New Optical Chip Claims 8 Trillion Operations/sec.

Richard Finney writes "Lenslet is announcing the 'World's First Commercial Optical Processor.'. Reuters has the story here. The Inquirer has a cool graphic here on it. The processor is specified to run at a speed of 8 Tera (8,000 Giga) operations per second, one thousand times faster than any known DSP. When Lenslet releases its Enlight processor in a matter of weeks, a unit using the technology will be 1.7 centimetres high and measure 15 by 15 centimetres."

Comments

what a ripoff!

[proj_2501]

It can't handle 8192 Giga Operations per second?

Re:what a ripoff!

[Directrix1]

I was almost impressed by this, until I read up on the technology on their website. It will have a pretty limited use as it only has 8-bit precision vector/matrix MAC which is where the 8 teraflops come from. This will be fine and all for just video but it isn't much of a quantum leap for anything else (besides having an optical core). I mean it has power, but there are other chips out there that do more with greater precision numbers.

Picture

[r_glen]

There's a nice picture of the processor here

Re:Picture

[r_glen]

Also, a demo video here

FYI

[Doesn't_Comment_Code]

Interstingly, optical processors aren't faster because light is faster than electricity. They are faster because they have much faster rise and fall times between digital on and digital off.

Re:FYI

[Detritus]

Really, I thought it was the cross-polarized emission of tachyons in a Potrezebie field.

Re:FYI

[QuantumFTL]

Interstingly, optical processors aren't faster because light is faster than electricity. They are faster because they have much faster rise and fall times between digital on and digital off.

While your statement may indeed be correct, that is not why this chip is faster. The reason is that they are doing analog signal processing using the physics of various optical elements to perform computationally intensive mathematics.

Think of it this way: We can use large, expensive mathematical operations to simulate optical components, which means we can also do the reverse - using optical components to perform the expensive mathematical operations.

I read about this about 2 years ago, and it was really quite fascinating to me. It turns out that with a simple lense, you can compute a fourier transform just by focussing the light (it doesn't focus down to an infinitesimally small point).

I managed to find an article about this, hopefully it should be apparent why this chip doesn't run quake:
Check it out here.

They are certainly not the only people doing this. I've seen plenty of references of this being used in missile guidance systems (turns out a simple fourier transform trick can be used to track objects in a camera). Someone I met while working at the Jet Propulsion Lab was working on this Optical Signal Processors. They prove to be very big in the next 10 years.

Cheers,
Justin

Re:FYI

[Doesn't_Comment_Code]

I studied this subject in depth and happen to know 6 six physics professors who agree with me on the subject. And I don't agree with a thing you've said.

To do anything at all with light, you need a material in with light beams can interact.
Light will interact in almost any medium. Many kinds of optical gates have already been created.

In this material, the speed at which electrons can change energy levels determine the speed.
I'm not even sure what you are talking about here. There are no signal carrying electrons in an optical fiber - that's the point. And if you meant photons instead of electrons, then the photons aren't changing energy levels. If you look back through your physics book, that would corrospond to changing color, and has nothing to do with optical computing. It is the absence or presence of light that determines the on or off state. Not the voltage, as in a regular processor.

In fact: the rise and fall time are determined by how fast you can (electronically) switch the light source on or off.
If you are using an isolated optical gate with electronic converters surrounding it yes. But that would be senseless and no one does it. Everyhing inside an opticl processor is connected by light signals. Each optical gate interacts with other optical gates optically. And as time goes on, the memory and bus of these systems will also become optical. Already, there are many physical processes that do not need any optical-electronic conversion : especially cpu bound operations that fit into the cache.

I don't beleive you have accurately grasped the concept of optical computing. If you have questions, please ask them. But don't assert your opinion as fact.

Re:FYI

[QuantumFTL]

They used to use regular electronic circuits to solve differential equations and similar problems too. They didn't get an exact solution, but they got a usable value. I think that's what you're talking about here.

