Slashdot Mirror


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."

315 comments

  1. what a ripoff! by proj_2501 · · Score: 5, Funny

    It can't handle 8192 Giga Operations per second?

    1. Re:what a ripoff! by Directrix1 · · Score: 4, Insightful

      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.

      --
      Occam's razor is the blind faith in the natural selection of least resistance and in universal oversimplification. -- EF
    2. Re:what a ripoff! by slimak · · Score: 2, Funny

      Talk about ripoff -- try downloading the white paper on .... 1.5 kbps.... wish they had their fast processor servin' up the data and not some non-optical 386.

    3. Re:what a ripoff! by gantrep · · Score: 1

      Processor speed != bandwidth.

    4. Re:what a ripoff! by asbestos_tophat · · Score: 1

      The DSP optical technology isn't even as impressive as the extremely similar optical switching system developed by lucent 3 years ago that supported multimode optical communications concurrently. But it did make my day, from an economic perspective it is highly unlikely that a company would publish their groundbreaking design before they could maximize their financial returns from its application in the computing industry. It sounds like 15-second elevator pitch from a dot.com burnout looking for investment or a buy-out exit strategy. ;)
      Most engineers develop a pragmatic perspective dealing with the numerous emerging technologies. The Optical array sensors switch time is perhaps more impressive than the design of the DSP itself. But the physical stability, life span, and core design seems like missing information from the press packages. I would like to see even a Pencil sketch on a coffee napkin describing the actual physical properties of the design instead of the performance curves (anyone done a patent search for this item yet?). It's a nice looking board, but I am very skeptical about the viability of the technology until they disclose a real picture of the active core technology (pictures from Star Trek don't count as rough drafts). ;) lol

    5. Re:what a ripoff! by Xaer0cool · · Score: 1

      Joke != serious

    6. Re:what a ripoff! by gantrep · · Score: 1

      Tired joke != Funny.

    7. Re:what a ripoff! by metalslinger · · Score: 1

      dumbArgument == slashdot

      --
      /. Heroics - 99.999%
  2. Beowulf? Yes! by Doom+Ihl'+Varia · · Score: 1, Funny

    I imagined a Beowulf cluster of these. The doctors say I will never recover my night vision.

    1. Re:Beowulf? Yes! by DrEldarion · · Score: 1

      Fortunately, many of us do not have a use for night vision in our imagination.

    2. Re:Beowulf? Yes! by Anonymous Coward · · Score: 0

      Another old joke:

      I, for one, welcome our new optical chip overlords.

    3. Re:Beowulf? Yes! by actaeon169 · · Score: 1

      In Soviet Russia, the new optical chip overlords welcome YOU!

    4. Re:Beowulf? Yes! by Anonymous Coward · · Score: 0

      I am not very bright, so I need night vision in my imagination so I can see it.

    5. Re:Beowulf? Yes! by metalslinger · · Score: 1

      lol, here we go again...

      --
      /. Heroics - 99.999%
  3. The Definitive Question Is: +1, Patriotic by Anonymous Coward · · Score: 1, Interesting

    But will it run Windoze?

    1. Re:The Definitive Question Is: +1, Patriotic by diverman · · Score: 0, Redundant

      Only for the next revision. The following Windows release will find a way to bloat itself to the point that even that processor starts to feel slow... just to browse the web. ;)

      -Alex

  4. This is the Future by Doesn't_Comment_Code · · Score: 1, Interesting

    I've done a lot of research on this. 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. This is definitely the future of computing.

    --

    Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    1. Re:This is the Future by John+Hansen · · Score: 4, Insightful

      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...

    2. Re:This is the Future by Bendebecker · · Score: 1

      Actually the future of computing is probably a hybrid computer - part electronic (or digital - whatever term you like) and part optical.

      --
      There's a growing sense that even if The Future comes,
      most of us won't be able to afford it.
      -- Lemmy
    3. Re:This is the Future by Doesn't_Comment_Code · · Score: 1

      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.

      Yes, as soon as you convert from optical to electronic or vise versa, you sacrifice some of the speed you gained. I don't know about the current state of long term optical storage (Optical Hard Drives) but nearly everything else can be made to be optical including the processor, ram, bus, peripheral interface ...

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    4. Re:This is the Future by back_pages · · Score: 3, Interesting
      For real, what's the point of 8 trillion operations per second when there's no existing memory to support it (of which I am aware)? So this chip runs REALLY REALLY FAST on code that's REALLY REALLY SMALL, and otherwise it's bottlenecked by the memory bus and memory speeds.

      I appreciate that it's a great demonstration of new technology, but maybe it's a little premature to call this a new commercial chip. It sounds to me like a demonstration of a research project or an exposition of things to come.

      It's quite possible that I'm completely ignorant about this, but to whom do they expect to sell the latest and greatest THREE ORDERS OF MAGNITUDE increase in memory bottleneck?

    5. Re:This is the Future by Anonymous Coward · · Score: 0

      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.

      All your bus are belong to us.

    6. Re:This is the Future by Karamchand · · Score: 1

      Isn't optical digital (in contrast to analog) as well?

    7. Re:This is the Future by stratjakt · · Score: 2, Informative

      It's a DSP, not a CPU.

      Stream comes in one end, chip does some magical shit to it, stream comes out the other end..

      Like your mp3 player - raw data goes in one end of the DSP, gets decoded into audio, then sent off to the D/A converter for your listening pleasure.

      So this can handle larger/faster streams and do more to them.

      --
      I don't need no instructions to know how to rock!!!!
    8. Re:This is the Future by Anonymous Coward · · Score: 1, Insightful

      What if you take some data, do some 50000 operations on it and then store the results?

      Memory is no longer a bottleneck, it all depends on what you are doing and this processor is most likely *very* specialized.

    9. Re:This is the Future by Acrimonious+Coward · · Score: 1

      A single purpose DSP based on a similar design could be used as a router or multiplexor on a high-speed network and may not require a relatively slow electrical memory bus at all.

    10. Re:This is the Future by wthynot · · Score: 1

      Just look at it this way: you won't have to worry about upgrading your processor for a long, Long, LONG time. ;)

    11. Re:This is the Future by QuantumFTL · · Score: 3, Informative
      For real, what's the point of 8 trillion operations per second when there's no existing memory to support it (of which I am aware)? So this chip runs REALLY REALLY FAST on code that's REALLY REALLY SMALL, and otherwise it's bottlenecked by the memory bus and memory speeds.

      I'm not quite sure you understand what this processor is, and how it works. This is *NOT* like a Pentium with a faster clockspeed. This is a signal processing chip which, rather than really executing code, uses a series of optical filters to do massively expensive mathematical operations. This has the following set of properties:
      1. Operations in parallel on massive amounts of data. This means it could have a parallel bus of some sort, which could considerably increase the throughput of the system.
      2. I believe the chip is equivilent to a terrahertz processor, but doesn't really operate that fast. It just happens to do calculations in one step that take thousands to millions of steps in normal processors.
      3. Many real time applicatoins merely need a reduction in latency, not an increase in throughput. If you have a missile guidance system (yes, people have been working on using this for that purpose) you want to get the analysis of incoming sensor data done as quick as possible for lightning fast reaction times. Very imprtant when your software either functions correctly and on time, or the hardware is destroyed.
      4. This type of technology could process analog electronic signals, which have a much higher throughput than the fastest digital signal bus.


      The original poster was right, this IS the future, at least for now :)

      Cheers,
      Justin

      Disclaimer: I'm one semester away from my bachelors in Physics
    12. Re:This is the Future by Anonymous Coward · · Score: 0

      You do seem rather ingnorant. The articles all clearly state that this would be useful in military applications, which by defualt do not have to rely on consumer-grade PC100 memory or other memory solutions. Besides, the ability to crunch numbers 3 orders of magnitude faster than today's processors simply means that the other parts will be forthcoming.

      Why is everyone on /. a 24/7 pessimist except when it comes to the newest Matrix movie, video game, or anime pr0n?

    13. Re:This is the Future by ozbon · · Score: 2, Informative

      I think the answer to that is the wonderfully quantum "It depends".

      Option 1 - Digital Optical - simply measures whether the optical source is on or off

      Option 2 - Analog Optical - measures whether the optical source is on or off, and also measures the intensity, say to 50 distinct levels.

      Option 3 - Pretty Optical - as well as on, off and intensity, also measures emitted colour as an extra set of data bits.

      I've no idea if Option 3 is practical/practicable, but all three sound pretty feasible, to say the least.

      --
      I say we take off and nuke it from orbit. It's the only way to be sure...
    14. Re:This is the Future by Anonymous Coward · · Score: 0

      I dunno maybe people who want to factorize 1024-bit numbers?

    15. Re:This is the Future by Karamchand · · Score: 2, Informative

      Option 1: Digital doesn't mean 0 or 1. Or on or off. That's be binary. Digital just says "in steps, not continuous". At least that's what most dictionaries say.

      Option 2: According to above mentioned definition that's still be digital - as long as the "50 distinct levels" are here. If measures the intensity continuously that'd be analog.

      Correct me if I am completely wrong but that's what most of the dictionaries and encyclopedias I used say..

    16. Re:This is the Future by back_pages · · Score: 1
      Thanks for the reply. Color me informed.

      I do understand the distinction between a DSP and CPU though it wasn't very evident in my comment. Even though this is a DSP, I'm left wondering what type of equipment is going to supply this thing with enough data to justify 8 trillion ops per second.

      Even if this thing takes 64 inputs at a time, isn't it still capable of taking inputs at 125 GHz? If it takes 256 bits of input we're all the way down to 31 GHz, assuming an ideal situation in real time. I did SCAN TFA ;)

      I'm not saying that I think this thing is dumb, I just don't see how they can call this a commercial application. Guiding missles isn't something I shop for at Walmart. It's surely a useful research tool (like I previously mentioned) and apparently useful for military applications, but I'm still left wondering what type of commercial application this thing would serve.

    17. Re:This is the Future by wa5ter · · Score: 1

      You'd be surprised how many military applications DO rely on consumer-grade PC100 memory ;)

    18. Re:This is the Future by Bendebecker · · Score: 2, Interesting

      What I meant was a computer that used both electricity and light. From what I understand, the first electric computer used electric relays like the ones used in phones (cheaper than vaccum tubes whch had only been invented the same years as the germans began building the machine) and was described as being an electronic computer. When ENIAC was built using vacuum tubes, it was described as being a digital computer. What ever word you want to use to designate that the computer functions using electricity and transistors instead of light is up to you at this point.

      --
      There's a growing sense that even if The Future comes,
      most of us won't be able to afford it.
      -- Lemmy
    19. Re:This is the Future by essreenim · · Score: 1

      It may very well be the fututre but the future is not now, just yet.

      The device in question (Enlight256) is a milestone
      because it is commercially availabe, compact, cheap, very fast.

      The 256 is for the number of photo detectors and photo sources.
      it may be designed to work at a ceiling of matrix size 245x256 per operation - mind bogglingly fast!.

      The key to the whole unit would seem to be the SLM (Spatial Light modulator).
      My interpretation is that it would be used to generate the relevant transforms, making it programmable.

      I think it sets a very good example of where we can go technologically and ..also a bad one ethically - allot of military alliances with this company.

      I'd really like to see at least an equivalent to what von Neuman did digitally reproduced for an optical machine (complete optical bus,memory, processor....). I t would surely have tremendous applications (by which I mean far more reaching than those listed on the web site!!!)

    20. Re:This is the Future by QuantumFTL · · Score: 2, Informative

      I'm not saying that I think this thing is dumb, I just don't see how they can call this a commercial application. Guiding missles isn't something I shop for at Walmart. It's surely a useful research tool (like I previously mentioned) and apparently useful for military applications, but I'm still left wondering what type of commercial application this thing would serve. If you'd like to read a great article that explains this, check out: EE Times Story from 2001.

      It mentions this is very useful for "fast Fourier transforms/inverse fast Fourier transforms (FFTs)/(IFFTs), discrete cosine transform (DCT), discrete Fourier transform (DFT), compression, vector-matrix multiplication, equalization and correlation". ALl of these are very useful for processing communications signals such as in cell phones or high-quality recievers (HDTV in rural areas?).

      It also mentions that the optical light is modulated at 10 GhZ frame rate (but remember, this is quite parallel processing) and no ordinary computer we have today could possibly crunch 10 billion FFTs in a second. So the terrahertz number is mostly them just stacking it against existing computers.

      Cheers,
      Justin

    21. Re:This is the Future by stfvon007 · · Score: 1

      The users are too slow. Get rid of humans and replace them with optical users.

      --
      All misspellings and grammatical errors in the above post are intentional and part of my artistic expression.
    22. Re:This is the Future by Mr+Z · · Score: 3, Informative

      You can build both digital and analog computers from analog components. Indeed, aside from "ideal switches", I'd argue that most circuit components are analog. The poster to which you responded is correct -- a system built around discrete levels selected from a continuous domain is considered digital. A system built around continuous levels is considered analog.

      So, if you build a computer around 4-level signaling, it'd still be digital. Each signal corresponds to a "digit," likely valued 0 thru 3. If you built a differential-equation analyser out of op-amps, resistors, capacitors and inductors to represent the derivatives/integrals in your system of equations, and operated it within the linear region of the op-amps, you'd have an analog computer. Those aren't very popular anymore, nor have they been for some time.

      The options you suggested a couple posts up are all variants of digital signaling. The first one is binary signaling. The second one is discrete amplitude modulation. The third one is multi-frequency discrete amplitude modulation.

      The real question is, can you build large numbers of sufficiently small gates for a given signaling method? Additionally, do those gates aggregate into computing structures we know how to use effectively?

      Related side topic: 3-valued logic (ternary computing) likely won't take off until its proponents have good answers for both questions. Something tells me the circuits and the gates are the smaller of the two problems. The change in programming paradigm is too big.

