Ageia Interview


posted 8/7/2006 by The GN Staff
other articles by The GN Staff
One Page Platforms:

GamingNexus: From a developer perspective how do you differentiate yourself from your competitors?  Why would a developer pick your product over something else in the market?
Michael Steele: There are a number of reasons why developers are embracing PhysX in such a huge way.  The open nature of our SDK licensing and usage helps, as does our world-class cross-platform engine software and tools coupled with the unique advantage of the PhysX processor.  Offering this complete, integrated solution to developers gives us the competitive edge.  Doing it without charging huge licensing fees makes it all the better.

GamingNexus: One advantage Ageia has is that your software engine comes with the Unreal 3 Engine.  How did that deal come about and how big of an advantage do you think that gives you over the other physics companies in the marketplace?  Are you trying to negotiate a similar agreement with other companies?

Michael Steele: AGEIA is working with a number of other engines and we see huge value in this as it allows many developers to leverage our technology while focusing on the creative aspects of their game, rather than implementation and integration details.  Epic – the developers of the UE3 engine – have always been at the forefront of new technology and showing off the best advances in gaming.  As such, we are very excited to be working with them on showcasing what PhysX can do and what the UE3 engine can do across many titles and genres.

GamingNexus: We asked the Havok folks about co-existence between your approach to physics and theirs last week. Do you think you two can co-exist in this market or will there have to be one standard that everyone supports?  Do you think Microsoft will ever bundle physics into DirectX or is this something outside their purview?

Michael Steele: AGEIA essentially makes its money from selling PhysX processors, so to a large extent we are API-agnostic.  Since there is no standard (like DirectX) for physics, we needed an access mechanism that could talk to our hardware and so the PhysX API and SDK are necessary and at the same time were designed and optimized to take best advantage of the processor.  As the only hardware physics API used in shipping games and with products on the shelf today, in a sense PhysX is the hw physics standard right now, but something like DirectX standardizing the access mechanism would truly establish hardware physics as a mainstream, required element of future games and that just drives more hardware sales for us, so we’re all for it!

GamingNexus: PCI card slots are starting to go the way of the ISA slots and the monochromatic monitors, how long before we see a PCI-E version of the PhysX card and do you think there will be a performance boost from the card?

Michael Steele: Right now we are looking at the trends from PCI to PCI-E and you’ll certainly see PhysX keep up as PCI fades away over time.  Initially though, we knew many gamers have PCs where their PCI-E slots are taken up by one or more graphics cards, and their spare PCI-slot was the most sensible place to drop in PhysX hardware.  Due to the very different nature of physics processing from graphics, PCI is actually not a bandwidth limitation today so any future PCI-E product would be more to support those gamers with PCI-E slots available over PCI ones

GamingNexus: One of the big trends in video cards right now is hooking multiple video cards up to increase performance (SLI/Crossfire), do you think there will ever be a point when you’ll have to consider such a configuration for physics cards?
Michael Steele: Currently, developers are just beginning to take advantage of the current PhysX processor so while there’s nothing precluding such configurations in the future, for now we believe the PhysX Processor has more than enough power and feature head-room that developers won’t be disappointed by the capabilities at their disposal with one card,

GamingNexus: One of the panaceas of game design is a fully interactive/destructible environment, do you think in game physics will get us there?  How far away are we from achieving that?

Michael Steele: We think this is certainly possible, but the challenges are less about the technical feasibility and more about smart level-design and compelling storytelling.  Games have goals built into them and destructible environments offer a far broader palette that game designers can paint from, but the results are still up to the artistic and storytelling skills of those designers.  If you look at a title like CellFactor: Combat Training (from Artificial Studios) or even something more simple but incredibly smart and compelling like the puzzle game Switchball (from Atomic Elbow), you can already see the power that advanced gaming physics offers smart game developers and we’ve only been available for a few months!

GamingNexus: How long do think you have before the game physics market reaches critical mass? (i.e. it goes from being a front of the box feature to being a back/side of the box checklist item)

Michael Steele: In many ways this of course depends on how games incorporate the technology and how it makes the gaming experience more compelling to the end-user.    The killer-apps and critical mass are just around the corner with titles like CellFactor and Unreal Tournament 2007 amongst many other titles.  Already the level of industry support, amazing effects and gameplay created that are impossible without dedicated physics hardware, and buzz about physics is amazing; we think the future for PhysX is extremely bright.  We’re over ten years into the 3D graphics add-in board category and not even four months into the PhysX add-in board category.  This is just the beginning and as avid gamers ourselves, we invite everyone to join us in pushing the next big thing in PC gaming.

We'd like to thank Michael for taking the time to answer our questions as well as Susan who helped to coordinate the interview.

Page 2 of 2