It's just that I wouldn't be surprised if they attacked it later on if it could raise their profits somehow.
And the shareholders wouldn't expect any less, as long as it's all aboveboard. It's when you pull an Enron that shareholders frown. If you just pull a Microsoft and give a big ol' "fsck you" to your former affiliates (read:IBM and OS/2) but manage to rake in dough by doing so, shareholders cheer.
First of all, unless the memory is embedded on chip, you've already lost-- slow bus speeds, limited pins. Second of all, once you have your on-chip cache, a single wordline might be very fast... if it's a kilobyte wide, then you can read 1024 bits every cycle no problem... but no cache has only a single wordline (unless it's a register), and no bitline has a single bit. So now you have a word address to decode, and a bitline with lots of bits on it and a sense amp at the end... and a bit address/mux if you want it. How does this affect latency? Well, that depends on the size of the memory not the speed of the memory cell. The bigger you make it, the longer the access. A single storage element is plenty fast... if the access isn't random, interleave intelligently and there's no problem. It's a size problem: you can't have every book in the library of congress at arm's length (physically). You've got your phone book, dictionary, and the last few books you read. The rest take a bit longer to obtain.
Re:The Heat Issue - The article is bullshit
on
The Diamond Age
·
· Score: 2, Informative
Granted this is interesting, and may one day revolutionize chip manufacturing. But don't hold your breath.
These days we make chips on silicon with copper wiring. Copper melts at 1064C, silicon melts at 1414C. Looks like finding an extremely high temperature conductor on par with copper to use for interconnect might be a problem the article completely misses.
Chip manufacturing is such a mature technology that just changing the interconnect from aluminum to copper took decades of work. Don't expect to see diamond substrate on real chips any time soon.
Japan is investing six million dollars in this research?? Come on! Intel is investing six billion in R&D just to figure out how to drive the next generations of silicon technology!
Great legal response to the patent infringement claim.
"But your honor, we didn't know we were infringing! Oh yeah, we didn't bother to read the nasty GPL thing either any more than we gave a hoot about IP until this March, but we didn't know the code we worked hard to develop, err, well, purchase was in nasty Linux... well, yes, the Unix code we developed, err, purchased is far older than IBM's patents... shit. You mean we have to care about other folk's intellectual property? This works both ways? I though we could just trash them without recourse. Well shucks."
Here's the patent
Prior art mentions graphical menus in MS Windows, so it seems the distinction here is allowing seamless navigation from one program's menus to another's, but I can't wade through all the legalese.
About one computer per soldier, since total personnel in 4/2002 was 481,266. Given the typical astronomical support to combat troop ratio, that's not huge at all.
News.com story.
Of course, IBM is actually selling services for the most part... when the Army's microcrap starts deteriorating, Microsoft probably won't answer the phone.
IBM's own POWER4 dual-proc eServer:
7029-6E3-221E (Linux ready)
POWER4+:
1.2GHz
2-way
8MB
2 x 36.4GB
2GB
$8,995.00 IBM Web price*
$237.74/mo. for 36 months**
I was under the impression that GM foods usually are modified versions of highly domesticated crops and animals that basically have no nautral version to compete with. Corn, wheat, cattle, etc. have been domesticated for thousands of years. They bear little resemblance to any "natural species" that might be eradicated by a GM version. GM seed creeping into a neighboring organic farm field is a different issue entirely than "natural" competiton. If GM dogs and housecats start competing in the wild with wolves and mountain lions, they're probably not going to make great pets. But yeah, species that aren't domesticated (Like fish) are a different issue.
The wing of the shuttle had a few hundred tons of thrust accelerating it upwards, while the piece of foam had lots of air friction and gravity accelerating it downwards. Who knows how many parameters you'd need to take account for to figure out the real impact velocity, or whether you just figure it out from the high-speed film of the wing strike, but I'm pretty sure if I throw something light out the window of my CAR driving HORIZONTALLY at 65 mph, the relative velocity changes pretty damn quickly.
I believe you somewhat blur the distincion between payload capability and "cargo" capability. It'd be neat if the Proton could launch combined payload of a 165,000-pound shuttle (yes, a ridiculously huge crew return vehicle, but still payload) and a 58,000lb "cargo," but I don't believe that there are any other current systems that can launch a few hundred tons into orbit. Yes, the fact that the shuttle reuses the three main engines blurs the payload/launch system boundary, but they are still reusable. Certainly the Soyuz rocket payload capability of about 12,000 pounds includes the crew capsule!
You campare the shuttle launch system to the Saturn V. Good comparison... every shuttle launch (except the last) is like launching Skylab and getting it back again in two weeks. Yeah, I can see why you might consider than inefficient.;) But Skylab only weighed 78,000 pounds, so maybe two Skylabs is a better number.
Yes, perhaps they promise less resistance than copper interconnect of the same size, but isn't a diameter of 100nm actually a bit large? Can nanotubes shrink, or is their diameter a chemical requirement? According to the International Technology Roadmap for Semiconductors, copper wiring pitch should now in 2003 already be 245nm. So with 50% spacing between those nanotubes, you're not even talking a 2x improvement in size over current interconnect. What if the things are too big to be used as interconnect for those 35nm gates we're supposed to see in 2007?
This article posted today at Silicon Strategies is probably just a rehashed press release pointing out that Motorola has made some nifty new nano-crystal flash with its 90-nm process technology. But if anyone's interested in it, it's those of you who've read this far.