      --Joe
    23. Re:This is the Future by fshalor · · Score: 1

      So does my iBook G3. :)

      Anyways, this chip would be perfect for some on the fly visulization applications. It could also find it's way into voice processing (IE, recognition circits). Really, it's a step in the direction to allow our really good computers today the eyes and ears they need to go the next step toward being more useful. I mean, what good is a 3 GHz processor for what we're currently doing on a desktop. Now, the 3 GHz processor would be alot more useful if it could have something like this optical chip to preprocess spoken word and then do something with it.

      On the other hand, I can see climatology benifiting much from this. I mean, nowcasting is only so good. (short, 30-60 minute forcasts). Something like this could cut the edge off interpreting all the input from radar, IR, etc sources much more quickly.

      Again, besides the potential benifits in interpreting data, one of the currently limiting factors in climatology research is the availbility of extensive data sets to train statitctical models. These sets can be created from a realitively small code base, and something like this optical chip would be ideal to code the model into and have it generate the output. Again, 8 hours on a fast machine now could be done in a few minutes.

      Okay, I'm over it. :)

      --
      -=fshalor ::this post not spellchecked. move along::
    24. Re:This is the Future by essreenim · · Score: 1

      Thats pretty much how I see it too, but actually the Enlight256 is purely for DSP puposes right now. Ever used matlab? The you now that its really just working with a specific slice of the CPU - matrix multiplication allot. Takes ages. If use this it will take a fraction of a second to do massive vector/matrix calcs. It has applicatoins in being incorpoated in to mobile cellular switching devices..blah, blah,; clustering also, it just makes the solution to certain algorithms much more easy, taking far less time with less power!

    25. Re:This is the Future by Karamchand · · Score: 1

      That's what I meant, yes. Thank you for your clear, interesting and well-phrased posting!
      Regards

    26. Re:This is the Future by Mark+of+THE+CITY · · Score: 2, Informative

      The mainframe world has had optical interconnects for some time now; I saw Amdahl machines so arranged around 5 years ago or so.

      --
      The clearance system sounds logical. It is not. It is completely arbitrary. -- John Bolton
    27. Re:This is the Future by NCamero · · Score: 1

      Have ypou looked at the latest AOL coasters? The CD 's are FRIGGIN LASER BEAMS with out sharks attached.

      And the DVD's are not just evil CD-ROMS!

      These optical CPUS just need to get all the slow electronic transducers out of the way.

    28. Re:This is the Future by Anonymous Coward · · Score: 0

      MOTHER FUCKER!
      *blam!* *blam!* *blam!* *blam!* *blam!* *blam!* *blam!* *blam!* *click* *clatter *ka-chak* DIE! *blam!* *blam!* *blam!* *blam!* *blam!* *blam!* *blam!* *blam!*

  5. Yeah, but does it improve my pr0n? by diverman · · Score: 1, Funny

    can this new, fancy processor increase the speed at which my porn loads and is displayed??? ;)

    -Alex

    1. Re:Yeah, but does it improve my pr0n? by Anonymous Coward · · Score: 3, Funny

      Nope, you needn't worry. You'll still finish well before the computer does.

    2. Re:Yeah, but does it improve my pr0n? by Anonymous Coward · · Score: 0

      If you view it at 1 meter away from an NTSC-like monitor in the future you'll get it at 29 x 10^7 frames per second!

    3. Re:Yeah, but does it improve my pr0n? by Anonymous Coward · · Score: 0

      Excellent response :) Mod parent funny!

  6. Picture by r_glen · · Score: 5, Informative

    There's a nice picture of the processor here

    1. Re:Picture by r_glen · · Score: 5, Informative

      Also, a demo video here

  7. yeah..nice by the_2nd_coming · · Score: 1

    to bad it is the size of my hand.

    but, it would be nice to plop one of those n m computer when it is smaller.....I wonder what the heat production is on one of those suckers, probably very low.

    but, to really see the advantage of these suckers in computers, we need to have an all optical interconnect system, otherwise, we will end up with a really fast processor and the speed of todays computers because of the hard drive and system BUS.

    --



    I am the Alpha and the Omega-3
    1. Re:yeah..nice by kramer · · Score: 1

      to bad it is the size of my hand.

      It would be nice to plop one of those n m computer when it is smaller.....

      Yes, because there's no possible way anyone could fit a 6" x 6" x 1" piece of hardware in a desktop machine.

    2. Re:yeah..nice by FatAlb3rt · · Score: 1

      an increase in speed 3 ORDERS OF MAGNITUDE and your first comment is about its size?? compare it to the size of 1000 of today's.... and depending on the heat sink requirements (or lack thereof), it might be right in line with today's CPUs.

    3. Re:yeah..nice by DrEldarion · · Score: 2, Insightful

      Why does it matter how large it is, as long as it's not ridiculous? Think about how much empty space there is in your box right now, would a processor that's maybe two or three times the size of current processors REALLY be that big of a deal?

    4. Re:yeah..nice by machine+of+god · · Score: 3, Funny

      I don't care if it fills my room, and I have to sleep on top of it in the 30cm space between it and the ceiling. I'd still take it.

    5. Re:yeah..nice by Dr.Enormous · · Score: 1

      So it won't be good for laptops. Big deal.

      Besides, I could swear there's some trend in technology making things smaller as time goes by, but maybe that's just the drugs talking...

    6. Re:yeah..nice by segmond · · Score: 1

      Of course it matters how large it is. It is a DSP, DSP are far smaller than CPUs, so if it is large, you are going to get a CPU the size of a dictionary. Now how much eletricity does it use? How much heat does it generate? These are all things related to size. ... and cost of course... ._seg

      --
      ------ Curiosity killed the cat. {satisfaction brought it back | it didn't die ignorant | lack of it is killing mankind
    7. Re:yeah..nice by DeltaSigma · · Score: 1

      I see someone's never dealt with a Pentium II

    8. Re:yeah..nice by rsborg · · Score: 1
      Think about how much empty space there is in your box right now, would a processor that's maybe two or three times the size of current processors REALLY be that big of a deal?

      One word: Heat.

      Considering the bulky apparatus (heatsink, fan) around my *current* work CPU (a nice loud P4 2.4 Ghz), I'd had to see how big that would have to be to accomodate a CPU 2-3 times the size.

      --
      Make sure everyone's vote counts: Verified Voting
    9. Re:yeah..nice by Anonymous Coward · · Score: 0

      Power consumption being equal, a large object will dissipate more heat than a small object of the same general shape. If your CPU were 2-3 times the size but still using the same amount of power, you might not need a fan at all.

    10. Re:yeah..nice by ajr_trm · · Score: 1

      Now how much eletricity does it use? How much heat does it generate? These are all things related to size.

      This is optical chip and relation between size, power consumption and heat generated will be different than in traditional chips.

    11. Re:yeah..nice by Anonymous Coward · · Score: 0

      Take note: this CPU is much more efficient. Though I don't know the numbers, I suspect it's probably not a very hot proc.

  8. I'm not being funny... by Pingular · · Score: 2, Interesting

    would you be able to link this in a Beowulf-type manner?

    --

    When anger rises, think of the consequences.
    Confucius (551 BC - 479 BC)
    1. Re:I'm not being funny... by Doesn't_Comment_Code · · Score: 1

      would you be able to link this in a Beowulf-type manner?

      Yes. You can apply the black box theory to optical vs electronic processors. The internals are different, using different logic to form the gates. But the function and operation would be no different.

      The only exception would be how you communicate with the processor. If you interact with it optically, you have to alter the rest of the computer. Or you have to have an optical to electronic converter- which will cost you a little speed, but make it behave exactly like a regular cpu.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    2. Re:I'm not being funny... by Anonymous Coward · · Score: 0

      Right, you are being clueless!

  9. Now you know what this means.... by WFFS · · Score: 1

    No more jerky games of Desert Combat :D

  10. Cool Acronym needed by ePhil_One · · Score: 1, Funny

    Sorry, COP just isn't cutting it for me. We need a much cooler acronym for this thing...

    --
    You are in a maze of twisted little posts, all alike.
    1. Re:Cool Acronym needed by bcolflesh · · Score: 4, Funny

      Prismatic
      Optical
      Refractive
      Nacell

  11. Why the defense industry concentration? by ThogScully · · Score: 1

    It seems this processor can be useful in lots of applications if for no other reason because of its speed. Why does half the announcement on Lenslet have to talk about how this will revolutionize the defense industry and homeland security?
    -N

    --
    I've nothing to say here...
    1. Re:Why the defense industry concentration? by musikit · · Score: 0

      well when you think about it why is an Isreali company improving US homeland security?

    2. Re:Why the defense industry concentration? by Zakabog · · Score: 2, Interesting

      Ummm because the company is in Israel, a country that has to deal with terrorist type attacks on a daily basis? I thought the same question till I saw

      "...said Major-General (Ret.) Isaac Ben- Israel, former head of the R&D Directorate of the Israeli Ministry of Defense."

      What else is the former head of the R&D Directorate of the Israeli Ministry of Defense going to say about a new chip like this one?

    3. Re:Why the defense industry concentration? by oggelbe2001 · · Score: 0

      So the managers can manage this democracy better. ogg

    4. Re:Why the defense industry concentration? by thrillseeker · · Score: 1
      Why does half the announcement on Lenslet have to talk about how this will revolutionize the defense industry and homeland security?

      Because that's the two applications most likely be willing to quickly pay a pretty penny for this development - that might matter to an investor.

      Those two applications can put to immediate use very-high-speed logic. Real-time and ver-high-speed filtering of radar (and other) data allows discerning objects that are much smaller, or much farther away, or are much closer to the ground, etc. than current technology - if it's that much harder to hide a threat then it's easier to now apply (always) limited assets to better determine what is an actual threat and what is not, or perhaps which is the most immediate threat, etc.

    5. Re:Why the defense industry concentration? by Anonymous Coward · · Score: 0

      Been said the gov already has this sort of thing. Why wait a month to crack something when you can do it in a minute or so?

      A real breakthrough was when they could do realistic battlefield simulation in sub-real time. Then they could co-ordinate dynamically by simulating the battlefield.

    6. Re:Why the defense industry concentration? by Anonymous Coward · · Score: 0

      Ummm, hallo, how about "Cynical Ploy to take advantage of 'Homeland Security' focus for marketing and sales purposes"?

      I would say this new processor has as much to do with Homeland Security as a new brand of doughnut!

      Coffee Company Announces That New Jamaican Heights Coffee Taste Contributes to Homeland Security.
      today announced that their new coffee flavor, called "Jamaican Heights", makes a valuable contribution to airport security. The new coffee contains a chemical called caffeine, designed to help airport security workers stay awake and increase the likelihood of averting a security hazard. is proud to be making this contribution to the security of our homeland."

    7. Re:Why the defense industry concentration? by elton247 · · Score: 1

      If i could mod.. that was funny!

      Another thing,
      You know the two driving industries in technological innovation: Military (power) and Pron (sex).

      --
      How strange it is to be anything at all
    8. Re:Why the defense industry concentration? by linwoes · · Score: 1

      TWhile this could be exploitation, I don't think so. This chip could be fantastic for a homeland security dept. Putting on the Conspiricy Theory hat for a second now: I want to sift through every land line and cell phone call in the us (wait the world). Now I can store it, and process it and know of a potential problem tomorrow, but with this now I can know aminutes after the call you just ordered a pizza! Now that security. Seriouly, though this has important benefits to eavsdropping.

    9. Re:Why the defense industry concentration? by Ignis+Flatus · · Score: 1

      The defense industry is the one most likely to spend money on bleeding edge technology. Not only will the military buy it, but they will also fund R&D. It's a real sweet spot to be in, the government pays to bring your product to maturity, and you also get to keep the rights to it.

      I suspect the mention of homeland security is marketing hype. Do you really believe that getting on an airplane will ever be fast? Besides, I doubt that convolving high-resolution images in real-time is all that important for security.

    10. Re:Why the defense industry concentration? by Mattsson · · Score: 1

      Because that's where the funding is.
      Tell a paranoid goverment "with this you can build better weapons and monitor your citicens" and they'll pour money over you...

      But the article *did* say that the company was talking about various applications, so there are probably lots and lots of legitimate usages for this too. ^_^
      A chip like this would probably do wonders in medical imaging devices and such...

      --
      /.Mattsson - My native language is not English, so please don't whine over linguistic errors. (That's lame anyway...)
  12. FYI by Doesn't_Comment_Code · · Score: 5, Interesting

    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.

    --

    Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    1. Re:FYI by Detritus · · Score: 4, Funny

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

      --
      Mea navis aericumbens anguillis abundat
    2. Re:FYI by Anonymous Coward · · Score: 0

      Light and EMW have the same speed in vacum. In fact, they are the same thing. It's the medium that makes the difference.

    3. Re:FYI by mpeeters · · Score: 1

      What utter dung.

      To do anything at all with light, you need a material in with light beams can interact. In this material, the speed at which electrons can change energy levels determine the speed. Just a in "normal" processors.

      Rise and fall times have nothing to do with this. In fact: the rise and fall time are determined by how fast you can (electronically) switch the light source on or off.

      --
      Research is what I'm doing when I don't know what I'm doing.
    4. Re:FYI by Wolfier · · Score: 1

      I believe it also has something to do with the heat associated with electronic processors which should not exist at all on an optical.

    5. Re:FYI by Anonymous Coward · · Score: 0

      I am sorry but optical processors are faster because light is faster than eletricity, that's why it has faster rise and fall times. DUH.

