I personally doubt Nintendo will be going for a chip like the Larrabee. Larrabee is an unreleased and unproven technology that Intel has publically buried as a "software development platform". This means any actual CPU/GPU based on it is years off. I suspect that the Larrabee's real world performance was so far behind the theoretical performance that Intel pulled the plug to save face.

Sony have tried completely new parallel processing architectures twice before (the Emotion Engine/RSX and the Cell) with mixed results. End performance was usually good, but they were an absolute pain to develop for. In these days of multi-platform releases, noone wants to get left behind because their console is too expensive. Nintendo should take note.

I'm sure Intel would be happy just to get the IBM PowerPC out in favour of x86. All they'd have to worry about then is ATI/AMD: good x86 CPUs and proven GPUs.

For what they were intended to be (cheap laptops capable of basic tasks), they don't really need to change that much.

The problem is that they way people are using them is changing. As they are become more mainstream, people expect laptop features from them.

Video is now important, so the old 1.6GHz Atom and intergrated intel graphics just doesn't cut it very well anymore. In saying that, I do watch video on my eeePC on long flights and it does well, although I do have to either resize or reencode HD videos on my desktop or the performance isn't great. I have had plenty of people comment on it (in a good way) as well.

The reality is that netbooks are evolving into something more than a cheap laptop, but not quite a full featured laptop or subnotebook. Ultraportability and long battery life are key elements of its success, as well as the price. For those reasons, we won't be seeing Core2Duos, massive HDDs/SSDs or any kind of optical drive.

I like my eeePC because it lasts all day (depending on the settings), it takes a fraction of the space of my old laptop, and it is cheap enough that I let it take a little more abuse when it is on the road with me.

For my next upgrade, what I'd personally like to see:
Slightly higher resolution screens (1024x600 is a rather awkward resolution for the work I do)
Touch screen capability
A real graphics chip (ION, or Radeon)
A more powerful CPU (Intel CULV, AMD NEO or a dual core Atom)
More RAM
HDMI instead of SVGA output
XP Pro (at least) or a better Linux distro
Expresscard slot
Similar or better battery life
and more importantly, a similar price.

MS don't care about individual users, but if you say that you're the CTO for a company with over 500 employees and your contemplating Linux, and their sales managers will be swarming you with free stuff.

The funding of a development centre is nothing special: MS likes throwing money at training developers into exclusive development for their software. It is all an attempt to lock people into a Microsoft way of life, and the group they try to lock in the most are developers (cue Steve Ballmer running about screaming "developers, developers, developers..."). $60 Million is a drop in the bucket for them, they spend ten times more throwing money at schools and training institutes in the US.

The more interesting new is the Memorandum of Understanding that LG has entered into with MS. This agreement is specifically about "Mobile Convergence", and is probably an attempt by MS to lock LG into Windows Mobile. LG is a member of the Open Handset Alliance, so this is most likely a preemptive strike to prevent more Andriod phones from entering the market. MS have already had anti-trust problems in South Korea: I wonder what the South Korean Fair Trade commission and the US DOJ will have to say about this agreement.

Gerenal purpose != web broswer.

Nokia's N800 demonstrates that nicely. It has a web browser, but is not considered a general purpose machine. My cellphone has a web browser, but that is hardly a general purpose ultraportable either. Yes, web browsing is important, but it isn't the only thing that makes a machine general purpose.

You are right in that the arrival of netbooks has served to drop the price of UMPCs. However netbooks were intended to be cheap and designed for basic tasks only. UMPCs were designed to be small yet still powerful, even if that was at the limits of usability. Devices like the Raon fit in nicely because some people do want a device that will fit into a briefcase (as opposed to a backpack) with plenty of room to spare, and yet not be stuck using Windows Mobile.

I'm wary of netbooks. There is an oversaturation of them in the market, all with the same specs - the only difference being minor funcionality and a brand name. Raon isn't a big company, and they could have easily made another netbook like everyone else, but they didn't. They have created something else and so far, given various reviews, it has been fairly successful for them.

This is a UMPC, not a netbook and not a smartphone. First of all, it is no comparison to a smartphone they are designed for entirely different purposes. It's comparison to netbooks is quite a bit closer, but netbooks are designed to be cheap and only powerful enough for most basic tasks.

Most netbooks have the same 1.6GHz single core Atom, and the same Intel GMA945 chipset with intergrated graphics. While they perform a bit better than the original Celerons, they can't outperfrom the 1.2GHz dual core Turion 64. Netbooks also have pretty awful 3D performance, while the Raon has a dedicated graphics core (yes it is intergrated into the chipset, but it is a true Radeon X1200 core) with 128MB of sideport RAM (not shared).

