I'm a fan of Nvidia and their products but this claim is fairly meaningless.
Nvidia is often "ahead" of the competition when they announce something and then right about in line with the rest of the industry once they get all of their delays out of the way and finally release something.
Just to pile on: mobile processing generations are gated more by power limitations and process technology. Unless nVidia chips are going to be exclusively printed on a 'generation-ahead' process, have access to a 'generation-ahead' battery, and/or be paired with a 'generation-ahead' display, they can't be a 'generation-ahead' in general performance when they wind up in a device. Extra bells and whistles that provide theoretical performance, performance that the device ultimately can't leverage, are routinely excised from shipping designs.
You don't have to look any further than the routine underclocking to see that mobile chip roadmaps are simply not as relevant as desktop or even laptop chip roadmaps.
Except that there is nothing that matches Fermi for GPGPU computing on the market. Certainly nothing from AMD and Intel who are still trying to figure out how to match NVidia's lead in GPGPU computing
Interesting piece on the roadmap: Tegra 3's maximum supported resolution (1920x1200) will still be a notch lower than the (unlikely) rumored iPad 2 display.
Yeah, seems like marketing bunk. Especially considering how their new shiny Geforce 5xx line is only a 10-15% improvement over the last gen. Show me the numbers!
5xx was conjured out of thin air like 9xxx, 1xx and 3xx. 580 and 570 use the gf100b chip -- just a minor revision of the chip that powered 480 and 470.
I like NV products, but for some very weird reason their marketing likes to boost the family number for no reason every 6 months.
Nvidia is often "ahead" of the competition when they announce something and then right about in line with the rest of the industry once they get all of their delays out of the way and finally release something.
See: Fermi, Tegra 2, et al.