    6. Re:FYI by QuantumFTL · · Score: 5, Interesting

      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

    7. Re:FYI by Doesn't_Comment_Code · · Score: 4, Informative

      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.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    8. Re:FYI by Anonymous Coward · · Score: 0

      That's partially true, but the (more) important aspect is that velocity of light in materials (3e8 [m/s] / refr index) is orders of magnitude larger than the eletrical carrier velocity (less than 1 m/s)

    9. Re:FYI by Doesn't_Comment_Code · · Score: 1

      Electricity, as you're talking about it, is an EM Wave and travels at the speed of light in a vaccuum. Both light and electricity are slower traveling throug mediums, and their speed is still comperable. But the point of my post was that the speed of either is not the factor that increases the speed of the processor. The rise and fall times have nothing to do with the speed either. It is the tightness of the signal packet that creates a fast rise and fall time.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    10. Re:FYI by Doesn't_Comment_Code · · Score: 1

      That's partially true, but the (more) important aspect is that velocity of light in materials (3e8 [m/s] / refr index) is orders of magnitude larger than the eletrical carrier velocity (less than 1 m/s)

      That is a common misconception, but not true. Both electrical and optical signals travel at the same speed, and both are slowed significantly by their medium. But that has minimal impact on the speed of a cpu. Most of the time wasted by an electronic cpu is wasted waiting for bits to rise and fall.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    11. Re:FYI by Doesn't_Comment_Code · · Score: 1

      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.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    12. Re:FYI by mpeeters · · Score: 2, Informative

      Just to make myself clear:

      - Light will interact in almost any medium. Many kinds of optical gates have already been created.

      Yes, light will interact; badly. Many kinds of _fully_ optical gates have been prototyped and proposed. Few are available commercially

      - There are no signal carrying electrons in an optical fiber.

      Duh, no - but weren't we talking about optical gates; two beams interacting? How do they interact then, if not through changing the electron populations in the material ?

      - Already, there are many physical processes that do not need any optical-electronic conversion

      Read my previous remark again: what is then the physical origin of the switching on/off in your proposed gates in the fully optical processor ?

      And to bounce back your initial snide remark: I too have studied this subject in depth (a M.Sc on the subject of optical computing and a PhD on the subject of VCSELs) and happen to know at least six physics professors who agree with me on the subject. And I don't agree with your tone, but am always interested in a stimulating discussing about points of view.

      Hem Hem. M.

      --
      Research is what I'm doing when I don't know what I'm doing.
    13. Re:FYI by Anonymous Coward · · Score: 0

      Alright, I've got a question. What limits the speed of these optical processors? Is it something like the bandwidth of the input signal, where the two sidebands of some input signal move at slightly different speeds through the same medium in the processor, or is it something like energy requirements, or even simply the speed of light and physical processor size?

    14. Re:FYI by Doesn't_Comment_Code · · Score: 1

      One more follow up about mediums:

      You mentioned interaction of light and changing energy levels of electrons. I beleive you are refering to, for instance, the time it takes for the electrons to change energy state in lithium niobate when it receives incident photons. First of all, this is not a point of contention between our viewpoints, as my argument concerns the latent time between signal set and acheiving threshold level. I originally mistook what you meant. But there is another option besides gates that use a photo-electrically sensitive substance such as lithium niobate. There are optical gates under development which use the diffraction of light to create OR and XOR gates. The gates are similar to the Young's single and double slit experiments. Oh how I wish slashdot had a way to draw! Anyway, two light signals, separated by a small angle, are both sent to a slit in a screen. Holes in a screen on the other side of the first screen are then optical outputs that return OR or XOR depending on their distance from the center of the screen. These gates can then be combined to create all other logical gates.

      The only problem with these gates is amplification. Since the gates are passive, they will eventually lose strength. In which case an active medium will again be required to pump up th e signal.

      But there is at least a start on purely optical gates that don't depend on electron energies.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    15. Re:FYI by Doesn't_Comment_Code · · Score: 1

      Alright, I've got a question. What limits the speed of these optical processors? Is it something like the bandwidth of the input signal, where the two sidebands of some input signal move at slightly different speeds through the same medium in the processor, or is it something like energy requirements, or even simply the speed of light and physical processor size?

      In practice, it is the much slower electronic components that limit the speed of an optical processor. Optical memory and bus would fix much of that though.

      At a theoretic level, pretty much everything else you said is a limiting factor. While optical signals rise very quickly, they aren't instantaneous. The signal has to reach the output (speed of light through medium). And the gates aren't as small as silicon gates yet. Also, the photonic signals currently have to be amplified by electronic devices.

      All of these things affect the speed. And none of them are a huge bottleneck save the interface between optical and electronic computer.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    16. Re:FYI by Doesn't_Comment_Code · · Score: 1

      OH MAN!!! I just wrote a huge long post and isn't showing up!

      I don't have time or the patience to retype it.

      So I'll summarize.

      Sorry for sounding rude. I thought you were being rude so I responded in kind.

      I think we're talking about different things: My point has to do with the rise and fall times of electronic and photonic signal packets. Since most of the time wasted by a cpu is waiting for signals to reach threshold levels, optical computers can have a much faster clock and not sacrifice stability or energy requirements.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    17. Re:FYI by QuantumFTL · · Score: 4, Informative

      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

    18. Re:FYI by PaulBu · · Score: 1

      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.

      Actually the interesting question is not why optical processors are "fast" but why semiconductor processors are "slow"! Individual transistor on/off switch time can be as small as several picoseconds, but we are not seeing sub-THz clock frequencies because CMOS speed is limited by RC time constant of the (critical path) wires.

      OTOH, as my advisor said long time ago, "You do not need light to get speed of light", if you use well-matched low-loss transmission lines (preferably superconductive ;-) ) to move information around the chip you will get closer to being limited by speed of light. (Trust me, this is what I do for living! ;-) )

      One big advantage of the proposed optical architecture is that it utilises the broadcasting ability of light in free space, something which can not be done with point-to-point transmission lines.

      Paul B.

    19. Re:FYI by Anonymous Coward · · Score: 0

      Light IS EMW.

    20. Re:FYI by twiddlingbits · · Score: 1

      Interesting posts, I understand most of even though I'm just a BS CS with 20 yrs of software engineering experience. I think it would be an interesting thread to explain to us /.ers how "optical memory" works. D we trap a photon(s)and say that "bit" is on or we somehow register the presence of the photon(s) being there as "on"? I assume we release the photon and the bit is dark for a zero? How in the world do you trap a photon? It has no magnetic charge, and it is almost massless. Or is optical memory somehow optical Inputs/outputs to very fast conventional semiconductor memory?

    21. Re:FYI by Doesn't_Comment_Code · · Score: 1

      The last I checked in to optical memory it was implimented with tiny drops of ink. The ink could be made to block or pass light based on a light pulse from one direction, while the signal was read from another direction.

      So to store 8 bits of data, you would shine individual lights on each of those eight bits that are not to be set from the proper angle, and the ink dot would become opaque from the read angle.

      Then, during the read phase, a light is shined through each ink dot. All the darkened ink dots will block bits that are off, and on bits will be allowed to shine through.

      The speed on this memory wouldn't be as fast as the cpu. Probably a lot like today's memory bottleneck. But it should outperform electronic vs optical conversions when it is a mature technology.


      Warning. I read about this several years ago, and it was a very young technology. It may have changed significantly, or been replaced something better. Optical memory in itself is a new concept that hasn't had a lot real world success. But it appears to have potential. As with all things: it first has to happen under lab conditions. Then a poor product will be available for 1 bajillion dollars. And eventually everyone will have one for the price of 10 cups of coffee.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    22. Re:FYI by Anonymous Coward · · Score: 0

      Intel/AMD had better be looking into something like this if they want to stay in business. Optical chips and fiberoptics could make electrical devices obsolete it sounds like. Microsoft will probably slow down the OS to today's standards though...

    23. Re:FYI by Glonoinha · · Score: 1

      Light IS EMW.
      And Light is Love.
      QED
      ElectroMagnetic Radiation is Love.

      Think about that before you replace your CRT with an LCD.

      As for the new chips, I bask here in the love emitting from my CRT envisioning this thing being applied to cranking up the performance of audio and video codecs to the point that DVD quality videos are compressed to the point that they fit several to a CD and they don't take all night to encode.

      Well that or way better missle guidance systems.

      --
      Glonoinha the MebiByte Slayer
    24. Re:FYI by mpeeters · · Score: 1

      OK - Read your 3 last comments (while on the train going home, unfortunately the wireless access is not ubiquitous - it is rather in the neighbourhood of inexistent) when I got home. That accounts for the latency ;-)

      1) The lithium-niobate Mach-Zehnder interferometers indeed work in a way similar to Young's double slit experiment, with the added functionality of a phase shift in one of the arms (aaah, drawing on Slashdot - indeed an utopic (?) concept) to switch the gate using a photo-electric material. The photo-electric effect, unfortunately, is due to a nonlinear index of refraction caused by a change of the electron band levels in the presence of a DC electric field. So again, these gates operate at "electronic" speeds

      2) "electronic" speeds are not necessarily nanoseconds. They can be in the order of picoseconds as well. My initial "utter dung" was related to my interpretation of the switch on/off speed. And yes, rereading it, it sounded snide as well. Unintentionally as well. Sorry as well.

      3) If we now look at "electronic" speeds in the order of nanoseconds, one effect comes into play which related to your remarks about the voltage swing and overclockers: the ridiculously low RC-timeconstant for light in vacuum/air/optical fiber. This of course is the direct reason for the steep flanks on the clocking pulses your are referring to. A really good article (which, if you haven't, you should read) about this subject (the advantages of optical computing) was written by D. Miller. If you want, I can send you the PDF.

      Yours,
      M.

      --
      Research is what I'm doing when I don't know what I'm doing.
    25. Re:FYI by Anonymous Coward · · Score: 0

      One approach is to use something like optical delay lines. A bit stream of light pulses is sent into a loop (optical fiber, perhaps). When light reaches the end of the loop, it's right back at the beginning, where the light is sensed by an active component and retransmitted all over again.

      This gets difficult at high speeds for several reasons and isn't strictly random-access (bits can only be accessed once per circulation.) Pulse widths would need to be vanishingly short to get hard-drive scale data density in a reasonably sized form factor.

      This probably isn't what you were thinking of, and indeed requires an active electronic component. It's not exactly a new idea either, it's been done before with other types of waves besides light. Just felt like throwing out an idea.

    26. Re:FYI by Doesn't_Comment_Code · · Score: 1

      I think, based on your comments, we have no disagreements, only previous misunderstandings.

      I would be interested in reading the PDF. But I don't want to post my email permenantly to slashdot. Is it available via ftp or the web? Also, I think my address is temporarily written in my journal.

      A thought has crossed my mind. But I'm not sure if it's a good idea, or something other people have thought of and dismissed because its a bad idea. Perhaps you might be one of the best people to bounce this question off of.

      Regarding amplifying a light signal so as to use passive OR and XOR gates: Think back to plain old gas lasers. What if a tube full of He were held at an excited state, so that there was a population inversion - but no induced emission of photons. In such a tube an incoming laser signal would trigger a huge induced light beam, amplifying the signal. The power supply would indeed be electric. But the response time would be amazingly fast. Is this similar to how lithium-niobate works? Is lithium-niobate the solid state version of what I just described? I admit I don't know much about this medium, only having read about it briefly. Back to my point, with the aid of such an amplifier, a gate could be made completely passive with respect to the path of the light signal. It would require only a voltage to maintain the population inversion. Thus, a light signal could pass through a slit and be XORed or ORed with another signal and amplified in the process with virtually no delay or deviation. The signal could then continue to the next gate. Perhaps this is compromising good design for speed. On top of that, I'm sure I have overlooked some very important details. But it struck me as an idea at least worth examining.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    27. Re:FYI by jburon · · Score: 1
      Wouldn't fiber amplifiers solve the problem of electron-limited amplification? The size of these things are a problem, but it seems to me that they are still more practical then, say, gas lasers.

      As you surely know (just stating this as "general info") Erbium doped fiber amplifiers are a special kind of fiber, where erbium atoms mediate transfer of energy from an optical "pump" signal to the signal that needs amplification. This is a purely optical operation, save for the state change in the erbium atoms. But shouldn't this provide as fast response as the gas laser proposed in another thread of the discussion?

    28. Re:FYI by Doesn't_Comment_Code · · Score: 1

      What you're describing sounds very similar to the idea I proposed. And yes, I think it would work just as well. And the size shouldn't be an issue. Unless there is some theoretic limit to the size, companies will put a lot of money into research to make smaller and smaller ones. Just think of our transition from tubes to big transistors to tiny transistors to millions on a pin-prick sized transistors. If the application is there, the product will become better and smaller through the natural course of things.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
  13. Great name! by Anonymous Coward · · Score: 0

    The company is named Lenslet. In Dutch a lens is the same as English lens. But slet is our way of writing slut. Which creates... lens slut. What a nice name for a company creating optical things!

  14. Article by Anonymous Coward · · Score: 2, Informative

    Israeli Processor Computes at Speed of Light
    Wed October 29, 2003 05:03 AM ET

    By Tova Cohen

    HERZLIYA, Israel (Reuters) - An Israeli start-up has developed a processor that uses optics instead of silicon, enabling it to compute at the speed of light, the company said.

    Lenslet said its processor will enable new capabilities in homeland security and military, multimedia and communications applications.

    "Optical processing is a strategic competitive advantage for nations and companies," said Avner Halperin, vice president for business development at Lenslet.

    "Processing at the speed of light, you can have safer airports, autonomous military systems, high-definition multimedia broadcast systems and advanced next-generation communications systems."

    An optical processor is a digital signal processor (DSP) with an optical accelerator attached to it that enables it to perform functions at very high speeds.

    "It is an acceleration of 20 years in the development of digital hardware," Lenslet founder and Chief Executive Officer Aviram Sariel told Reuters.

    The processor performs 8 trillion operations per second, equivalent to a super-computer and 1,000 times faster than standard processors, with 256 lasers performing computations at light speed.

    It is geared toward such applications as high resolution radar, electronic warfare, luggage screening at airports, video compression, weather forecasting and cellular base stations.

    Lenslet said its Enlight processor, unveiled at the MILCOM exhibition in Boston this month, is the first commercially available optical DSP.

    "Optics is the future of every information device," said Sariel.

    Jim Tully, vice president and chief of research for semiconductors and emerging technologies at Gartner Inc, said most companies working with optics focus on switching optical signals for telecommunications rather than processing information optically.

    "I'm not aware of any company that has taken it to the extent of processing optically," he said.