Slashdot Mirror
Old event, new "news" story from NASA's press people. The story is datelined Sept. 12 2003.
I would think that a "blank line" pretty accurately reflects the intellectual property of SCO. This opinion is mine, and not my employer's.
First of all, unless the memory is embedded on chip, you've already lost-- slow bus speeds, limited pins. Second of all, once you have your on-chip cache, a single wordline might be very fast... if it's a kilobyte wide, then you can read 1024 bits every cycle no problem... but no cache has only a single wordline (unless it's a register), and no bitline has a single bit. So now you have a word address to decode, and a bitline with lots of bits on it and a sense amp at the end... and a bit address/mux if you want it. How does this affect latency? Well, that depends on the size of the memory not the speed of the memory cell. The bigger you make it, the longer the access. A single storage element is plenty fast... if the access isn't random, interleave intelligently and there's no problem. It's a size problem: you can't have every book in the library of congress at arm's length (physically). You've got your phone book, dictionary, and the last few books you read. The rest take a bit longer to obtain.
These days we make chips on silicon with copper wiring. Copper melts at 1064C, silicon melts at 1414C. Looks like finding an extremely high temperature conductor on par with copper to use for interconnect might be a problem the article completely misses.
Chip manufacturing is such a mature technology that just changing the interconnect from aluminum to copper took decades of work. Don't expect to see diamond substrate on real chips any time soon.
Japan is investing six million dollars in this research?? Come on! Intel is investing six billion in R&D just to figure out how to drive the next generations of silicon technology!
Great legal response to the patent infringement claim.
"But your honor, we didn't know we were infringing! Oh yeah, we didn't bother to read the nasty GPL thing either any more than we gave a hoot about IP until this March, but we didn't know the code we worked hard to develop, err, well, purchase was in nasty Linux... well, yes, the Unix code we developed, err, purchased is far older than IBM's patents... shit. You mean we have to care about other folk's intellectual property? This works both ways? I though we could just trash them without recourse. Well shucks."
Here's the patent
Prior art mentions graphical menus in MS Windows, so it seems the distinction here is allowing seamless navigation from one program's menus to another's, but I can't wade through all the legalese.
I second that e-motion. Bad pun, sorry.
About one computer per soldier, since total personnel in 4/2002 was 481,266. Given the typical astronomical support to combat troop ratio, that's not huge at all.
News.com story.
Of course, IBM is actually selling services for the most part... when the Army's microcrap starts deteriorating, Microsoft probably won't answer the phone.
IBM's own POWER4 dual-proc eServer: 7029-6E3-221E (Linux ready) POWER4+: 1.2GHz 2-way 8MB 2 x 36.4GB 2GB $8,995.00 IBM Web price* $237.74/mo. for 36 months**
I was under the impression that GM foods usually are modified versions of highly domesticated crops and animals that basically have no nautral version to compete with. Corn, wheat, cattle, etc. have been domesticated for thousands of years. They bear little resemblance to any "natural species" that might be eradicated by a GM version. GM seed creeping into a neighboring organic farm field is a different issue entirely than "natural" competiton. If GM dogs and housecats start competing in the wild with wolves and mountain lions, they're probably not going to make great pets. But yeah, species that aren't domesticated (Like fish) are a different issue.
IANAL = I am not a lawyer. Fragging = Killing someone with a fragmentation grenade, usually. Vietnam connotations. Try google next time.
The wing of the shuttle had a few hundred tons of thrust accelerating it upwards, while the piece of foam had lots of air friction and gravity accelerating it downwards. Who knows how many parameters you'd need to take account for to figure out the real impact velocity, or whether you just figure it out from the high-speed film of the wing strike, but I'm pretty sure if I throw something light out the window of my CAR driving HORIZONTALLY at 65 mph, the relative velocity changes pretty damn quickly.
I believe you somewhat blur the distincion between payload capability and "cargo" capability. It'd be neat if the Proton could launch combined payload of a 165,000-pound shuttle (yes, a ridiculously huge crew return vehicle, but still payload) and a 58,000lb "cargo," but I don't believe that there are any other current systems that can launch a few hundred tons into orbit. Yes, the fact that the shuttle reuses the three main engines blurs the payload/launch system boundary, but they are still reusable. Certainly the Soyuz rocket payload capability of about 12,000 pounds includes the crew capsule! You campare the shuttle launch system to the Saturn V. Good comparison... every shuttle launch (except the last) is like launching Skylab and getting it back again in two weeks. Yeah, I can see why you might consider than inefficient. ;) But Skylab only weighed 78,000 pounds, so maybe two Skylabs is a better number.
Yes, perhaps they promise less resistance than copper interconnect of the same size, but isn't a diameter of 100nm actually a bit large? Can nanotubes shrink, or is their diameter a chemical requirement? According to the International Technology Roadmap for Semiconductors, copper wiring pitch should now in 2003 already be 245nm. So with 50% spacing between those nanotubes, you're not even talking a 2x improvement in size over current interconnect. What if the things are too big to be used as interconnect for those 35nm gates we're supposed to see in 2007?
This article posted today at Silicon Strategies is probably just a rehashed press release pointing out that Motorola has made some nifty new nano-crystal flash with its 90-nm process technology. But if anyone's interested in it, it's those of you who've read this far.