Slashdot Mirror
Does A Pentium 4 Need A Weapons License?
WindBourne writes "It appears to be that the U.S. house of Reps. want to classify Pentium 4 and above CPUs as weapons. This would mean that all these will require export licenses. Apparently, they have not heard about that the far east has developed large CPUs as well that are used in beowulf clusters." According to the article, this clause is unlikely to appear in the final version -- but stranger things have happened.
Foreigners could simply obtain SPARC or MIPS specs and fab a multi-GHz version of those. Since these chips are better designed for multi-processing, foreign powers could scale them just as high as a PIV cluster, and run their nuclear simulations. Time to worry more about refined Uranium-235 and Plutonium-239.
Javascript + Nintendo DSi = DSiCade
It wasn't "the House of Representatives", it was Representative Duncan Hunter, a San Diego Republican who makes Rush Limbaugh look moderate...and that's coming from a registered Republican!
The amendment will never leave the House.
-h-
It was actually the G4 because it was the first processor that could pull a gigaflop.
Most are manufactured in US; the only foreign countries where they are made are Ireland and Israel. They are packaged in various places around the world (Costa Rica, Malaysia, Philipines, etc.).
Doh! Quoth the article:
Section 1404 of the appropriations bill would roll back the licensing equation to a level not seen since 1994.
"The President shall require a license...for the export of goods or technologies included on the Militarily Critical Technologies List," Section 1404 of the House bill states. That list cites a level of 1,500 MTOPS as being militarily critical.
tasks(723) drafts(105) languages(484) examples(29106)
Yes it does. "Fab 30" and "Fab 36" are in Dresden. They make the Athlons at "Fab 30". I'm not sure if "Fab 36" makes Athlons or Opterons.
You forgot Japan. Fujitsu makes Sparc's. NEC makes pretty good computers, at least the Earth Simulator seems to be fairly fast...
The only way to know if a bomb will fission properly (i.e. it will blow up and not just very hot) is to test it.
Well, it's the only way to know 100%, but if competent engineers build a Little Boy (Hiroshima) gun-type bomb, they can be very, very confident without bothering to test it.
The Little Boy bomb design was never tested because it was such a no-brainer that it would work. Built as a back-up to the Fat Man (Nagasaki) implosion type bomb, it was always taken for granted that it would work, while no one was that confident about Fat Man, which was why the design was tested in the Trinity test.
In a gun-type bomb, you take a slug of fissile material with a hole in it, and build a gun into the bomb to literally shoot a fissile projectile into the hole. Nothing could be simpler in principle. You need precision and competence in the design, and you need to know that projectile will assemble into the slug, but not fly right through it, and you need to tend to some details I'm not going to enumerate, but that is pretty straightforward engineering.
Little Boy was not very efficient. It had an 85 lb slug of U-235 and a 55 lb projectile of U-235, with what IIRC was a modification of a common 3" gun to shoot it. Only 1.38% of the U-235 actually fissioned, but that was enough to produce an explosion equal to 15,000 tons of TNT.
Little Boy wasn't very "little" either (10 feet long, 9700 lb). But that isn't much of a package requirement to take out a city with a very high assurance factor.
It always escaped me why the US (or someone else) didn't simply mass produce gun-type bombs, rather than apply the tremendous amount of science and engineering to perfect the implosion assembly type, of which Fat Man was the first design of many.
It always escaped me why the US (or someone else) didn't simply mass produce gun-type bombs, rather than apply the tremendous amount of science and engineering to perfect the implosion assembly type, of which Fat Man was the first design of many.
The gun design requires a lot of refined material, which is expensive. It also doesn't scale. You can make implosion bombs use fantastically small amounts of material, or you can scale their yield up greatly, or use them as the trigger to a fusion bomb, and they will be cheaper to produce (even if more expensive to design) than the gun design. When you're making thousands of them, using less material is a significant gain. When you're planning on using thousands of them in a full-scale war that you want some people to survive, using less material is also a significant gain. And when you want to stick them on top of missiles or inside bombers to launch them at your enemies, making them as small and as light as possible is yet another significant gain.
Mod down posts with a "Free Mac Mini/iPod" sig, they're spam!