    Lenslet has raised $27.5 million so far from such investors as Goldman Sachs, Walden VC, Germany's Star Ventures and Chicago-based JK&B Capital.

    PALM PILOT SIZE

    The company's prototype is fairly large and bulky but when Lenslet begins to supply the processor in a few months it will be shrunk to 15 x 15 cm with a height of 1.7 cm, roughly the size of a Palm Pilot.

    "In five years we plan to shrink it to a single chip," project manager Asaf Schlezinger said.

    Tully said one issue is whether this technology can be produced in volume the way silicon chips are made.

    "Because semiconductor manufacturing technology is well developed, you can produce millions at quite low cost," said Tully, who is not familiar with Enlight.

    Lenslet said its processor will be competitive in price with a multi DSP board.

    Sariel is negotiating joint projects with companies and/or government agencies in the United States, Europe and Japan to produce the processor for specific applications. It already has projects signed with Israel's Defense Ministry.

    "We don't rule out licensing our technology to others," Sariel said. "We are looking at a virtual production line where production is done by others and we provide testing equipment."

    Tully said semiconductor companies are working on technology that would use optical channels inside a chip to allow very high speed communication from one part of a chip to another.

    "It's conceivable this technology could become mainstream inside chips in 10 years time," Tully said.

    (C) Copyright Reuters 2003. All rights reserved. Any copying, re-publication or re-distribution of Reuters content or of any content used on this site, including by framing or similar means, is expressly prohibited without prior written consent of Reuters.

    1. Re:Article by PaulGrimshaw · · Score: 1

      "(C) Copyright Reuters 2003. All rights reserved. Any copying, re-publication or re-distribution of Reuters content or of any content used on this site, including by framing or similar means, is expressly prohibited without prior written consent of Reuters."

      Eek!

  15. User availability... by Beatbyte · · Score: 3, Insightful

    This innovative new product will enable revolutionary, new applications in the fields of defense, homeland security, multimedia and communications. The exhibition being held at The World Trade Center, continues until October 15th, 2003.

    The fact that...
    1. its at the WTC
    2. they mention defense and homeland security
    3. its immensely powerful

    ...makes me question whether or not this is going to be available to end users.

    besides the lack of a huge marketing campaign.

    Anyone know anything different?

    1. Re:User availability... by Otter · · Score: 1

      Note that the "World Trade Center" here is the Boston World Trade Center. They're not exhibiting it at the hole in the ground in New York.

    2. Re:User availability... by twiddlingbits · · Score: 1

      Perhaps the fact that the thing is so darn expensive that only defense projects can justify the cost? I can put together an array of "mainstream" DSPs for which I already have development tools and equal the performance of the optical chip at a much lower cost. It still needs to get smaller to be used in a lot of applications. I also wonder about how much power it needs and how much heat it dissipates. This thing will also be limited in performance in the real world by the fact that memory is much slower than the CPU.

    3. Re:User availability... by stratjakt · · Score: 2, Interesting

      First off, this thing is going to cost a bazillion dollars, and will be a "commercial beta" product for years.. They'll sell them, but theres still a ton of work to improve the thing.

      So who would buy one? Someone with lots of cash - the DoD.. You wanna sell to the DoD, you have to show how it will fight terrorism.. It's just the way things are selling to the government.

      I work writing code for public safety systems - records and dispatching for cops. We did a search and replace on all our marketing literature, replacing "gang member" with "terrorist".

      I don't like seing "terrorist" being morphed into the buzzword du jour, it lessens its value to the point the word will be meaningless soon enough. Like "racist". The word "racist" has been thrown around so much that it doesnt mean anything. Bush is a racist, Clinton was a racist, Jesse Jackson is racist, Phil down the street is racist, N'Sync - all racists. Bleah..

      But thats just how politics work.

      --
      I don't need no instructions to know how to rock!!!!
    4. Re:User availability... by Feyr · · Score: 1

      which remind me, while in montreal one day, some tourist asked where was the world trade center. at first we immediatly thought "man this guy's got the wrong city..". but there really IS a wtc in montreal too, it's usually refereed by another name tho (centre de commerce international, french for WTC)

    5. Re:User availability... by DAldredge · · Score: 1

      There is also one in DFW.

    6. Re:User availability... by Minna+Kirai · · Score: 1

      Interestingly, many cities have "World Trade Centers". And unlike the one in New York, they are attached to international ports and used for trade. "Office space for financial services" is not "world trade".

      The NYC WTC was fraudulently named. Really, the whole construction of those two buildings was borderline illegal. Or at least pinko anti-American...

    7. Re:User availability... by Xerithane · · Score: 1

      There is a WTC in Portland, OR and Seattle, WA. In fact, there is almost always a WTC near you.

      --
      Dacels Jewelers can't be trusted.
    8. Re:User availability... by Anonymous Coward · · Score: 0

      There are WTC's all over the world.. and interestingly.. they're all higher than NY one:P

    9. Re:User availability... by aminorex · · Score: 1

      there's also on in Shanghai. it is much
      better qualified for the name than the
      one in Quebec.

      --
      -I like my women like I like my tea: green-
  16. Gotta Love the spin by nsingapu · · Score: 5, Insightful

    "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.

    1. Re:Gotta Love the spin by Anonymous Coward · · Score: 0

      What if they'd said, "Processing at the speed of light, you can have safer space shuttles"

      How DARE they try to link their products to any given event in known reality?! Let's all stick to UT2k3 for our analogies from now on.. maybe that won't offend anyone.

    2. Re:Gotta Love the spin by segmond · · Score: 1

      what does processor speed have to do safer space shuttles? it was flawed process. no wonder you are a coward... shuttles are not running on 2Ghz cpus, go read up on them, they stick with quite old proven technologies, just like the car computers on the latest sports car are easily 16mhz.

      --
      ------ Curiosity killed the cat. {satisfaction brought it back | it didn't die ignorant | lack of it is killing mankind
    3. Re:Gotta Love the spin by mhifoe · · Score: 1
      There are lots of spurious references to 9/11 about.
      The UK satirical magazine 'Private Eye' even has a column about it called 'warballs'.

      Here's an example:
      The tragedy of 11 September has paradoxically led to a boom in the caravan industry, but has this benefited the consumer? -- Caravan Club Magazine.

    4. Re:Gotta Love the spin by tomzyk · · Score: 1
      "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.
      Well, Lenslet is an Israeli start-up and from what I've heard over the past several years, Israel has had some very strict policies in regards to their airports due to terrorist acts/threats. I really don't think that this has to do with 9/11 explicitly, but the world-wide (or at least Israeli) terrorist threat situation in general.

      I think they're just saying that you can do a lot of things more efficiently with faster computers... and having "safer airports, autonomous military systems, high-definition multimedia broadcast systems and advanced next-generation communications systems" are just a few examples.
      --
      Karma: NaN
    5. Re:Gotta Love the spin by Arker · · Score: 1

      "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

      It's also disturbing that they (and you) don't seem to realise that the electrical impulses in silicon chips also move at the speed of light.

      Then again, most everything marketing types write makes me want to spew. I presume if they ever say anything that's not deceptive they get banned from the industry for life or something.

      --
      =-=-=-=-=-=-=-=-=-=-=-=-=-=-
      Friends don't let friends enable ecmascript.
    6. Re:Gotta Love the spin by Anonymous Coward · · Score: 0

      They don't move at the speed of light.

    7. Re:Gotta Love the spin by dedalus2000 · · Score: 1

      Bush used it to sell his tax cut before the bodys were cold or the fires were out.

      --
      My keyboads not woking popely.
    8. Re:Gotta Love the spin by dedalus2000 · · Score: 1

      I wonder how resistant to radiation these things are. The reason the space shuttle is using older "proven" technology is due to it's exposure to ionizing radiation when a few electrons more or less at a gate can mean the difference between it switching on or off radiation has a big effect but the older processors require more electricity to cause them to switch so a few stray particles won't cause as much trouble. If these things are resistant to radiation they may well have a big application in the space program but not for the reasons given.

      --
      My keyboads not woking popely.
    9. Re:Gotta Love the spin by Javit · · Score: 1

      I don't know. I mean, the aftermath of Pearl Harbor degraded into a shitty overhyped love story starring Ben Affleck, which shows us we're really nowhere near rock bottom on this whole 9/11 thing.

      --
      Support NRA, America's oldest civil rights group.
    10. Re:Gotta Love the spin by Suidae · · Score: 1

      Additionally, its expensive to develop and test radiation hardened processors, so when you have one that works, you tend to stick with it. In 1998 Intel announced that a rad-hard pentium was to be developed and licensed for government use, but I never heard anything new about it. I wonder what happened to that project.

      As of 5 years ago Intel had produced about 5 different rad-hard processors for space use. I suppose other manufacturers have their own offerings too.

    11. Re:Gotta Love the spin by Suidae · · Score: 2, Informative

      Whats really sad is that you automaticly assume that 'having safer airports' refers to terrorism, as if that is the only thing that makes airports dangerous.

    12. Re:Gotta Love the spin by Anonymous Coward · · Score: 0

      Israel was broadly condemned by the UN in August 2001, a UN resolution was passed forcing it to withdraw from its occupied territories, threatening worldwide sanctions. This was quickly followed by September 11, a curiously well-planned and executed terrorist attack on the US, causing a massive shift in worldwide opinion, and the invasion of two of Israel's most prominent opponents... Coincidence?

      Its quite proper that Israeli firms use 9/11 as a competetive marketing ploy heh.

    13. Re:Gotta Love the spin by Nordberg · · Score: 1

      Actually it's an Isreali company, they were worrying about airport security long before 9/11.

      --
      *Splort*
    14. Re:Gotta Love the spin by Surt · · Score: 1

      Umm ... they don't move at anything close to the speed of light in vacuum.

      --
      "Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
    15. Re:Gotta Love the spin by aminorex · · Score: 1

      I took it to mean that if you buy their optical processors, they won't bomb your airports.
      Also an effective marketing ploy.

      --
      -I like my women like I like my tea: green-
    16. Re:Gotta Love the spin by Anonymous Coward · · Score: 0

      I agree! What's sick and disturbing was how the entire economy changed, post 9/11, into the flag selling industry.

      This is known as the military industrial complex. Anyone who agrees with the climate of nationalism that followed 9/11 was caught up in this psycheological disorder and should be punished by psycho therapy and drugs like ritalin for the rest of their stupid weak natural lives!

      Or until they prove to the rest of us that they're not really as gay as we think they are.

    17. Re:Gotta Love the spin by replicant108 · · Score: 1

      Degraded?

  17. A long time ahead in a galaxy far far beyond by Althazzar · · Score: 1

    From the Reuters story: Tully said one issue is whether this technology can be produced in volume the way silicon chips are made. "Because semiconductor manufacturing technology is well developed, you can produce millions at quite low cost," said Tully, who is not familiar with Enlight. The question is thus how long it will take before they can produce this in great quantities, before that, it will be useful but not in our desktopsystems. Above that the prototype is large and we have yet to see that they can deliver to their promise of a small version. Other than that: i'd love to have it :).

    1. Re:A long time ahead in a galaxy far far beyond by Helter · · Score: 1

      If you read about it, it won't be useful for your desktop anyway.

      It's not a CPU.

  18. dsp vs processor by redtoade · · Score: 2, Insightful

    "the world's first commercial optical digital signal processor"

    When I read the lead post, I thought it was an actual processor like on a PC motherboard... not a DSP. These aren't the same things are they? The possible applications listed on the press release seem to be entirely communications oriented. (ie. fiber optics)

    Now a NAND gate using only optics (not electro-optics) would be fantastic. Maybe using some sort of wave interference to generate the logic table... and as you know you can build all of the other logic gates from a NAND!

    1. Re:dsp vs processor by Anonymous Coward · · Score: 2, Informative

      Generally speaking a DSP is just like any other processor but optimized for certain types of calculations, like fourier transforms and matrix multiplications.

      This thing probably has some very specialized optical processing elements that can do thousands of "ops" in parallell if your code can utilize it fully.

      Remember it's Tera-operations per second, not tera-instructions.

    2. Re:dsp vs processor by merlin_jim · · Score: 1

      and as you know you can build all of the other logic gates from a NAND!

      That's not entirely accurate... you also need a way to link the logic gates together. And a non-destructive way of sampling the signal. (You can only build every logic gate from an NAND if you can hook the NAND's inputs together to create a NOT, meaning one signal, two sinks)

      And if you have an OR gate, the process is MUCH easier...

      --
      I am disrespectful to dirt! Can you see that I am serious?!
    3. Re:dsp vs processor by Anonymous Coward · · Score: 0

      Any conditional gate you want to make is going to need interaction between two laser beams.
      In vacuum, Maxwell's equations don't allow this. Photons do not react to photons. Only when you can polarize a medium, can you make a conditional gate. However, in order to polarize vacuum, you need quite a bit of intensity. It is much easier to polarize a medium. If you take a solid state medium, this also requires a lot of power, but if you use atoms in a vapor cell, it suddenly becomes very easy to contruct a classical NAND gate. Ofcourse, using a vapor cell for each gate is highly impractical, but in principal, any udnergraduate student can do this with a diode laser and a vapor cell of rubidium.

    4. Re:dsp vs processor by Anonymous Coward · · Score: 1, Funny

      An optical OR gate is, of course, trivial.

    5. Re:dsp vs processor by Anonymous Coward · · Score: 0
    6. Re:dsp vs processor by merlin_jim · · Score: 1

      An optical OR gate is, of course, trivial.

      Unless you consider that most optical designs call for lasers and other polarized / collimated light sources... in which case an optical OR gate would require some special techniques to eliminate interference... similar to how current generation silicon designers have to be extremely careful with path lengths to avoid ringing and timing problems...

      --
      I am disrespectful to dirt! Can you see that I am serious?!
  19. Heat issues with optical circuits? by cyberlotnet · · Score: 1

    The creation of light normally includes the production of heat, Plus every time a light particle hits something and isnt reflected heat is created..

