> Firstly, you need a mod chip for each Xbox. Secondly, you need to install > them into each Xbox. Thirdly, how many Xboxes, mod chips, expensive Myrinet > networks, and how much labor would it take to replace *one* dual Athlon 4? > > Striclty speaking, for some hard problems, where "connectivity" is very > important, even 1000000 XBoxes working together will not reach the same > performance. > > Luckily, most interesting number-crunching problems fall somewhere in the > middle: connetivity is important, but some parallelization is possible. > > Our lab had to write a specific grant application for a Linux cluster. I just > don't think a grant application asking for well over $20,000 to buy Xboxes, > mod chips and network hardware has even the slightest chance of being > approved. > > And even *if* one goes through the process of raising the money necessary for > this, if Microsoft comes out with the new Xbox version that resists hacking > by that time, you may just get stuck with 100 game consoles. > > As to the legal issues, IANAL and I don't even know under which jurisdiction > you live. Consider my ealier statement a friendly suggestion to consult one.
Ugh. I have to tell you, why is everyone so damned worried about "why would you want to do it." Why? Because it's possible. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
