Stardock seems to have made a way to test your theory. You can use Start8 to bypass the start screen on boot, heading directly for a desktop with start menu, then you can use ModernMix to run metro apps in a window; the app layout still appears to be full-screen sized with two scrollbars, however, so it's not precisely what you're talking about. Those apps are $5USD each though, so I've only personally acquired the first (as no metro apps I've seen seem useful on desktop).
I think I read that Windows Blue is adding the ability to use metro apps at different sizes, so perhaps it'll be possible to do it properly at some point. At least it's proof that implementing what you're talking about would be really easy for Microsoft to do.
NVIDIA Surround and AMD Eyefinity are both fairly clumsy technologies; both approaches merge two or more physical screens into one logical screen. Whilst active the spanned mode results in oddities like a stretched task bar, the inability to properly borderless maximise windows to one monitor only, and things such as full-screen movies which would usually fit on one monitor with black bars above/below will instead stretch across the three and look terrible.
The best approach is to get 3D software to support three screens without crutches like Eyefinity. I've seen it with some mild success in Supreme Commander 2 and OpenSceneGraph, but it looks like it'll be down to graphics API and game developers to support multi-monitors properly. Given the only groups of people that seem to be really interested in multi-monitor solutions are the simulation crew (driving or otherwise), it doesn't seem like the available support is going to improve any time soon.
TechReport analysed the nVidia 680 a bit after its release and had a piece on adaptive vsync which should answer your question.
Quoted from an nVidia software engineer:
There are two definitions for triple buffering. One applies to OGL and the other to DX. Adaptive v-sync provides benefits in terms of power savings and smoothness relative to both.
- Triple buffering solutions require more frame-buffer memory than double buffering, which can be a problem at high resolutions.
- Triple buffering is an application choice (no driver override in DX) and is not frequently supported.
- OGL triple buffering: The GPU renders frames as fast as it can (equivalent to v-sync off) and the most recently completed frame is display at the next v-sync. This means you get tear-free rendering, but entire frames are affectively dropped (never displayed) so smoothness is severely compromised and the effective time interval between successive displayed frames can vary by a factor of two. Measuring fps in this case will return the v-sync off frame rate which is meaningless when some frames are not displayed (can you be sure they were actually rendered?). To summarize- this implementation combines high power consumption and uneven motion sampling for a poor user experience.
- DX triple buffering is the same as double buffering but with three back buffers which allows the GPU to render two frames before stalling for display to complete scanout of the oldest frame. The resulting behavior is the same as adaptive vsync (or regular double-buffered v-sync=on) for frame rates above 60Hz, so power and smoothness are ok. It's a different story when the frame rate drops below 60 though. Below 60Hz this solution will run faster than 30Hz (i.e. better than regular double buffered v-sync=on) because successive frames will display after either 1 or 2 v-blank intervals. This results in better average frame rates, but the samples are uneven and smoothness is compromised.
- Adaptive vsync is smooth below 60Hz (even samples) and uses less power above 60Hz.
- Triple buffering adds 50% more latency to the rendering pipeline. This is particularly problematic below 60fps. Adaptive vsync adds no latency.
So triple buffering is bad because it could cause an intermediary frame to be dropped, resulting in a small visual stutter despite being 60fps. There's a video of adaptive vsync on YouTube.
Perhaps those who are getting used to it are simply ignoring the changed components and continuing to use their PC as they did under previous versions of Windows, much as I am.
It's actually pretty easy to ignore metro entirely if you install something like Start8 to replace the start menu. It's a bit daft that we have to - particularly with a third-party program - but once you've done it the general experience is more or less like using Windows 7 with a couple of minor feature upgrades (file history, multi-monitor taskbar).
Attempting to explain how to shut down the PC to older computer users has been a challenge though, particularly for my folks with dual monitors; it's quite hard to summon the charm bar properly in the gap between the two screens, and going to "Settings" to shut down is just...weird. I've just told them to press the power button once and let the PC shut itself down that way, or installed Start8.
You may be suffering a little from rose-coloured glasses. The third LOTR movie still had some pretty obvious CGI issues that were a bit jarring, such as the beacon flames (distant ones), Pippin climbing down the beacon tower, the lava behind Frodo and Sam whilst they exited Mount Doom, and others; most of the CGI worked really well, but some parts were immersion-breaking at critical moments. That its latest installment continues to have the problem in parts is probably to be expected, and not necessarily the fault of the frame rate.
It'll be interesting to see if this spawns some new creative building options in Minecraft - specifically tailoring a construction to act as a facade for a building. Theoretically you could transform a street when viewed through your device, which could be an interesting building challenge or competition.
The AR tracking component is about what you'd expect - requires localised areas of high contrast to track from and doesn't work well on non-flat surfaces - but for what it is, it could be pretty neat. Perhaps it'll end up working with Google Glass or similar later.
They advertise that they're using the PointCloud library for the tracking component, which is apparently free if you wanted to build your own app.
"Just think, with VLSI we can have 100 ENIACS on a chip!" -- Alan Perlis