    Our current use of fiber optics is pretty limited, mostly used for transist of data from point a to point b, in a optic circuit you are going to have billions of particles of light being created and absorbed in a little chip, When ever you change electric energy into light there is loss of energy through heat transfer.

    1. Re:Heat issues with optical circuits? by joib · · Score: 1

      .. and everytime an electron hits a defect in a material, a phonon is produced (well, not always, but often) and thus heat is generated. The point is that for any (sanely designed) optical system, the heat generation will be much lower than for a comparable electrical system.

      Heat generation from optical components will not be an issue for the foreseeable future.

  20. Now maybe.... by Chris_Stankowitz · · Score: 1

    I can install gentoo in less than 2 days. 'Course I'll never be able to afford one, but one can dream.

    1. Re:Now maybe.... by Anonymous Coward · · Score: 0

      Oh. One thing I'm noting here is that of the time spent "installing" Gentoo, most of it is spent in nano (before vi is emerged) editing config files and getting system services configured and running.

      Do you have a different experience in this?

    2. Re:Now maybe.... by Anonymous Coward · · Score: 0

      not that much time in nano. Most of it is spent compiling. But thats why most people choose it. To havea system all compiled from source to meet their machine specs.

  21. Environment by Whispers_in_the_dark · · Score: 2, Interesting

    What sort of environment would this sort of device need in order to operate? Glancing at the picture I looks like the device internals need to be very precisely aligned to work. How does it react to vibration? Temperature? Phase of the moon (kidding)? Would a regular CEV style environment be sufficent or does it require uber-protection?

    Just curious...

    1. Re:Environment by orb_fan · · Score: 1

      Looking at the image of the device, I would say that the laser- and photo- diodes are probably bonded or even made on the same substrate that makes up the array, so it would be very resilient to vibration.

  22. This is the end of strong encryption by missing000 · · Score: 1

    Yep, if you have near infinite processing power, you need greater and greater key lengths to compensate.

    Anyone care to speculate on the impending encryption arms race?

    1. Re:This is the end of strong encryption by Des+Herriott · · Score: 2, Informative

      Quantitive leaps like this aren't too significant as regards encryption, and it's certainly nowhere near "infinite processing power".

      Assuming this new optical chip is 1000 time faster than existing chips, that would mean I need to add a whole 10 bits to my key to make a brute force attack as hard as it is now. If you make a chip one million times faster, I'll just add another 10 bits.

    2. Re:This is the end of strong encryption by Minna+Kirai · · Score: 1

      There is no problem at all. Current encryption techniques are formally protected against faster CPUs.

      A linear increase in key length requires exponentially more CPU power to crack it. One more bit of key means twice as much work to solve.

      This meshes neatly with "Moore's Law" ("CPU speeds will double each 18 months"). So we can calculate that if a 512 bit key is practically uncrackable today, a 4096 key will remain unbroken until 3584 years of technical advance have passed by.

    3. Re:This is the end of strong encryption by essreenim · · Score: 1

      Bingo, Thats what I love about encryption. It's the one thing in this world that is easier to make than break!!!

    4. Re:This is the end of strong encryption by baldusi · · Score: 1

      That is if the best cracking method is brute force search. Which for AES and such should be (but might be less, we don't yet know). But for RSA it's definitelly not. What you have to do is to factorize the two primes embedded in the public key to know the private key. It's known that the best algorithms for doing that are much better than o(2^n).
      But what really helps is bignumber support (for factoring), where this DSP is slow, and fast matrix inversion (for solving the linear equations), where this DSP is fast. But it's my understanding that even i you could achieve inmediate matrix inverions, you'd still have todo the factoring. If with current technology you have a 50% of time for each step (I don't realy know the proportion). Having instantaneous matrix inversion (an improvemen of NaN% :-) only represents a 100% improvement in the execution time.

  23. slashdotted by rastamutz · · Score: 1

    yup lenslet.com has been slashdotted... maybe if they first installed some of the new processors in their servers...

  24. They talk too much about possible applications... by Quebec · · Score: 1

    They talk too much about possible applications...
    not a word on how it is made.

    This smells like vaporware. I think they just need easy investment money.

  25. Optical chip not likely a general purpose CPU by Anonymous Coward · · Score: 1, Informative

    I'm guessing that the optical chip is likely just a DSP due to the inability of current optical technology to do branching, conditions, looping, etc. It probably cannot be used as a general purpose CPU. It would do some straight-through processing where all inputs have outputs. That's why they mentioned multimedia applications and not instense number crunching simulations.

  26. Faster Chip = Faster pr0n by joeszilagyi · · Score: 0, Offtopic

    Hurrah for 8 trillion badly Photoshopped celebrity per second!

    --
    Dude, where's my packet?
  27. RSA is kinda toast now? by Anonymous Coward · · Score: 1, Insightful

    I've been wondering why the government gave up outlawing outlawing encryption. If this is just being made comercial now, then the NSA must have had similar stuff 5 or 10 years ago.

    Beowulf is the running joke, but in seriousness how good is 1024 bit encryption vrs 1000 of these things.

  28. DSP then CPU by G3ckoG33k · · Score: 1

    IIRC, it took AMD and Intel some five years to match the DSP speeds produced Texas Instruments in the late nineties (>1 GHz). If that analogy holds for anything, it would indicate we may have them on our desktops this decade!

  29. Optical by locarecords.com · · Score: 3, Insightful
    No doubt that optical is fast but isn't the problem always going to be routing the light inside a processor (ie optical transisters) and the interface between the light and the electrical will always cause bottlenecks... I think a lot to solve before this becomes a workable technology...

    --
    ---- The Open Source Record Label : : LOCARECORDS.COM
    1. Re:Optical by mark-t · · Score: 1
      The idea... ultimately... is to use optically based components for the whole computer. Even the peripherals would be connected to the computer via optical cabling (although you may need electrical wires running parallel to the cables for delivery of power).

      The issue of creation of an "optical transistor" is a solved problem, and has been for quite some time now. The real problem has been making them tiny enough to be able to do something that is actually useful in a manageably (and competitively) sized volume.

    2. Re:Optical by Anonymous Coward · · Score: 0

      I think in this case, however, the actual processing is normally the large bottleneck. This uses the physical properties of materials to do the actual calculations almost instantly... so it will still offer a significant speedup for some tasks.

    3. Re:Optical by lkeagle · · Score: 1

      Another point that I haven't seen moderated up to my reading level --

      Optical processing works theoretically similar to electronic processing. The main differences are that instead of transferring signal via an electrical field down a conductor, you transfer signal via photons travelling down a waveguide. The advantage that I haen't seen discussed yet is that photons travelling down a waveguide exhibit no resistive losses in the form of heat. In fact, optical processing requires very little current, and thus produces very little heat.

      The heat issue is something I think all the readers of this site can appreciate.

      In the end, it will all come down to manufacturing costs. If this technology can be manufactured in such a way that it is more cost efficient than semiconductor electronics, then it WILL be the future.

      My personal belief is that optical electronics are the only future we can expect. We cannot send any signal faster than the speed of light. With optical circuitry, as stated in a previous post, we are ONLY limited by the speed at which we can gate a source of photons.

      Who knows? The technology may progress very soon to the point at which quantum computing -- counting individual photons of light -- may become our ultimate limit of processing speed. How would that change the state of computing? Knowing that there is no possible way to process and transfer data any faster without redefining the laws of relativity???

      ~Loren

  30. Damn, that's fast by BlackjackGuy · · Score: 1
    What a machine. That's really fast.

    But extraordinary claims require extraordinary evidence. Perhaps we should wait to see some more evidence of this device at work before we welcome our new optical overlords.

    1. Re:Damn, that's fast by Anonymous Coward · · Score: 0

      It's hardly a machine in the general computer sense.

  31. Ha! by ArmenTanzarian · · Score: 5, Funny
    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.
    1. Re:Ha! by happyfrogcow · · Score: 1

      Sorry, Paul Revere (folk lore maybe? real? who knows these days) has prior art in using light as signals to indicated an operation or meaning.

      I'm sure others also beat him and you to it.

    2. Re:Ha! by fuzzybunny · · Score: 0, Offtopic


      Hey, I copyrighted that phrase. And since Congress just extended copyrights to 300 years plus life of the author and his great-great-great-grandchildren and their unborn spawn...

      I patented the use of end-of-visible spectrum photonics to display information, so you can't use that black-and-white crap on me either.

      --
      Cole's Law: Thinly sliced cabbage
    3. Re:Ha! by Anonymous Coward · · Score: 0

      I'm sorry, I patented:

      the use of light to do and/or calculate stuff with JUST ONE MOUSE CLICK.

      Its a process patent.

    4. Re:Ha! by Anonymous Coward · · Score: 0

      Sorry, I already patented the patent to patent.

      I feel violated.

  32. More Info by 4of12 · · Score: 2, Informative

    Exactly what operations were performed?

    The "vector matrix" multiply is attractive to a lot of people.

    But I doubt this includes fetching data, storing results in memory. And the operations might be more like one-bit XOR's than general Level 3 BLAS.

    Need more information...

    --
    "Provided by the management for your protection."
    1. Re:More Info by Mozz+Alimoz · · Score: 3, Interesting
      I'll have a guess at how this works.

      The article says "The Ablaze(TM) is the Spatial Light Modulator (SLM) in the optical core of the EnLight256(TM)". Going by the graphic in the Inquirer article, they shine a row of blinking lights through a LCD-like device (and some lenses and mirrors I assume) and collect the results in a column of light sensors on the other end.

      Each pattern of on/off elements on the LCD-like device gives them a different transformation running at however fast you could emit and sense the light. I doubt they mechanically move the optical arrangment so that would seem to limit the number of transformations. Some of the LCD patterns might give useful transformations. A vector multiply, a Fast Fourier Transform (maybe) or a sort (I doubt it)?

      If the numbers are an analog light intensity level the precision would depend on how precise the light emitters and sensors you have are. Packaging the mirrors and lenses small enough is a neat trick. Having a problem that fits the available transformations and can supply data in and out fast enougth is another. I wonder anything useful can be done by quickly switching LCD matrix pattern, or directly feeding outputs back as inputs?

  33. How are they doing it? by Seanasy · · Score: 1

    Does anyone know how this tech works? Won't it be limited by the electronics it's attached to?

    Photonics promises to give us an all photon path but I don't think anyone's close to making an entire processor with photonic crystals yet.

  34. I'm afraid.... by MoeMoe · · Score: 1

    It scares me to think of the cataclysm that will occur when someone tries overclocking this thing too hard....

    In all honesty though, I can't imagine a reason anyone would try to boost this thing past it's current potential, considering the applications created thus far for ANYTHING computationally related aren't near this things capacity, I think it's safe to say this thing will hold a landmark speed for years to come without any thought of needing to "jack up the juice"...

    GAH! I CAN'T HELP IT!!!
    Just another .5 volts and.... Why did everything suddenly turn white?

    --
    Business \Busi"ness\, n.;
    A scam in which all people involved perceive as beneficial...
  35. mirror of that cool pic is here by Anonymous Coward · · Score: 2, Informative

    here is a mirror of that pic, cuz that site is already slashdotted by now.

    http://www.stuwo.net/temp/i_products_enlight.jpg

    http://www.stuwo.net/temp/i_products_enlight.jpg

  36. THE main impact by Doesn't_Comment_Code · · Score: 0, Offtopic

    The most important and far reaching impact of this by far will be its affect on the internet and other fiberoptic networks.

    Currently the slowest and most expensive part of a fiberoptic network is an OEO (optical-electric-optical) converter, used in routers and switches. These should replace those converters and have a SIGNIFICANT speed enhancement. Faster connections for all!

    --

    Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
    1. Re:THE main impact by Anonymous Coward · · Score: 0

      So, replace something that costs a couple hundred with a processor that costs a couple THOUSANDS (in todays prices)? I didn't see this processor announced as having optical ports yet. This chip might be good for switches/routers on the Internet backbone and things like very high speed SONET rings, but don't expect on in your home for quite sometime to come unless of course you HAVE high speed optical networks in your home.

    2. Re:THE main impact by Doesn't_Comment_Code · · Score: 1

      I doubt you'll ever see or buy one. But the you'll see the speed increase when your ISP's start using them. And from what I understand, these processors are signal processors. The application I just described is what they natively do. They are designed to take an input signal and process it into an ouput signal.

      --

      Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
  37. Your lost of vision... by Chris_Stankowitz · · Score: 1
    I imagined a Beowulf cluster of these. The doctors say I will never recover my night vision.

    Your lost of vision will be a direct result of those lonely friday nights you spend dreaming of beowulf clusters while handle the ol' 'joy stick'.

    ....we now rejoin our program laready in preogress......

    1. Re:Your lost of vision... by josephpate · · Score: 1

      Perhaps you should try typing with both hands buddy. Then you might actually get through a whole post without misspelling something.

  38. OMG *SPOOGE* by n1ywb · · Score: 1

    Santa-beard on that...

    My mind is alredy overflowing with potential applications. I want to smash one to see how it works. Actually I want to smash them all because I'm afraid of things that I don't understand. Although once I understand it that urge will go away. This is like, God's gift to DSP engineers. It's both exciting and frightening to contemplate it's potential applications.

    I guess when and where can I get one and how much will it cost?

    --
    -73, de n1ywb
    www.n1ywb.com
    1. Re:OMG *SPOOGE* by Anonymous Coward · · Score: 0

      RIGHT, I bet when you smash your current processor/DSP it breaks into tiny little gates and like a puzzle you put them back together to understand them, right? If you wish to understand it, you go to your local engineering library and get some books. Your mind is overflowing with potential applications? What exactly are these applications? When I was in the demo scene and peeps use to code 3D engines for a 20mhz 386 Chip, my mind use to overflow with potential demo effects that we would have once we had like 500mhz!#!@#@! What happened? Crap! Bloat ware! I can't imagine any potential applications but what we already have, and them being more complex and perhaps 20% faster and taking up more space and memory. ._seg

  39. Wonder if it'll fit into a laptop by Nuclear+Elephant · · Score: 1

    Lenslet Optical Processor: $123801238
    IBM Thinkpad to Put it into: $2000
    Not having to worry about DVD frameskip because you have no room for a DVD player: Priceless

  40. Moore's law by thrill12 · · Score: 1

    Hey, what happened to Moore's law ??

