face of gaming with physics. You need strong title to influence the
world and that's exactly what the graphics industry learned through 3DFX and the original Quake GL.
We weren't able to run any meaningful benchmarks on the card. We'll be able to do so as more titles tip up.
And, at $299 it is not a cheap piece of marchitecture. But if you have enough money to buy SLI you could spare some and spend it on a PhysX. You will be able to play a total of two games and these you will get with the card anyway So, at this time, it is more of a show-off feature than a real necessity. Cell Factor is cool but it's not in retail yet, it's just a cool-looking demo.
PhysX has a great future and by around Yule time we expect to see a nice selection of titles available on this hardware. Its time is coming but it might be a tadpole too early for $/200 -$/300 investments, just to play two games, well, a game and a demo, to be precise.
But bear in mind you could be investing in a bit of gaming hardware history and that gives you a bit of a buzz. ?