Hands-On with NVIDIA Tegra-Powered Tablet

Terry Olaes

I Used to be the [H] News Guy
2FA
Joined
Nov 27, 2006
Messages
4,646
Engadget went to London to check out NVIDIA’s Tegra-powered tablet and shared a bunch of pics and a video of the hands-on demo. It’s still a prototype so it obviously still needs work but it looks pretty cool, especially with that magnetic inductive charger.

So what we're looking at here is a Windows CE-powered, resistive touchscreen display spanning somewhere around 15 or 16 inches, with the same Tegra internals as may be found in the Zune HD or one of them smartbook devices. As we reported earlier, the company behind the machine is ICD, and this particular unit was built to try and entice T-Mobile into placing a few orders.
 
*rolls eyes" The MS ARM platform is being worked on for version 7; but was more aptly named when it was called wince (WinCE) than win mobile. Completely different codebases.
 
*rolls eyes" The MS ARM platform is being worked on for version 7; but was more aptly named when it was called wince (WinCE) than win mobile. Completely different codebases.

Verison 7, of Windows Mobile, not Windows 7.

Just to clarify, since you make an excellent point.
 
I'm sorry, but I don't get the point. Why would I want an ARM processor in a 15" tablet?
 
on a screen that big it's not going to make that much of a difference. A 10" netbook screen is about 5W; at twice as big this would use something like 10W. Versus the current 5W Atom an ARM could get about 50% more battery lifetime. Versus the upcoming 2W (32nm) Atom the gain could be at most 20%; for a 1W ARM chip ~10%.
 
The only way I can see the ARM making sense is if the 32nm atom's GPU isn't crappier than the Tegra.
 
on a screen that big it's not going to make that much of a difference. A 10" netbook screen is about 5W; at twice as big this would use something like 10W. Versus the current 5W Atom an ARM could get about 50% more battery lifetime. Versus the upcoming 2W (32nm) Atom the gain could be at most 20%; for a 1W ARM chip ~10%.

Yeah, but on something this big I would fully expect a Core 2 Duo, at least one of the CULV ones. Asus coupled a 1.3ghz C2D CULV (10w TDP) and an 8 cell battery into a 13" laptop and it gets 10 hours of battery life doing web surfing (6 hours of continuous video playback). I wouldn't even settle for an Atom in something that big.
 
At the point of comparing vs a CULV chip the arm should get roughly double the battery life...
 
heh... double the battery life but it would take 5x longer to encode a movie or whatever moderate task you throw at it. if they made something VERY thin, like 1/2" closed up max, then i could see the real appeal.
 
It's a tablet; encoding a movie isn't on it's design requirements list. :rolleyes:

Surfing the web and watching video is all it needs to do, and the tegra650 supports 1080p video playback, so you don't even need to transcode your blueray rips to watch them on it.
 
It's a tablet; encoding a movie isn't on it's design requirements list. ;eyescutout;

Surfing the web and watching video is all it needs to do, and the tegra650 supports 1080p video playback, so you don't even need to transcode your blueray rips to watch them on it.

The refernce was the slower ARM CPU might make having a longer battery life worthless, as it will just take longer to do any task, so it would have to consume more battery life in order to finish it's job.

The 650 can display 1080p...

it doesn't mean it can accelerate 1080p video.

And even so, what codecs?
 
The refernce was the slower ARM CPU might make having a longer battery life worthless, as it will just take longer to do any task, so it would have to consume more battery life in order to finish it's job.

The 650 can display 1080p...

it doesn't mean it can accelerate 1080p video.

And even so, what codecs?

http://www.nvidia.com/object/product_tegra_600_us.html
1080p H.264 and VC-1 AP Decode

If you are making encodes of 1080p videos of your trip to Timbuktu, then yes there's better hardware for that. But that isn't what this is for.
 
I don't know about you; but when websurfing my Atom CPU typically idles at less than 10% load. An Intel CPU consumes more power at idle than an arm does at full load. Assuming the ARM is fast enough to do everything it needs to in real time it's mostly a non-issue. Where it matters is that you've got a much smaller total set of applications to play with. The only major gotchas I expect are the availablility of GPU accelerated video codecs for all the common web formats, and the availability of major browser plugins at the same time as x86 versions. If these can be provided an ARM based system will walk all over a intel one in the web tablet role. If not I think an 32nm atom based system will be the way to go because the power consumption gap vs a CULV chip will be fairly large again.

Yesterdays leaked intel roadmap includes a graph showing the TDP for the 32nm atom+chipset being one third that of the 32nm CULV equivalent+chipset. Idle for the new atom platform is reported to be 2W; I haven't seen one for the low voltage platform.


http://translate.google.com/translate?js=y&prev=_t&hl=en&ie=UTF-8&u=http%3A%2F%2Fpc.watch.impress.co.jp%2Fdocs%2Fcolumn%2Fkaigai%2F20091127_331818.html&sl=ja&tl=en
 
Back
Top