A much more recent article has appeared at Tom's, further lambasting Ageia's performance. It's a considerably more interesting follow-up, as it focuses on how to play PhysX demo's without the need for the card - allowing the direct comparison of software vs hardware-enabled physics, and again it isn't good for Ageia.
The report is originally based on a post from 'Lowrider007' at HardForum which demonstrated how to work-around the PPU check that was put in place within the PhysX demo. Both Lowrider's and Tom's benchmarks show only a few fps difference between software and hardware-enabled tests, which is pretty negligible in the real world.
Although Tom's did go on to mention that various features such as cloth-emulation, caused brutal slow-down in software-only mode, the review still concludes that the PhysX card still doesn't warrant expenditure stating: "Even using the specially-designed game demo gives us nothing compelling to recommend purchasing a PhysX based card.", adding that although Ageia promise 65 titles to utilise the hardware by the end of the year, multi-core processors should be more than capable of taking the physics burden.
Ageia has taken a lot of criticism since the launch of their PhysX PPU, and further driver revisions have done little to appease hardware review gurus. Sites and users continue to comment on the negligible differences that come with using the card, and are still banging on about the lack of support for the Ageia physics SDK, several months after release. Fudo's not going to be happy. µ
Sign up for INQbot – a weekly roundup of the best from the INQ