(Page 2 of 2)
"This would be very useful for mainstream computing applications," says Julian Miller, lecturer in electronics at the University of York in England, who has used FPGAs for evolutionary computing applications. Currently, for his purposes, FPGAs are simply too slow. "It's a huge problem," he says. Being able to reconfigure a chip within a single clock cycle would be a great benefit, he says.
ChaoLogix has gotten to the stage where it can create any kind of gate from a small circuit of about 30 transistors. This circuit is then repeated across the chip, which can be transformed into different arrangements of logic gates in a single clock cycle, says Ditto.
Despite having attracted the attention of both Intel and AMD, the technology is still in its early days, says Ditto. ChaoLogix is raising $2 million to produce a range of prototypes. But even if the company can gain only a tiny slice of the chip markets, it "will be huge," says Ditto.
Besides being extremely fast, the use of a single circuit has huge advantages over FPGAs. The way FPGAs are designed takes up a lot of silicon real estate and consumes a lot of resources. With ChaoLogix's chips, "you have one car in a smaller garage, and it can change between one hundred different car types," says Ditto.
It's not the first time anyone has tried to develop single clock cycle reconfigurable chips. "It is well-trodden ground," says Cantle. "Most of the companies that have tried have come and gone." One of the challenges lies in the software required to reconfigure the chip, says Mark Parsons, commercial director of the Edinburgh Parallel Computing Centre in Scotland, who is using FPGAs to make a supercomputer as part of joint industry and academic project. "They are still very difficult to program," he explains. Not only is it complex to design each configuration, but each software template describing the configuration takes up computational resources.
Others agree. The success of a reconfigurable chip does not depend only on what it can do, says André DeHon, assistant professor of computer science at the California Institute of Technology in Pasadena. If it proves to be too complex for most programmers, it may never get off the ground.
Krishna Palem thinks a little uncertainty in chips could extend battery life in mobile devices--and maybe the duration of Moore's Law, too.
Computer memory could soon earn the ultimate commercial validation: the cheap plastic knock-off.
Guest (Sigma)
Will this technology be difficult to program like the article says?
Guest (Bruce Considine)
Ah Ha!
A perceptive question without runaway hype!
We all know Lt. Cmdr. Data was actually Brent Spiner, but we still run away with the fairies (as the Brits say) ever seeing mirages of intelligence in the next clever technological development.
Robby the Robot of "Forbidden Planet" was smarter than a human on his fistful of mechanical relays. That must be some absolutely amazing code. At one time a rack of vacuum tubes was going to take over the planet. The newest technology is always assigned power that in hindsight it is completely incompetent to deliver.
Guest (Jason)
I'd think that the chaotic gate implementation (if it's meant to sort of replace the de-facto LUT flavored FPGAs of today) would be relatively simple to program.
You program FPGAs using a language called VHDL (some people use Verilog). VHDL is a hardware definition language where you abstract yourself away from the architecture of the chip you're targetting. so in the end you specify everything in terms of state machines and such.
So the short answer to your question would be, no..no I don't think this technology would be harder to program. No more harder than it is to program an FPGA.
However I'd think that dealing with these chaotic states and finicky gate behaviors would make timing closure a royal pain.
Guest (Sigma)
Could various materials be used in the future?
Could other materials be used like diamond-films, or does the process require CMOS to function?
Guest (Bill)
We are also looking at silicon lasers and other options on the research side. The only real requirment is that the system be chaotic and those are all too ubiquitous in nature and materials.
Guest (Sigma)
Sounds like a match for genetic algorithms.
Sounds like the chaos chips would be a perfect match for genetic algorithms and synthetic evolution. I could also see it being very useful for distributed computing. I wish you and your company much luck and a speedy entrance into the computer industry.
Guest (Siddharth)
If chips can be reconfigured into different arrangements through software, don't you think an internet-connected pc would invite malicious programs to manipulate the chip adversely?
Guest (Sigma)
but you will probably have some kind of "control" chip that searches for unknown hardware configurations.
Guest
Absolutely. These viruses would be horribly nasty. Instead of looking for logical loopholes that are on the OS level..you now can completely change the behavior of your hardware.
As in you could physically fry someone's router/PC/whatever by causing a bus contention.
Manufacturing in the United States is in trouble. That's bad news not just for the country's economy but for the future of innovation.
On Twitter
Become a Fan on Facebook
Subscribe to the Feed
Guest (kitk)
brain advance?
I have been expecting this sort of developement for some time--it really resembles the way organic brains and neural systems operate, and is an example of how circuit advacement, intended or not, mimics living organisms, and will continue to do so. This sort of computing could really lead to learning, self-repairing systems in time, that can tailor themselves to meet user requirements. Amazing!
Reply
Guest (Bill Ditto)
Thanks for the kind words of encouragement
I read your post and appreciate your comments. We are working hard to commercialize this and have some cool applications in the works (think search engine chips as one cool direction). I agree with your comment about biology, we have been working on getting neurons to do some interesting computations as well, just not as commercial right now.
Reply
Guest (up9start@charter.net)
The next basic step toward AI
Sounds like a step toward an electronic version of a "Positronic Brain". Make them small enough and Put enough of them together and we could have self aware AI. Caution is advised though unless you can figure out a way to build in Asimov's three laws!
David Warner
Reply