Hmm... I guess I didn't make myself clear in my last post. What I meant was, if the camera could rotate 180, there could be a hardware (or software) switch at about 90 degrees that causes image/video capture to "flip" upside down to compensate for the physical rotation of the camera. That way images would be recorded correctly, without having to look at the finished product upside down.

I have to agree. Until someone gets some "quality" time with the tablet, these crumbs of information are all we're likely to get.

Let's see, what's left to know about it?:
- Release date
- Price
- Internal specs
- Camera details
- Storage capacity
- Screen cover [?]

I'll add a little more fuel to the fire here:
In the file /etc/osso-af-init/af-defines.sh there is code to define the INTERNAL_MMC_MOUNTPOINT to /media/mmc2. This is within an IF structure that tests if uname -m contains the string "armv6l".
This suggests a new processor that supports the ARMv6 architecture.

This Whitepaper on the ARMv6 architecture lists some interesting new features especially for multimedia.

Happy speculating

Let's make some bets

My guess is that the 880 CPU will indeed be a armv6 CPU. And I would say that this one is a good candidate:
http://www.arm.com/products/CPUs/ARM1176.html

Notable features:
- ARM Intelligent Energy Manager (IEM) technology which can reduce processor energy consumption from 25% to 50% - making them ideal for cell-phone, PDA, hand-held game and other portable
- Optional Vector Floating Point coprocessor (ARM1176JZF-S) (Powerful acceleration for embedded 3D-graphics)

My conclusion are based on the fact that:
1/ there are some if [ "`uname -m`" == "armv6l" ] all over Maemo source code
2/ Some folks may already be playing with armv6l CPUs on tablets: http://synthesize.us/Maemo/HTTP_User-Agents

But of course I may be wrong

DC, I understand what you're saying about the 90 degree correction point and was thinking the same myself-- BUT what you're hoping for is contingent on where the rotation of the webcam stops and starts. I wish I could say more but I can't.

Well, I don't have ARM11 based device to do real performance tests yet, but from the instructions timings description, it looks like Pentium4 from the ARM world Surely it contains some additional multimedia instructions (just like P4 with latest SSE) and it may get faster on some special tasks because of this (multimedia as you mentioned), but it will be somewhat slower on ordinary code if compared with ARM9 cpu per clock. See more info in this thread: http://www.internettablettalk.com/fo...ead.php?t=2406
I just hope that it will be clocked high enough to outperform Nokia 770.

But much more interesting features for me are unaligned memory access support and availability of fpu. If they get supported, porting applications will get much easier (a drawback is that if applications start depending on these features, they will be not usable on Nokia 770 and support only Nokia 870 (or whatever name if will have)).

Also I wonder about DSP core, will it be still used in this new device?

I gues it is OMAP 2420 chip. I think they will keep TI's DSP architecture for multimedia codecs. They have big know-how in OMAP. Also they probably have good deals at TI (OMAP is used in phones a lot) .

Good work :-)
When did those arm6 http requests start?

Uh-oh...none of those armv5 requests is using Opera 8.5; whereas the armv6 are. Could mean we are screwed for updates...

Texrat: I'd say that given that camera output is most likely displayed to the screen that when it's facing the user it'd be 'right-side-up'... flip it past the top side of the device and it rotates 180 degrees. That doesn't explain if it would automatically accomodate holding the unit in 'portrait' mode, and there's no word of tilt sensors. I just want to know how many megapixels it can pull.

You could say that. You might be right. Then again, if the designer was focusing strictly on webcam applications, he might be inclined to limit the rotation mostly to the front plane of the device... or thereabout.

As for megapixels, back to that hypothetical designer: the webcam is likely to be cheap, to keep cost down, and not designed for anything more than video cam use where super high resolution would be impractical.

I'd offer to bet you on which one of us is closer to reality, but that wouldn't be fair.

Ok, Mr. Texrat :-) Since you obviously have a little more info than most of us would you be willing to comment on whether this hypothetical device will be released this year? Or maybe if it's any faster?