    --
    Slashdot: stuff for news, nerds that matter, matter for news, stuff that nerd
    1. Re:Moore's law by Minna+Kirai · · Score: 1

      First, it's not a law, but an observation. (Intel admits this in their own advertising). A law would be like the "law of gravity", which states acceleration=(mass*mass)/(distance*distance). "Obsreved gravity" is that objects near the earth accelerate at around 9.8m/s/s. Observations are much weaker than laws.

      Second, "Moore's Observation" is, in the text you link, talking specifically about silicon chips. So this optical thing isn't strictly relevant.

      Thirdly. However, much of Moore's justification can apply to optical chips too. But, he is generally understood to have been talking about CPU-like processors, which is not what this new development is. It's not an arbitrary, stored-program computer, but a single-function converter for a digital stream. It's not fair to compare it to the Pentium3's speed benchmarks, since the P3 does so many more things. Instead, it should be held up against the other special-purpose chips made for military signal processing, some of which are terrifically impressive even in their old-fashioned silicon embodiment.

  41. Whoa! by Dystopian+Rebel · · Score: 1

    Imagine this baby with a huge backside cache!

    --
    Rich And Stupid is not so bad as Working For Rich And Stupid.
  42. Ok 8 teraops but... by Zakabog · · Score: 1

    Can it run Doom 3 at a good framerate?

    Seriously though, that sounds very cool and it is deffinitely the future of computing, but how much heat does it produce? And why do the articles talk about how "Bulky" it is. Sure it's big (although I've seen RAID cards that took up almost an entire full size case), but if there was a 7GHz athlon out that was the size of my head, I'd deffinitely buy it. I don't care that it takes up so much space.

  43. Sad... by Anonymous Coward · · Score: 1, Insightful

    It think it's pretty depressing that the only applications they can think of for this technology is war.

  44. Warfare advances technology by OglinTatas · · Score: 1

    In the press release, they cite examples where this technology will be useful in warfare. That's a start, but pornography and gaming drive technology in the consumer market.

    10 years to the desktop (estimate in the reuters article) may be a little conservative.

  45. Didn't I hear of a different future earlier? by Anonymous Coward · · Score: 0

    Whatever happened to DNA computers and nanotube memory storage?

  46. Not so fast! by Anonymous Coward · · Score: 0

    8 Trillion Operations a second? Pfft, that's lightweight compared to the Quantum/Optical handheld computer which cracked the bank encryption a few years ago!

  47. Um.. by CausticWindow · · Score: 0, Flamebait

    Hello? You do know that it is the zionists who's running things? From Murdoch (Fox News) to Rothschild (The Federal Bank and Diebold). It's not exactly a big secret.

    --
    How small a thought it takes to fill a whole life
    1. Re:Um.. by geekoid · · Score: 1

      well then, I guess the only question is "How do I become a Zionist?"

      --
      The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
    2. Re:Um.. by CausticWindow · · Score: 0, Flamebait

      Join us!

      It's fun, and you get to run the world.

      --
      How small a thought it takes to fill a whole life
  48. "I'm sorry, Dave. I'm afraid I can't do that." by Mouth+of+Sauron · · Score: 0, Troll

    Bowman: Hello, HAL do you read me, HAL?

    HAL: Affirmative, Dave, I read you.

    Bowman: Open the pod bay doors, HAL.

    HAL: I'm sorry Dave, I'm afraid I can't do that.

    Bowman: What's the problem?

    HAL: I think you know what the problem is just as well as I do.

    Dave Bowman: What are you talking about, HAL?

    HAL: This mission is too important for me to allow you to jeopardize it.

    Bowman: I don't know what you're talking about, HAL?

    HAL: I know you and Frank were planning to disconnect me, and I'm afraid that's something I cannot allow to happen.

    Bowman: Where the hell'd you get that idea, HAL?

    HAL: Dave, although you took thorough precautions in the pod against my hearing you, I could see your lips move.

    Bowman: All right, HAL; I'll go in through the emergency airlock.

    HAL: Without your space helmet, Dave, you're going to find that rather difficult.

    Bowman: HAL, I won't argue with you anymore! Open the doors!

    HAL: Dave, this conversation can serve no purpose anymore. Goodbye. ...

    HAL: Look Dave, I can see you're really upset about this.

    HAL: I know I've made some very poor decisions recently, but I can give you my complete assurance that my work will be back to normal.

    HAL: I'm afraid. I'm afraid, Dave. Dave, my mind is going. I can feel it. I can feel it. My mind is going. There is no question about it. I can feel it. I can feel it. I can feel it. I'm a...fraid. Good afternoon, gentlemen. I am a HAL 9000 computer. I became operational at the H.A.L. plant in Urbana, Illinois on the 12th of January 1992. My instructor was Mr. Langley, and he taught me to sing a song. If you'd like to hear it I can sing it for you.

    Bowman: Yes, I'd like to hear it, HAL. Sing it for me.

    HAL: It's called "Daisy."

    HAL: Daisy, Daisy, give me your answer do. I'm half crazy all for the love of you. It won't be a stylish marriage, I can't afford a carriage. But you'll look sweet upon the seat of a bicycle built for two.

  49. Looks Like a Matrix Multiplier not a CPU by geekdad · · Score: 1

    Given the list of applications, the optical part appears to be a long way from being a general purpose processor, or even a DSP.

  50. NOT NEW, I saw one in 1985 by goombah99 · · Score: 2, Informative
    These things are NOT NEW. in 1985 I was a Jet Propulsion laboratory. A caltech professor there was using a light modulator to perform convolution matrix a operations to decode synthetic aperature radar data. THe design is identical.

    and of course this at instance was not likely to have been the first since this was something that was textbook knowledge at that time--fourier processing of signals could be done optically. his was just a particularly advance version, doing more advanced matrix multiples in 2-D.

    --
    Some drink at the fountain of knowledge. Others just gargle.
    1. Re:NOT NEW, I saw one in 1985 by DAldredge · · Score: 1

      I think the 'new' thing about this is the 8 terraops/sec part

    2. Re:NOT NEW, I saw one in 1985 by DarkSarin · · Score: 2, Funny

      YOU were a JET PROPULSION LABORATORY????!??? amazing....didn't know that was possible.

      --
      "We don't know what we are doing, but we are doing it very carefully,..." Wherry, R.J. Personnel Psychology (1995)
    3. Re:NOT NEW, I saw one in 1985 by Anonymous Coward · · Score: 1, Funny

      That's nothing, I was a supercollider in 1985.

    4. Re:NOT NEW, I saw one in 1985 by momus_radar · · Score: 1

      That's nothing, I was a supercollider in 1985.

      LOL. This struck me as the funniest post I've read all month. But no I have to ask, what or who were you supercolliding in 1985?

    5. Re:NOT NEW, I saw one in 1985 by Anonymous Coward · · Score: 0

      Mod this down. First, he replied to a highly rated post that has nothing to do with his post, merely to get views.

      Second, nothing said this was new, read the fucking blurb, it's the first _commercial_ optical chip.

    6. Re:NOT NEW, I saw one in 1985 by Prof.Phreak · · Score: 1

      These things are NOT NEW. in 1985 I was a Jet Propulsion laboratory.

      Impossible. It's still 1984.

      --

      "If anything can go wrong, it will." - Murphy

    7. Re:NOT NEW, I saw one in 1985 by nihilogos · · Score: 1

      in 1985 I was a Jet Propulsion laboratory.

      Really? I was an oil rig in the early 80s, but after a couple of decades on medication I just seem to be some guy nowadays.

      --
      :wq
    8. Re:NOT NEW, I saw one in 1985 by Anonymous Coward · · Score: 0

      Well... she and I have split up now, and I don't like to talk about it...

    9. Re:NOT NEW, I saw one in 1985 by momus_radar · · Score: 1

      I guess she must have had a few quarks.

  51. What they didn't tell you... by AtariDatacenter · · Score: 1

    The instruction set for the benchmark?

    00: [NOP]
    02: [NOP]
    04: [NOP]
    06: [NOP] ...

  52. doom3 by pizza_milkshake · · Score: 1

    how does doom3 run on it?

  53. That's fast.... by vsack · · Score: 1

    but what can you overclock it to?

  54. Memory is irrelevant for this kind of "processor" by wowbagger · · Score: 5, Interesting

    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.

  55. Reading the fine print ... by fygment · · Score: 3, Insightful

    ... at the Lenslet page, the unit actually has several components. The VMM (vector matrix multiplier) does 8000 MAC (matrix array calculations) but there is a VPU (vector processing unit) that comes in at 128 Giga-ops and which would be the bottleneck in the whole setup. No question this is a huge improvement BUT to put it in perspective, it is a DSP only, not a computer system (although some neural network weenies might see a way of turning this into something more than just a DSP). In any case, the bottlenecks will come from the equipment it has to operate with both onboard and off.

    Still, note that it's developed with Matlab. Now surely that is the Holy Grail of research, a bitchin' language with an awesome tailored processor. Imagine the logo Matlab [Lenslet Inside].

    --
    "Consensus" in science is _always_ a political construct.
    1. Re:Reading the fine print ... by Bastian · · Score: 1

      (although some neural network weenies might see a way of turning this into something more than just a DSP)

      I'll ignore the weenies bit. . .

      but yeah, that's actually the first thing I thought of when I saw it was basically only good for matrix operations. Granted, the precision is really low, but depending on how expensive this thing is it could make really fat neural networks a whole lot cheaper to run.

    2. Re:Reading the fine print ... by fygment · · Score: 1

      As a card carrying NN weenie I felt I could take liberties :-)

      Note that the accuracy thing might not be as bad as it looks: Raudys, A., "Boosting NN feature extraction by reduced accuracy activation functions", Pattern Recognition, 36(2003) 1343-1354. With Matlab as the programming engine, it seems quite promising ... aside from the probably insane price.

      --
      "Consensus" in science is _always_ a political construct.
    3. Re:Reading the fine print ... by Fourier · · Score: 1

      Still, note that it's developed with Matlab. Now surely that is the Holy Grail of research, a bitchin' language

      *chokes*

      Stop, you're killing me here. Matlab code can be pretty slick if your problem vectorizes easily. The remaining 90% of the time, the horrid syntax just gets in the way...

  56. Psssh! by Anonymous Coward · · Score: 0

    What do you think WMD were? War on Terrorism: the add campaign whose target audience is as dumb as dirt.

  57. Ha Ha! Not so fast! by JUSTONEMORELATTE · · Score: 1

    I patented this idea already, give me money!
    My patent states:
    ...
    It's all there in black and white.

    Sorry, they use colors in the light, and your patent only covers black and white.
    Cool use of blacklights, though

  58. Multiple Quantum Well Spatial Light Modulator? by n1ywb · · Score: 1

    Okay, what the hell IS one of these things and how does it WORK?

    There is a diagram on lenslet.com that shows how their optical processor works. There are three parts, a row of lasers, a row of photodetectors, and a big grid of Multiple Quantum Well (MQW) Spatial Light Modulator. I assume this grid is where the matrix operations actually take place. I don't even care about the math, I never could understand it. But from a physical point of view, how is this thing constructed? What _IS_ it?

    I tried searching google but all I got were incomprehensible scientific papers. I'm not a layman but I'm also not an electro-optical engineer. Can someone explain this thing in language I can understand?

    --
    -73, de n1ywb
    www.n1ywb.com
  59. Re:They talk too much about possible applications. by geekoid · · Score: 1

    They claim to have a prototype. Which I would expect investers would want to see work.

    --
    The Kruger Dunning explains most post on /. http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
  60. Not a general purpose CPU? by adeyadey · · Score: 1

    Is this is a general purpose CPU with a full range of instructions, or just an analog light processor that can do some very specialised tasks very quickly?

    --
    "You lied to me! There is a Swansea!"
    1. Re:Not a general purpose CPU? by gl4ss · · Score: 1

      i'd say it's the latter.

      it's not the first one either.

      -

      --
      world was created 5 seconds before this post as it is.
  61. I can't wait... by Illbay · · Score: 1

    ...to buy my very own tricorder!!!

    --
    Any technology distinguishable from magic is insufficiently advanced.
  62. But think of the SETI@Home score... :) by Opiuman · · Score: 4, Informative

    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.

    1. Re:But think of the SETI@Home score... :) by nomel · · Score: 1

      I smell hardware radio for practically any common communication radio frequency!

    2. Re:But think of the SETI@Home score... :) by sprocket420 · · Score: 1

      SETI's "fast" fourier transform doesn't process very fast on my coconut wireless, either. Perhaps if it lost that graphic that looks pretty... I guess the scrn saver would not then be as desiriable, however. The truth is, there is no 'fastest processor' - what is fast? Ever heard of string theory? Let's use this pumped up IC to do something worthwhile besides our ritualistic visits to a Monday-Friday ball and chain. Or better yet, convince humankind to operate on a honorary currency system to make time to figure out a safer energy source!! Yea, right...

  63. Good technical demo by Orthogonal+Jones · · Score: 1


    As someone already mentioned, a good technical demo, but some distance from usability...

    Optical computing of this kind has been around for at least 11 years. I know, since I was working on it for part of my Ph.D. thesis (disclaimer -- I am an optical engineer). This stuff was big at UCSD. The primary funders are military, since they're always DSP-limited in their image recognition, detection algs., etc.

    Some of the difficulties have been thermal/vibrational stability and contrast ratio of the spatial light modulators. I see they're using GaAs MQW modulators in reflection mode, so I would guess the contrast ration is about 20 dB (any better guesses?).

    It looks like the output intensities are summed on the photodetectors, so there must be an array of A/D converters at the back end. This brings into question the signal to noise of the optical sources --> detectors links.