You're talking about old analog electronic computers... yeah those weren't very precise (one of the reasons they are no longer used).

What I'm talking about is a little different. Those electronic ciruits would solve differential equations in the time domain (requiring a bit of time to compute) whereas these optical processors process information in the frequency domain (almost instantly, the bottlneck is as you say how fast they can moduate the light from an electronic signal).

Frequency domain computing is fundamentally different from the time domain computing in that in time domain analog computers, tiny errors accumulate very rapidly. For instance, an operational amplifier that is used to perform an integration will have a small bias current which will slowly charge the integrating capactor(s), requiring the integration to be rezeroed every so often (at least every few seconds, if not many times a second). In frequency domain computation, the error is not accumulative like that. There is error, and it does add up, but its pretty much orthogonal (the error is spread throughout the frequency space, rather than adding up towards the end of the time space in a time domain computer).

A really great article I found (this is the one I originally read back in 2001) is here. Anyone interested in the more technical side of the processor should read it. It explains why the processing is so fast (because it's essentially parallel rather than serial, along with being based on photons rather than electrons).

That's where I got most of my information from, along with my optics and mathematical physics classes :)

Cheers,
Justin
Disclaimer: I'm still a semester away from my BS in physics

Re:This is the Future

[John Hansen]

Optical processors have incredible potential. And if you think that's good, just wait. The combo of an optical processor with optical memory is a one-two punch.

But if you want to get the full speed out of your processor and memory, as I recall, all the buses must be optical as well.

Otherwise you're limited by silicon and PCB boards again...

Gotta Love the spin

[nsingapu]

"Processing at the speed of light, you can have safer airports"
Its really quite sick and disturbing that the aftermath of 9/11 has degraded to a marketing ploy.

Re:Cool Acronym needed

[bcolflesh]

Prismatic
Optical
Refractive
Nacell

Ha!

[ArmenTanzarian]

I patented this idea already, give me money!
My patent states:

Using light to do stuff and/or calculate stuff.

It's all there in black and white.

Memory is irrelevant for this kind of "processor"

[wowbagger]

This is NOT a Harvard architecture part - this isn't fetching instructions from RAM and executing them, like a regular DSP would.

Think of this more like an FPGA - you have a device that is configured for a specific processing algorithm, and data is fed in at wire rate and processed at wire rate.

An example of how a device like this might be used may be in order:
I'm trying to find a radar pulse buried in the noise coming in from my receiver. I want to know the phase delay of the radar pulse - how long from when I sent it till I got it back.

Now, I know what my radar pulse looks like as it goes out. I know that any reflection is going to consist of versions of that pulse shape, delayed and of varying strengths. So what I do is called a correlation - the easiest way to think of this is to imagine having 2 transparencies, one of my outgoing pulse, and one of the incoming signal. Now, I hold them up to the light, and slide the incoming signal across the reference pulse until things match up - that's the point of maximum correlation, and that give me the delay of the signal.

A real correlation function is a bit more complicated as you have to allow for the signal level to be changed - if I am looking for a signal of N samples in a received data stream of M samples, I have to do M*N multiply and add operations to get my correlation. Now, for a radar signal I might be sampling at over a billion samples a second, and looking for a chirp of a 100 ns would give me over 100 billion MAC operations a second. There are ways to do that with conventional DSPs, but they are a galloping BITCH to do (you basically make a cluster of DSPs, and each DSP takes a part of the signal. Synchronising that is a bitch.)

This device would work by having the shape of the outbound pulse represented in the structure of the device itself, and the MACs are done by taking the incoming data stream and projecting it on the structure - thus you do all your processing in parallel, and at wire speed. You get a pulse out when the incoming signal matched the signal you ar looking for.

But think of the SETI@Home score... :)

[Opiuman]

Seriously though, basically this chip can do very quickly what the SETI@Home software does on PCs. Fast fourier transforms and the like... Think about completing a calculation unit every 30 seconds instead of 8 hours and 40 minutes. That is the ball park. I wonder if the precision will be the same.