It is far smaller than a regular netbook, at about half the size of an MSI Wind. Yes, it is more expensive, but it is smaller and more powerful. It is clear they were designed for two different things: the Raon is a UMPC not a netbook, and the price reflects that. It is certainly cheaper than Sony's UX series or an OQO.

Let's look at your wishlist:
browser - yes (with most cellphones having a browser I don't see how this is cutting edge)
video and still cam - yes (built in webcam)
wifi - yes (bluetooth too)
touch screen - yes
PMP - it is portable and should handle most types of media, but it may not quite fit in your pocket
UMTS - no (but 3G will be an option, and it has a USIM slot)
GPS - no (but not hard to add with USB dongle)
TV - no (but again a USB dongle may help there)

The Raon isn't perfect: it's battery life isn't great, the RAM is just 1GB (soldered in) and it should have an HDMI port instead of SVGA (yes, the RS690E chipset will happily output to HDMI). Despite its faults, it has plenty of advantages. Just because the Raon doesn't suit your needs, or your budget, doesn't mean that other people won't find it useful.

It seems like everyone is treating this like the demise of AMD.
It isn't.
They are simply spinning off their fabs into a separate company. They will still most likely use the company for most their fabrication, but it also lets AMD decrease the fab sizes of their designs without having to throw millions of dollars into building a new plant each time. They can outsource fabrication of smaller die sizes to companies who already have that technology (namely TSMC, IBM and Chartered Semiconductor) at least until the ex-AMD fabs match the technology. Outsourcing does add extra cost, but that is more than offset by not having the expense of maintining fab plants. In addition, by outsourcing to multiple companies they can guarantee their supply.

Subnotebooks are a category that describes really small form factor laptops, typically without bulkier features of larger laptops (such as optical drives or floppy drives). The MacBook Air is a subnotebook in comparison to the rest of the MacBook line, much like the Powerbook Duo was a subnotebook to the Powerbook line in the 90s.

Netbooks, as Intel described them, are a class within the subnotebook category that are designed to be far more affordable. SSDs are not a requirement for netbooks, as there are plenty of netbooks out there with HDDs.

@kccboy2004

You really were condescending, and for all the wrong reasons.

You also need to check your sources. Intel didn't invent the concept, they merely coined the term 'netbook'. Intel's own website is hardly going to be an unbiased source there. Intel introduced the term (in its current form) in Feb 2008, well after the ASUS EeePC success in 2007. Subnotebooks have also been around for quite some time before the EeePC, and while many were tiny like the Toshiba Libretto, they were also expensive. The Compaq Contura Aero, in the 90s, was intended to be really affordable, and it was for the time. To me this is the long lost ancestor to the 'netbook'. Intel didn't invent the 'super affordable' concept either, that was done by the OLPC project that Intel was a member of before leaving to make the ClassmatePC. The ASUS EeePC was supposed to be ASUS's answer to the OLPC and although it was more expensive, it was available to the general public (and long before the OLPC shipped too).

It is nice to see that you're willing to spend the extra $500 for a graphics chip, a fingerprint scanner, Windows Vista, and an upgraded WiFi module while getting stuck with a single core Atom. To me that isn't value for money.

For the same money, I'll be going with the Raon Everun Note. It's dual core processor easily trounces the single core Atom. Its graphics may be intergrated, but the chipset has a real GPU core and it has good 3D performance. It has a touchscreen, webcam, blutooth (and, as rumour has it, 3G capability). It does have a smaller screen, but the smaller form factor makes it much easier to travel with. It also doesn't have Vista, and to me that is a plus.

Adderz was right. Netbooks were supposed to be low cost (and even Intel emphasied the low cost part), and ASUS is going the wrong way with this. It seems that they are trying to edge into the subnotebook market, but with specs barely above the average netbook they are going to have a hard time.

How is the coding going to be easy? Intel haven't released an SDK for this yet, so there is no way to tell. Programmers today have a hard enough job effectiovely using two or four processors, yet alone a 32 core GPGPU. Just because it is an x86 instruction set, doesn't automatically mean it will be easy to program. Going massively parallel does seem to scale well, acording to Intel's figures, but the associated coding is by no means easy. Have a chat to a PS3 developer and ask about how hard it is to program for a Cell processor and it's 7 SPEs.

Offloading hardware specific features into software is an interesting way of doing things, and it is clear that Intel intend the Larrabee to be more than just a GPU. However, this extra layer of software translation comes at a significant performance hit, and adds another layer of complexity to coding.

As far as SLI and Crossfire, the performance would not double. SLI or Crossfire does not allow the Larrabee chips to share their coherent cache. It is that coherent cache that is a key part to the scalability of the cores on the Larrabee die. Because the chips do not share that high speed cache (and associated interconnects), and because there is some overhead to both SLI and Crossfire, the performance will not simply double.