If you read the article (no i'm not new here) you would know that the claim of demo boxes was made by neweggs supplier and that newegg rejected that story. Ofcourse the summary here on slashdot changed that into newegg rejecting the story of fake cpu's all together.

Among graphics specialists it has been well known it is not linear. However many people think it is linear because most of the time you can get away with it. But if you make a grid of black and white pixels and a square of pixels with a value of 127 you can easily see there is a significant difference where you wouldn't expect one.

Actually a good scaling algorithm should perform a lowpass filter when downscaling. This is similar to downsampling of digital audio where you do need to filter out frequencies above half the sampling rate. Leafing these higher frequencies in would cause noise because they can not be faithfully represented in a lower resolution file.

Actually it is good to use a particular gamma for storing an image because the human eye is not linear. If a gamma of 1 was used to steps at the black end would be to large and at the white end to small (when using 8-bits per channel).

We are not using a gamma of 2.2 because the old CRT's did the CRT's had a gamma of 2.2 because it was determined that would work well.

Indeed while you would like to minimize body pitch and roll as caused by forces like acceleration and going through corners it actually is important also for a car to be able to quickly follow the angle of the road. That is also one of the reasons most supercars have their engine in the center. Centering the majority of the mass at the center of the car will make it turn, roll and pitch easily!

However there is another reason why you do not want the gyroscopic effect. To store enough energy in a flywheel that doesn't weight very much requires extremely high rotational speeds which means the gyroscopic effects will be very strong. Strong enough to put a high strain on the chassis and destroy the bearings of the flywheel in the process.

I actually expect they are using counter rotating flywheels with a vertical axis. Suspended in a construction allowing (limited) free rotation on the other axis's and an electrical system to transfer the energy.

Two counter rotating flywheels will NOT cancel out each other! Only the reaction (precessional is the official term i think in english) forces are canceled out!

Let's say the three axis are x, y and z. Then when you have a single flywheel which is rotating about the x axis it will resist rotating along the other axis and while react with a force that is perpedular to the the rotation and the force. When adding a second counter rotating flywheel it will cause a reaction force opposite to that of the first flywheel so the reaction forces are canceled out. However the combination still resist rotating along any axis other then it's axis of rotation.

Propably not

However if they made some software for some Microsoft branded computer they would be allowed to limit the use of that software to that computer.

Okay, you give me a million euro's and i'll give you a million dollars...

A good touch interface is easy to use and can be fairly efficient however in many cases keyboard and mouse beat it speed wise. I can know, I work on a touch based point of sale application. I also can tell you those things are only comfortable to work on if you are standing. Which often puts the keyboard to low to be able to type comfortably. In my opinion apple is doing the right thing by using large multi touch touchpads and no touch screens. Especially on laptops touch screens have the problem of requiring a fairly strong hinge to standup to the poking or the touch screen has to be so sensitive that it triggers each time your fingers just brush the screen accidently.

In the property window that is true however windows explorer will report the filesize in the size column of the details view. Even a file of only 1 byte will be reported as 1kiB and this is NOT the cluster size, that's 4kiB (just verified this claim on Windows XP Professional SP3).

As far as I know NTFS and HFS+ and most other filesystems all use 4kB pages in most cases.

One nice touch about file size oreporting in OSX is the way it handles related files. I mean when you have an image there is a very good change there is also a hidden file containing a thumbnail for the file. The size of this hidden file is added to the size of the normal file in the finder. So you do not get the effect that the total of the files doesn't match the total folder size. On the other hand it can be confusing when your images decrease in size when you upload them because the hidden thumb is not uploaded.

There will be a box however for tiger users containing Snow Leopard, iLife 09 and iWork 09 for $169.

But maybe they could make more money in the same time doing something else which makes doing a linux port the wrong choice for a commercial business.

Then get a Mac atleast you will have unix underneath.

Nope that's the wrong word. It is called copyright violation. Maybe you like to call it stealing because you think it is just as bad. Well maybe it is but that doesn't give you the right to change the meaning of the word stealing. Stealing in the context of music would be if person A composed a song and person B registers the copyrights for it. Now person B has stolen the rights on the song instead of violating them.

I started off in the PC world with a 486DX/33, with 256K or RAM and a 20Mb hard drive

Was this a system build from spare parts? It is a weird configuration. Allmost every 486 that was sold had 4 MB of RAM in it and a standard harddrive at that time was about 100MB. The amount of memory is especially low as a 486DX would require 4 SIMMS because it had a 32-bit data bus and the simms where only 8 bits. So you had 4 simms of 64 kB each!?