Hasta La Gadget!

A new chip could improve error correction in flash memory, and might also lead to more efficient spam filtering and shopping recommendations.

Lyric Semiconductor, a small MIT spinoff, has created an error correction chip that uses a technique called “probability processing” to guess the right answer or solve a problem.

The chip, called LEC, is 30 times smaller in size than current digital error correction technology. That means manufacturers can create higher density chips that offer more storage at lower costs.

“This is not digital computing in the traditional sense,” says Ben Vigoda, founder of Lyric Semiconductor. “We are looking at processing where the values can be between a zero and a one.”

Error rates in flash-based storage are of concern to both consumers and manufacturers.

“The issue with flash is you get higher and higher bit errors as you move to smaller geometry,” says Greg Wong, an analyst with research firm Forward Insights, “so to discern data that is in there you have to use probability type of algorithms.”

Today, one in every thousand bits stored in a flash memory comes out wrong when the memory is read. With the next generation of flash memory, the number of errors is expected to approach one wrong bit out of every hundred.

For consumers, this  means a music file that they play from their flash storage disk could sound wrong — or a file could get corrupted. To avoid that, flash memory makers have to use error correction, much of which is currently done using software algorithms.

The problem with software-based solutions is that they use digital signal processing circuits that add to the size of the chip, says Wong.

“This is an area where cost is a very sensitive factor,” he says. “So if you can reduce the size of the circuitry, there’s a big benefit there.”

Despite its tiny size, the Lyric LEC contains “a Pentium’s worth of computation,” says Vigoda.

Story continues.

Updated at 18:30 EST to correct timeline of prediction to 2030 from 2020

Reverse-engineering the human brain so we can simulate it using computers may be just two decades away, says Ray Kurzweil, artificial intelligence expert and author of the best-selling book The Singularity is Near.

It would be the first step toward creating machines that are more powerful than the human brain. These supercomputers could be networked into a cloud computing architecture to amplify their processing capabilities. Meanwhile, algorithms that power them could get more intelligent. Together these could create the ultimate machine that can help us handle the challenges of the future, says Kurzweil.

This point where machines surpass human intelligence has been called the “singularity.” It’s a term that Kurzweil helped popularize through his book.

“The singular criticism of the singularity is that brain is too complicated, too magical and there’s something about its properties we can’t emulate,” Kurzweil told attendees at the Singularity Summit over the weekend. “But the exponential growth in technology is being applied to reverse-engineer the brain, arguably the most important project in history.”

For nearly a decade, neuroscientists, computer engineers and psychologists have been working to simulate the human brain so they can ultimately create a computing architecture based on how the mind works.

Reverse-engineering some aspects of hearing and speech has helped stimulate the development of artificial hearing and speech recognition, says Kurzweil. Being able to do that for the human brain could change our world significantly, he says.

The key to reverse-engineering the human brain lies in decoding and simulating the cerebral cortex — the seat of cognition. The human cortex has about 22 billion neurons and 220 trillion synapses.

A supercomputer capable of running a software simulation of the human brain doesn’t exist yet. Researchers would require a machine with a computational capacity of at least 36.8 petaflops and a memory capacity of 3.2 petabytes — a scale that supercomputer technology isn’t expected to hit for at least three years, according to IBM researcher Dharmendra Modha. Modha leads the cognitive computing project at IBM’s Almaden Research Center.

By next year, IBM’s ‘Sequoia’ supercomputer should be able to offer 20 petaflops per second peak performance, and an even more powerful machine will be likely in two to three years.

“Reverse-engineering the brain is being pursued in different ways,” says Kurzweil. “The objective is not necessarily to build a grand simulation — the real objective is to understand the principle of operation of the brain.”

Reverse engineering the human brain is within reach, agrees Terry Sejnowski, head of the computational neurobiology lab at the Salk Institute for Biological Studies.

Sejnowski says he agrees with Kurzweil’s assessment that about a million lines of code may be enough to simulate the human brain.

Here’s how that math works, Kurzweil explains: The design of the brain is in the genome. The human genome has three billion base pairs or six billion bits, which is about 800 million bytes before compression, he says. Eliminating redundancies and applying loss-less compression, that information can be compressed into about 50 million bytes, according to Kurzweil.

About half of that is the brain, which comes down to 25 million bytes, or a million lines of code.

But even a perfect simulation of the human brain or cortex won’t do anything unless it is infused with knowledge and trained, says Kurzweil.

“Our work on the brain and understanding the mind is at the cutting edge of the singularity,” he says.

See Also:

• Cognitive Computing Project Aims to Reverse-Engineer the Mind
• Darpa’s Simulated Cat Brain Project a ‘Scam’: Top Scientist
• Mouse Versus Supercomputer: No Contest
• Never Mind the Singularity, Here’s the Science
• Futurist Ray Kurzweil Pulls Out All the Stops (and Pills) to Live

Photo: A graphic overlay shows neural connections on a scan of IBM researcher Dharmendra Modha’s brain/IBM

You’re a klutz. Admit it. I bet you have soaked at least one gadget in your life, either a cellphone in the toilet, a glass of wine on an Amiga keyboard or even, if you are really not paying attention, an iPod Nano in the washing machine (full disclosure: I have done all of the above).

