@(Unverified) like the 5xxx? What? The Nvidia 5xxx series was universally lampooned as a crap overpriced rebadge of the excellent 4xxx. And the 6xxx introduced SM3.0... if you bought a top of the line one when it came out in April 2004, it'll still run almost any game at a tolerable clip with decent effects, albeit likely not at full HD resolutions.

That said, good luck nvidia. Hopefully you'll consider selling ARM IP, and not just your particular Tegra; the mobile world really needs more freefloating IP to power it.

@shnoogie lay off the crack pipe. nvidia's chipsets have traditionally been luxury sedans, laced with fancy features. to this day, people still seek nforce2 chipsets for their realtime dolby encoding capabilities. nvidia introduced close-to-core accelerated gigabit pipes, & the first series of usable non-shiet onboard video. nv lost relevance as more and more features became standard and less important, but they've been the king of throwing on ridiculous amounts of accelerators and gimmicks for years. after all, they were tasked with differentiating themselves from baseline chipsets-- it was up to them to make it clear to consumers there was a reason to choose nv, and not go with the vendor providing the cpu's chipset (amd/intel).

on the other hand, luxury as they once were regarded-- nv chipsets have always been huge power hogs.

@JS sorry, but MALI has always been regarded as a joke. Its performance is abysmal. PowerVR is the only player in the ARM SoC space, although Creative's Zii may shape up to be a contender. AMD used to sell some good IP, but they sold the lot to Qualcomm. Neither Creative or Qualcomm appear to have any interest in licensing IP, which is where PowerVR and MALI both target.

@kleptophobiac AMD catalyst drivers have been non-problematic for ~3 years now. AMD apparently realized the maxim "fancy chips with bad drivers are just really expensive glass", and reformed the hell out of ATI's completely unworkable drivers. Even in infamous and universally reviled Vista (thanks Aero!), little irks like mode switching have been smoothed over.

And more importantly, AMD open sources its device specs, which is why in a growing number of cases the open source AMD drivers are the most performant drivers out there. You might not care now, but when you're running WebGL in ChromeOS in another three years and its leaving everything else in the dust, I think people will begin to reconsider.

NVidia chipsets have always weighted in with relatively high power consumption. Good luck transitioning to SoC, where power is of the utmost concern.

I still cant believe AMD sold their relatively successful 3d System-on-Chip (SoC) intellectual property (IP) to Qualcomm. They were one of two vendors, themself and PowerVR, and now its just PowerVR selling 3d IP.

VGA port, I know, the travesty of it! Even the blind deserve modern standards like HDMI, for shame BrailleNote.

replying to the only sensible comment here--

yes, hdmi d connector is great-- half the width and half the height of the common type a.

the only sensible upgrade from the hdmi type d would be some kind of wireless spec: wireless hd, or whdi.

720p 60fps is the real use for this over the ContourHD. Also, an external mic and/or BT headset communication is basically a must.

Only if USB3 can radically alter the latency. Which it might be able to do. Theres probably a big enough pipe to pack mildly encoded 720p video down USB2. But the latency and jitter of USB2.0 makes them painful to watch; a .5 second delay between you and the screen is half an order of magnitude worse than the latency to your friend on the other end. USB3.0's big draw for me is its more asychronous nature, which has the potential to significant reduce latency, and bring USB on par with the acting "respectable" interface, Firewire.

USB3.0 does open up interesting possibilities of 60+ fps webcams. It'll take a while for a market to realize around that, but it should provide a compelling experience.

should be supported via upnp/dlna, no?