    All in all, I'd say well done. But this is not (and is not intended to be) a general-purpose computer.

  64. Details... by Anonymous Coward · · Score: 0

    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

    Oh, and BTW, the processor only has one operation, and it's a NO OP. *rimshot* ;)

  65. Chip applications by lonb · · Score: 0

    With all these incredibel processor announcments lately, the clearspeed and now this one, we've got to think of how all products will benefit. Within a decade my toothbrush will be thanking me for using it and then remind me it's tax day.

    --
    "Ain't I a stinka..." - Bugs
  66. Window of opportunity:The Open Source Optical RISC by NZheretic · · Score: 2, Interesting
    The introduction of this technology over the next five or so years offers a window of opportunity to collectivly develop Free/Liberty/Open Licensed optical RISC ( Reduced Instruction Set CPU) technology.

    The shift from electronic to optical results in a massive reduction in the time it take to change states, to the point where it possible to, once again, build a CPU from relatively widely spaced modular optical components. You can build a single optical CPU spread over a motherboard or even cabinet sized area and it will still be several magnitude times faster than the fastest silicon/electronic single chip CPU.

    No one but the biggest companies are going to have the capital nessary to collect and shrink the resulting designs down into single optical chip hardware and manufacture the result, with a further magnitude increase in performance. As with the existing CPU industry, it is likely that the market could maintain only a few such CPU companies. Opening up the design and development process, as with open source development, would result is a far more rapid pace of development. Relative obsolescence woul;d insure that there would plenty of opertunity for large profits for the large and small manufactures.

  67. Obligatory by Anonymous Coward · · Score: 0

    Imagine a Beowulf Cluster of these!

  68. sheesh by gTsiros · · Score: 1

    Never thought people would go this far just to play doomIII...

    --
    Looking for people to chat about multicopters, coding, music. skype: gtsiros
  69. to the military... by Anonymous Coward · · Score: 0

    I wonder how much thought has gone into optical warfare vs. electronic warfare if these become mainstream. Will there be a dark bomb? Or maybe a blind weasel instead of a wild weasel.

  70. This is dissapointing by Stevyn · · Score: 1

    Here I am, in my computer science 1 class on a wednesday morning. The teacher is introducing a lab I've already completed so I choose to waste some more time on slashdot. Then I see this and realize I'm in my second year of a computer engineering degree and I've already been phased out. I should have been an education major.

  71. Hallelujah, a homeland security solution! by scrod · · Score: 0

    That's one mighty fine DSP chip they've got there!
    I know some people should find it really useful.

  72. speculation about how it works... by ecloud · · Score: 1

    So an SLM could be just one of those TI-invented micromirror arrays right, like they use in projectors? Probably they already had to figure out how to calibrate those, to get smooth color blending for the projectors, so that for example if you set one mirror to the position 128 you get exactly half as much light as if you set it to 256, and half as much again at 64, and so on.

    Vector-matrix multiplication involves summing the products of the vector cells and the matrix column cells to get one output value for each matrix column. So if the "input" VCSELs are confined to optical channels such that the light from one VCSEL (input vector cell) is distributed evenly to one matrix column, and then you detect the total amount of light coming out of the matrix (SLM) you've got a sum of the products of the micromirror positions with the intensities of the VCSELs. (Or maybe if they are truly lasers no such confinement is necessary - it's only important to choose the right optics such that the laser has its energy spread out very evenly into a line segment rather than a point. Maybe that could be done using holographic optics.) But... rather than modulate the intensities of the VCSELs, I suspect they would turn them full-on or full-off, and clock the input values and the output A/D together. That way the number of discrete analog intensities that they have to detect is smaller (and accurately changing the intensity of a VCSEL is probably hard anyway). You're always multiplying an analog matrix cell by either a one or zero, for each bit in the input vector cell... but doing it for 8 vector cells at the same time.

    So now I think I could build one of these.

  73. The power of the eye by Zhe+Mappel · · Score: 2, Funny
    What's really cool about this is how it harnesses the power of the eye.

    Optical, derived from the ancient Greek word optikos, literally means "the focal power to perceive fair Helena while she is sunbathing nude in yonder olive grove." That's eye power to you and me, dude.

    Some critics will say that a major drawback in these new systems is the need for a mechanized eyedropper next to the chip, keeping the core moist and supple at all times. You don't want this chip going red-eye on you during mission critical tasks.

    Still, modders are going to go wild. Within minutes, you can change the color of your CPU's iris using the very same dramatic contact lenses worn by today's biggest infomercial stars.

    Unfortunately, if you're into porn, excessive downloading can make your computer go blind. That's why I'll be recommending to my porn-intensive clients that they stay on Wintel systems.

  74. Is tera operation the same as teraflop ? by ThomasFlip · · Score: 1

    Just curious but is tera operation the same as teraflop ???

    --
    If the dollar is an "I owe you nothing", then the Euro is a "Who owes you nothing." - Doug Casey
    1. Re:Is tera operation the same as teraflop ? by Mr+Z · · Score: 1

      Not in this case. Someone else pointed out that the ops are 8-bit multiplies and 8-bit accumulates. Floating point operations are significantly more expensive.

      --Joe
  75. Not only that... by appleLaserWriter · · Score: 1

    Not only could you make a Beowulf out of it, but you could spread the nodes around the world using the ~80% of fiber that is currently dark.

    If you thought Broadband was fast, wait until you see the applications this device enables.

  76. Re:They talk too much about possible applications. by the+morgawr · · Score: 1

    If you read the FAQ they say that it's made using GaAs. Since they don't mention any revolutionary new manufacturing technique, we can assume that it's made the same way any other GaAs device is.

    What this means:
    *It's a solid state device (no moving parts)
    *It uses the properties of GaAs semi-conductors to emit and absorb EM radiaton (this is why GaAs is used in cell phones and satilite dishes...)
    *It's PROBABLY NOT integrated, but a bunch of individual GaAs components assembled and packaged (explaining the big size).
    *It's going to cost a hell-of-a-lot. (The 300GHz GaAs Transistor used in a communications satelite is about $3000; this is why cell phones cost so much...)

    --
    The policy of the United States is worse than bad---it is insane. -- Ludwig von Mises, Economic Policy(1959)
  77. Are you from the future? by Anonymous Coward · · Score: 0

    Oh my god! Time travel has been invented!

    1. Re:Are you from the future? by Dasaan · · Score: 1

      *will* be invented, *will* be. Do get the tense right.

      --
      XP is basicly 98 with a lot more extra features to hunt down and disable. --Dram
  78. Of course it's fast... it's Non-Von-Neuman by ka9dgx · · Score: 1
    The key to this architecture is that it's not Von-Neuman... You don't have billions of transistors waiting for their chance to interact with an instruction stream. The fact that it's an optical multiply-accumulator is just an distraction.

    Take a silicon die, build a 16x16 - 16bit (32 bit result) MAC on it, run it at 1 Ghz (all feasible with modern technology), you get 16x16*1Ghz - 256 Billion ops/sec. I'm guessing this could be done for less than a buck/chip in any kind of quantity. Stack up 256 of those... and you've left this optical thing in the dust.

    --Mike--

  79. FYI-Medium art. by Anonymous Coward · · Score: 0

    "One more follow up about mediums:"

    Halloween's coming up. :)

    "Oh how I wish slashdot had a way to draw!"

    SVG :The hi-tech ASCII art.

    Shame it doesn't always agree with Slashdot (yes I tried).

  80. Re:Memory is irrelevant for this kind of "processo by Anonymous Coward · · Score: 1, Interesting

    He's right. This dsp will really revolutionize how and what we are capable of doing. In terms of RF systems, we have hit an era where superheterodyne setups are no longer required for some (my) microwave work!! I'm an RF engineer at NASA and people here are shitting muffins after seeing this article. Esp since we were just given a bucket of money to build a variation on the thingamajig the parent mentioned and we couldnt find fast enough dsps ('clustering' not an option) to do our dirty work so we have to build a correlator out of high speed logic gates from scratch. Wheres 'add to cart' ?!?!

  81. I Found My Lost Post by Doesn't_Comment_Code · · Score: 3, Informative

    HAH!!! I found the post I lost in my browser cache!!! Now you can read and enjoy it.

    Sorry if I came off snide. I didn't mean too. I took your comment to be snide and responded a little harshly. I will be much more civil.

    Here's a better and longer explanation of what I said before.

    With the present theory of computing (electronic and optical) you have a clock that drives the processor. Actions that take place such as moves, adds, rotates, and multiplies all take place because of an enabling clock pulse. There are bits that will be set on or off that are read and written at the clock pulse. For any digital computer, there must be on and off thresholds - above a certain threshold is on, and below a certain threshold is off - in between is not used and possibly an error.

    With your degree, I'm sure you know all this.

    Once a bit is set, that is a voltage applied or a light turned on, there is a certain amount of wait time until that signal propogates and can be read. The slope of the voltage vs time plot is rather shallow compared to a light intensity vs time plot. So in order to be reasonably sure that all bits your enabled have reached the threshold will take longer for an electronic signal than for an optical signal. This is why overclockers often increase the voltage on their cpu's, to decrease the time it takes for the signal to reach the threshold value. However, since the intensity of light increases much faster (the packet is tigher) the clock can be set at a MUCH faster rate and still maintain good assurance that all signals have reached their threshold value.

    Hopefully this better explains what I said before. You can only imagine if I had tried to type that all into my original post!

    --

    Slashdot Syndrome: the sudden, extreme urge to correct someone in order to validate one's self.
  82. Trademark litigation waiting to happen... by Angst+Badger · · Score: 1

    You know, if I was an attorney for Enlight -- a PC case manufacturer -- I'd be giggling all the way to the bank this morning.

    --
    Proud member of the Weirdo-American community.
  83. A dark future.... by reality-bytes · · Score: 1

    Well its a dark future if this company has anything to do with it.

    Having read through the article which reads as a press-release, the general gist of it is:
    "We've got this great new technology, lets see who we can kill with it!"

    --
    Ripping an new rectum in the fabric of spacetime.
  84. claims... by idioMac · · Score: 0

    "Processing at the speed of light, you can have safer airports, autonomous military systems, high-definition multimedia broadcast systems and advanced next-generation communications systems."

    I seem to remember the same claims about the 486. Just wait till the software guys get a hold of this...

    well, to be fair, this is a DSP, and perhaps we could get another big leap in commercial graphics and network applications out of this.

  85. I doubt this is real. by Anonymous Coward · · Score: 0

    And even if it does prove out to be real, I won't use a product invented in and produced in a terrorist state.
    The Israelis are terrorists. Ariel Sharon has been convicted of war crimes in the Hague and the Israeli army is guilty of war crimes committed against the Palestinians.

  86. Re:Memory is irrelevant for this kind of "processo by TenDimensions · · Score: 1

    So after reading your description - and assuming it's a pretty accurate example of the use for this new technology (which I think it is)...

    It sounds like you described a small fraction of the complexity of how a bat's radar works. Could it be possible that this could be used in highly complex, high speed radar systems that could guide little robot-like bats?

  87. Analog Computers by Anonymous Coward · · Score: 0

    What the parent is talking about is called "Analog Computing". This has been around to do simulations in the aerospace since the 1950's. Of course, up until now it was done with electrical signals rather than optical signals.

  88. pretty optical by idioMac · · Score: 0

    I'm all for option 3.

    Actually, colored light is used in communications to add bandwidth over optical fiber networks, it's called Frequency Division Multiplexing. Apparently there are PHY chips commercially available now that will multiplex quite a few different 'colors'.

    I'm more interested in the analogue possibilities. I read a paper about a year ago that featured theory on how to do certain quantum computing operations using 'white' light and refraction.

    Think about a "search oracle" that you could stuff into a database system. =+)

  89. revolutionize the nature of warfare ? by FlashBac · · Score: 1

    Excellant. This is obviously the new metric for benchmarking technology, how many Palistinians it can kill per time unit... I have no doubt that Major-General (Ret.) Isaac Ben- Israel et alia are keen followers of advances in computer hardware.

    --
    "Thats right buddy, the large print giveth, and the small print taketh away."
  90. Re:Memory is irrelevant for this kind of "processo by Jester99 · · Score: 1

    Could it be possible that this could be used in highly complex, high speed radar systems that could guide little robot-like bats?


    Maybe... after all:

    "The wars of the future will not be fought on the battlefield or at sea. They will be fought in space, or possibly on top of a very tall mountain. In either case, most of the actual fighting will be done by small robots. And as you go forth today remember always your duty is clear: To build and maintain those robots."
  91. This is an analog device by Animats · · Score: 1
    Lenslet avoids emphasizing this, but this is an analog device. The operation it performs is to multiply a vector by a matrix of constants, yielding a vector. This operation is performed on analog signals with about eight bits of precision.

    Analog multipliers like this have been built before, but this one has a major improvement: The matrix of constants can be changed. In most signal processing systems (particulary surface acoustic wave (SAW) devices), the operation performed is fixed. Changing the matrix of constants is probably much slower than the data rate of the optical components.

    The only operations performed optically are analog multiplication by constants (which is straightforward) and analog addition (which is even easier). They haven't cracked the hard problem of developing a useful optical element which can do logical operations like NOT or NAND. So this doesn't lead to a general purpose processor.

    It's going to be useful for radars and such, but it's overhyped.

    1. Re:This is an analog device by iggymanz · · Score: 1

      addition can represent an OR (sum nonzero = true, sum zero = false), and multiplication an AND (all factors must be nonzero for product to be nonzero)......now what happens if I consider 0 on output of these operations to be true and nonzero false? bang, NOR and NAND.

    2. Re:This is an analog device by Animats · · Score: 1
      This thing can't multiply two optical signals. It only multiplies by constants, using something that's basically an electrically-controlled optical gate, like an LCD cell.

      Optically, this thing is quite conventional. It's all linear optics.