If you lived in Japan, you big butter-fingers, you could just drop the poor, bedraggled device into the Dryer Box, a gadget-saving cube currently finding its way into various Yodobashi Camera stores in Tokyo. After a half hour inside, even a fully soaked cellphone will be dry again.

Whether it works or not is a different matter: if water already shorted out the electronics, you’re out of luck. If you ripped the battery out fast enough, then you may fare better. If your phone or iPod or whatever is resurrected, there will be a ¥1,000, or $12 charge. Should it not rise from the dead, the session is free.

The specs are somewhat impenetrable, but it appears that the box is simply a big hair-dryer, blasting the drowned gadget with hot air to dry it out. The best alternative, should you not happen to be in Tokyo when you dunk your cellphone, is a long spell in a warm, dry place: I used a sunny window-ledge for my Nano.

Those who suggest putting it in a bag of rice are dead wrong. If rice could pull moisture out of the air in your phone, it would do the same for the air around it, and all the rice in the world would be a soggy mess. It isn’t, so don’t bother.

Dryer Box [JMC via Gizmag and Core77]

See Also:

• IPod Nano vs. Washing Machine
• IPod Nano vs. Washing Machine Update: iPod Wins!
• Don't Try This At Home

Follow us for real-time tech news: Charlie Sorrel and Gadget Lab on Twitter.

A new research breakthrough from Intel combines silicon chips and lasers to transmit data at 50 gigabits per second — and someday, maybe as fast as a terabit per second.

The 50-Gbps speed is enough to download an HD movie from iTunes, or up to 100 hours of digital music, in less than a second.

The technology, known as silicon photonics, can be used as a replacement for copper wires to connect components within computers, or between computers in data centers.

“The fundamental issue is that electronic signaling relying on copper wires is reaching its physical limits,” says Justin Rattner, chief technology officer for Intel, which announced the breakthrough Tuesday. “Photonics gives us the ability to move vast quantities of data across the room or planet at extremely high speeds and in a cost-effective manner.”

Photonics refers to the generation, modulation, switching and transmission of light, and can be done using lasers or light-emitting diodes.

Over the next two years, Intel hopes to perfect the technology by improving the efficiency of the lasers, as well as the packaging and assembly of the silicon chips and the manufacturing techniques needed to churn out millions of these modules.

“We have a good sense of the challenges here and what it takes to put all the components together, so we expect the technology to be widely deployed by the middle of the decade,” says Mario Paniccia, director of the Photonics technology lab at Intel.

Copper cables are the lifeblood of computing today. But they are limited by length because of the signal degradation that comes with using them over distances.

“At speeds of 10 Gbps and higher, it is difficult to move electrons fast enough and with enough signal strength to beat the tradeoffs,” says Rattner.

This limits the design of computers, forcing processors, memory and other components to be placed just inches from each other, says Intel. The alternative is to transmit data over optical fiber, but that is expensive and also limited.

“It’s not an issue if you are using only a few of them in an undersea cable,” says Rattner, speaking about optical fiber cables. “But if you want to have optics widespread, from consumers to supercomputers, the cost has to be taken down or it is not practical.”

That’s where integrated silicon photonics could come in. Using silicon-based chips and the same manufacturing process currently used for those chips, photonics modules could replace copper connections.

It could change how computers and data centers are designed in the future, says Intel. Earlier this year, the company showed its Light Peak technology that uses optics to deliver bandwidth of 10 Gbps and higher. Silicon-based photonics can go much higher, reaching tera-scale data rates, says Intel.

Here’s how the silicon photonics prototype works to achieve the 50-Gbps rate. Each module has a silicon transmitter and a receiver chip. The transmitter chip has four lasers whose light beams travel into an optical modulator. The modulator encodes data onto them at 12.5 Gbps. The four beams are then combined to output a total data rate of 50 Gbps.

The receiver chip at the other end of the link separates the four optical beams and directs them into photo detectors. The detectors convert the data back into electrical signals.

“In the labs, we ran this for 27 hours with no errors and transferred about a petabit of data,” says Paniccia. “And all this at room temperature with no fancy cooling.”

The silicon-based photonics chip could be used within a computer or to communicate from server to server in a data center. “If we are talking about CPU-to-memory connection, we would take our photonics chip and put it close to the CPU to bypass the copper interconnects,” says Paniccia. “For now we are not talking about integrating with the CPU.”

As the next step, Intel researchers are trying to increase the data rate by boosting the modulator speed and increasing the number of lasers per chip.

“If you increase the data rate of the modulator and put more than four lasers on a chip you can scale the whole thing,” says Paniccia. “The 50-Gbps rate is just the beginning.”

See Also:

• Intel Researchers Turn Counter Tops Into Touchscreens
• Intel Shows 48-Core Processor for Research
• Butler Robot Can Fetch Drinks, Snacks
• New Processor Will Feature 100 Cores
• Optics and Silicon Elope
• Chip the Light Fantastic

Photo: A 50Gbps Intel Photonics module/Intel