The new thing seems to be that you can actually switch between the onboard and 'real' GPU on the fly and fast while everything is running.

The previous laptops with switchable graphics, such as my Sony Vaio which had a Geforce and an Intel chips, did have to at least reboot the graphics system (on OS X) or reboot the whole computer (Windows) in order to go to the power saving mode.

In my experience, I usually was too lazy / didn't want to close my work and kept using the good GPU all the time. The only times I'd work up the enthusiasm to actually switch over was before a flight or something where I'd know I'd not need the power.

Are you sure?

An atmosphere is defined as roughly the pressure exerted by one kilogram of mass in Earth gravity on one cm^2. The article claims the area is also around one cm^2 large, so that means the mass would be about 3500 kg. A 2010 VW Golf weighs about 1,451kg, so this would actually be less than 3 cars.

Actually, GPUs started out doing all rendering in fixed point arithmetic, i.e. the equivalent to integers. That worked fine for rasterizing and shading for quite a while.

Then, they started doing limited-precision floating point support (with some 16-bit 'half' floats and other non-standard-conforming weirdness). Only later did they actually go on to support full IEEE floats. The current (as in, can buy them now) generation added IEEE double for both manufacturers, but of course performance is about an eighth or so of single precision.

I assume that's what the parent meant.

As an addendum, the newest CUDA 2.2 (with chip of the newest generation, i.e. GT200) actually has support for reading directly from (page-locked) host memory inside of GPU kernels... something I believe ATI cards have allowed for a while.

Because his system has 4 CPU sockets (with a quadcore CPU each) and I believe the licensing for all non-Server Windows version will only let you use at most 2 CPU sockets.

Any card that supports CUDA (i.e. anything since the 8800) should also support OpenCL.