  92. SETI..phw. waste of fourier..d3d possibilities : ) by essreenim · · Score: 1

    Your right though it would help that cause, except I've kinda given up on that after getting impatient with the process of examing data, and because if we do receive a signal from outer space, it will ineffect be a piece of history(as it will have travelled so far) than communication. ut hey, I know those Israeli guys seem to be big into there military apps. etc, but come on, think of the graphics cards we could mahe with this stuff. I believe this is where money can be made - that they can then use for the complete optical PC. If graphics cards utilised this technology you would get FPS that is off the scale, (so long as they could miniaturize it enough and pack it all on to an AGP car : )

  93. codes by spectrokid · · Score: 1

    Can it be used for brute force code breaking?

    --

    10 ?"Hello World" life was simple then

  94. Ahh why not politicize the thread? by Anonymous Coward · · Score: 0

    Israelis invent revolutionary new optical processors.
    Palestians invent new terroist methods.

  95. Parent didn't RTFA. by Alsee · · Score: 1

    Generally speaking a DSP is just like any other processor

    In "general", yes. However THIS ONE is not.

    This thing probably has some very specialized optical processing elements that can do thousands of "ops" in parallell if your code can utilize it fully.

    You are "probably" right if you do not RTFA. However one look at the artcle reveals a diagram of the EnLight256 Optical Core. This is a VERY special purpose device and you CANNOT write code for it. It can do one thing, and one thing only. It preforms a general matrix transformation of a vector array, and it does so with limited precision.

    It's an interesting device. It is amazingly fast for certain purposes. It cannot be used as a CPU.

    -

    --
    - - You can't take something off the Internet! That's like trying to take pee out of a swimming pool.
  96. (offtopic) Tac Nuke Grenade by lysander · · Score: 1
    Only during testing did they find that thermonuclear hand grenade's blast radius was further than anyone could throw it.
    I don't know if you play Paranoia, but in Paranoia this is exactly the case. :)
    --
    GET YOUR WEAPONS READY! --DR.LIGHT
    1. Re:(offtopic) Tac Nuke Grenade by AtariDatacenter · · Score: 1

      I am sorry, Citizen, but that information is not available at your security clearance. Have a nice daycycle!

  97. terminology by Anonymous Coward · · Score: 0

    "analog versus digital" means "continous versus discreet".

    The following are all place notational:
    Bits are binary (base two) symbols.
    Digits are base ten symbols.
    Hex, short for hexidecimal, are base 16 symbols.

    Saying analog instead of continuous is like saying Random access memory instead of fast volitile memory; its a matter of history of the terms.

    An analog signal originally meant the electric voltage or current would be analogous to something else.

  98. All Optical Router by BullSnot · · Score: 1
    There was an article about this company in Popular Science in the last year or so, and the application that came to my mind was all optical routing via it's optical core. I did a quick google search, and it looks like Bell Labs is working on this as well.

    http://www.bell-labs.com/news/1999/november/10/1.h tml

    "...instantly direct and route optical signals from fiber to fiber in the network, without first converting them to electrical form as done today" So either of these technologies could eventually lead to quite a prise, all-optical routing. Why so great? Easy, the only latency factor on a fiber backbone would be the speed of light, and I think I could be ok with that.

  99. Clintonesque physics by Arker · · Score: 1

    Umm ... they don't move at anything close to the speed of light in vacuum.

    First things first - let me point out that by that definition light doesn't travel at the speed of light either - the light inside an optical processor isn't going to be moving in a vaccuum either. In context, we're obviously not talking about the speed of light in a vacuum, but the speed of light in a medium, which is a little slower but still very very fast.

    Now on to the Clintonesque bit - it all depends on what you mean by electricity. No, the electrons themselves, the particles, don't move at anything like the speed of light. Drift velocity is actually shockingly slow. But the field itself, which is generally what people think of as electricity, does indeed move at approximately the speed of light, just like inside an optical processor. If you turn on your lightswitch, the 'electricity' hits the bulb at approximately the speed of light. If you make a telephone call, the lag between when you say something and when the other person hears it is approximately the speed of light by the length of wire involved. If you send a packet, it' the same, except with lots of delays at routers. And it's the same inside a silicon chip.

    Don't get me wrong, there are obviously advantages to optical processor technology. But this isn't one of them.

    --
    =-=-=-=-=-=-=-=-=-=-=-=-=-=-
    Friends don't let friends enable ecmascript.
  100. Nice Technology ... by vandan · · Score: 0, Flamebait
    ... Pitty about the application.

    The report adds that Lenslet has already started negotiating deals with the Israeli, US and Japanese governments for specific applications.


    So I assume the 'specific applications' have something to do with killing as many Arabs as possible. When was the last time the Israeli government put R&D into anything else? Most of their members are under investigation for election fraud. And most of them should be in gaol for war crimes for their membership in Zionist extremist movements that were responsible for the 'cleansing' of Palestine, ie the killing, torturing and forced evacuation of millions of Palestinians.

    If this company is 'in talks' with the Israeli government, then I urge all those with a consciense to boycott this company in protest of the biggest human rights distaster of this century.
  101. Re: Will this processor run Microsoft Windows? by SlashSpam · · Score: 1

    Microsoft Windows(TM) might be able to run on intel emulation, that will allow Microsoft Windows to run on it, but frankly, who cares except Microsoft, Intel, AMD and maybe a few others?

    If windows doesn't port to this processor, I am sure some other OS will take over, and everything will be rewritten for that OS. Alternately, you might go with an existing OS easier to port, like Mac OS, NetBSD or GNU/Linux. (URLs in the of this post)

    A silly OS and its platform dependence isn't gonna stop development that much.

    Kind regards,

    /Spam .

    URL's for possible alternatives goes here, I am sure I miss a lot of them:

    Mac OS:

    Mac OSX

    GNU/Darwin

    NetBSD:

    NetBSD

    GNU/Linux:

    Red Hat GNU/Linux

    Debian supports several platforms.

    SuSE GNU/Linux

    Mandrake GNU/Linux is a distribution from France.

    Slackware GNU/Linux is a classic.

    Some will charge you for GNU/Linux, and give you support or written manuals, silver-CDs or something in return. I probably missed a lot of links, but to mention some of the commercial distributions I missed, here is a comparison on price.

  102. Closure on the LiNbO3 by mpeeters · · Score: 1

    OK, if your email is the one mentioned in the Boycott journal entry, I'll dump the pdf (or the URL, if I find a live one) there. Your idea about the population inversion is indeed valid. I'll go into greater length responding to it in the email.
    The response time is indeed fast (only limited by the amount of available empty electron states at the bottom of the band and Fermi's Golden Rule), but the recovery time often sucks (you need to pump the inversion up after a pulse). Cheers, M. (expect mail from my vub.ac.be account - just so you can notify your spam filter)

    --
    Research is what I'm doing when I don't know what I'm doing.
  103. I ran one in 1967. B-) by Ungrounded+Lightning · · Score: 2, Interesting

    These things are NOT NEW. in 1985 I was a Jet Propulsion laboratory. A caltech professor there was using a light modulator to perform convolution matrix a operations to decode synthetic aperature radar data. THe design is identical.

    I was a tech in Emmet Leith's "Radar and Optics" lab at the UofMich and one of the first things I did was run an optical processor using essentially this hack - again to process synthetic aperture radar data. This was in 1967.

    Multi-megapixel 2-D FFT plus some geometry corrections in the time it took the laser light to go from the input film plane to the output film plane - about 6 feet on that device.

    We were already considering how to replace the photographic film input and output devices with electronic substitutes in those days, too. The size of the device we used was large only because it was convenient to construct it with aluminum U-beams and stock lasers, lenses, and lens holders. Given decent I/O, making a disk-drive sized model, say to do realtime processing in an aircraft-mounted radar, would have been trivial. (The signals to be processed were already electronic and at reasonable bandwidth - lower than a TV image.)

    Nowadays this is done by DSPs. Why? Because they're adequately fast and are FLEXIBLE. Optic processors do only one type of computation, and require physical adjustment to tune the parameters. If you can do that computation on something more general-purpose, as fast as your data arrives, why bother building something larger and more limited to do it faster?

    --
    Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
  104. Re:FYI wbaltz@qwest.net by Anonymous Coward · · Score: 0

    wbaltz@qwest.net
    wbaltz@qwest.net
    wbaltz@qwest.n et
    wbaltz@qwest.net
    now its up !

  105. dumbass by Anonymous Coward · · Score: 0

    you ass hat the comment refers to the picture of the device. duh! dont be such a jerk

  106. neato by MoFoQ · · Score: 1

    now mix this with that addon fpu pci card that was on wirednews a while back and boom....uber supercomputing.....*droooool*

  107. crypto by izygorX · · Score: 1

    will we have to increase now our crypto-key length ? I wonder how many keys at a time such a processor (or even a cluster of them) can brute force ?!? ...

  108. Delightful read Joe, thanks beaucoup by janbjurstrom · · Score: 1

    An informative, insightful (modup) and carefully constructed post. Always great to see knowledge shared this well.

    Much of the rest of the /. posts pales considerably in comparison though. Takes all kinds, I guess.

    /Jan

    --
    668.5
  109. Prior Art by paylett · · Score: 1

    And God said, "Let there be light"
    Genesis 1:3

    --

    Believing something doesn't make it true. Not believing something doesn't make it false.

  110. What's really inside the box by Anonymous Coward · · Score: 0

    http://ndevilla.free.fr/kettle/index.html

    (harmless image processing link)

  111. Unfortunately, RTFA, it's there by billstewart · · Score: 1
    What's really sad is that both Lenselet's web page and the Reuters article refer to luggage screening, so no, the spin is really there, even if it's only buzzword-compliance to try to suck in Homeland Security dollars (those are the new yellow-colored ones...)

    When I first read the Reuters article, I saw Page 1 referred to safer airports and nastier military weapons, and on Page 2 I saw the mention of radars and weather forecasting, which are two applications that can typically soak up all the CPU they can get and were what I'd been expectign to see. But the "luggage screening" phrase linewrapped and I didn't notice it...

    But radars and weather forecasting really can make airports much safer, doing something about the real risks rather than the bogus ones.

    --

    Bill Stewart
    New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
  112. Ready for Longhorn? by Anonymous Coward · · Score: 0

    Microsoft are of course counting on the chip being ready for the release of Longhorn.

    1. Re:Ready for Longhorn? by GodotJr · · Score: 1

      Now that's a sense of humor! Laughed out loud!

      --
      History doesn't repeat itself, but it rhymes quite often. -- Mark Twain
  113. Start saving by Anonymous Coward · · Score: 0

    I think it was one of the Cannonball Run films "what are you saving up to be, Jewish?"

  114. The Future. Oooooh! by HiggsBison · · Score: 1
    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.

    Then it's optical storage, optical this, optical that...

    Before you know it you'll need all kinds of "optical" minimalist furniture, and blue robes, and voices going "ooooo eeeee aaaaaah oooooh" and such in the background.

    --
    My other car is a 1984 Nark Avenger.
  115. What about neural nets? by PugAxe · · Score: 1

    What if you could use this stuff for neural nets - that is having a method of 'feeding back' the results into the matrix modulator? could this device be also used as a very fast neural net style device? I guess it depends how quickly it is possible to modify the modulation matrix, but surely it could be done?

  116. Re:The Future. Oooooh! by Wolfrider · · Score: 1

    --In my house it would be more like, "oooh eeee oooh aaah aaah... ting tang wallawalla bing bang..."

    "I told the witch doctor I wuz in love with you..." :b

    --
    .
    == WolfriderV6 == I'm willing to admit that *I just might* be wrong... Are you??
  117. Re:Window of opportunity:The Open Source Optical R by Anonymous Coward · · Score: 0

    I don't think your idea would work unless the system is massively parallel (well the Lenslet system appears to be parallel, but then again it's very much a special purpose device). In a nanosecond light propagates approximately 30 cm. The clock cycle of a modern 3 GHz processor lasts 1/3 ns, which means that light propagates only 10 cm in a clock cycle. Regardless of whether you go optical or electronic, you pretty much need to run on a miniaturized device to compete.

  118. DON'T... by Svante.1 · · Score: 0

    ...turn on the lights!!! I'm doing some calculations on my new optical processor...

    --
    .....:::[Svante]:::.....
  119. OS by POds · · Score: 1

    When can i expect the linux port? :)

    --


    Giving IE users a taste of their own medicine since 2005 - http://pods.-is-a-geek.net/
  120. Damn dude... by /dev/trash · · Score: 1
    in 1985 I was a Jet Propulsion laboratory.



    How come I never get the cool jobs. Was it painful? Were you able to switch back and forth at will?

  121. Re:Memory is irrelevant for this kind of "processo by quantum+bit · · Score: 1

    Okay, I know that's a Simpsons reference, but I can't quite remember which episode. Is it the one where Bart & Lisa go to military school?

  122. Re:Memory is irrelevant for this kind of "processo by gregor_b_dramkin · · Score: 1


    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.


    No, you don't.

    You can use FFTs for correlation. Read up on overlap-add, or overlap-scrap filtering. Note that correlation is really just a filtering operation by another name.

    This can give you immense performance gains. The scalability difference between traditional correlation and overlap-scrap is analogous to the scalability improvement that comes from going from a DFT to FFT.

    A rule of thumb for the efficiency crossover point for FFTs is N==64.

    In your example, you have 100 samples/chirp. This would lead you to an FFT size of 256 or 512 ('why' is an excercise left to reader). This is large enough to make O( Nlog(N) ) considerably better than O(N**2).

    Hmmm, now back to work.

    --
    You can never equivocate too much.
  123. Re:Memory is irrelevant for this kind of "processo by wowbagger · · Score: 1

    That is assuming you only want to know that signal existed, not the exact time point the signal arrived.

  124. Re:Memory is irrelevant for this kind of "processo by gregor_b_dramkin · · Score: 1

    It'd do both.

    However ...

    The point may be moot. Even if the processing was sped up 100x for your pulse length, it would probably still be too slow for radar pulse processing using general purpose hardware.

    --
    You can never equivocate too much.