Forgot your password?
typodupeerror
Technology Hardware

The Arrival of Very Small Memory 175

Posted by Hemos
from the mixing-our-terms dept.
Roland Piquepaille writes "After the ages of DRAM and SRAM memories, is this time for nanotech memories? ExtremeTech says that "molecular memories" as well as memories based on carbon nanotubes are emerging. With these nanotech memories, several startup companies are envisioning future chips mixing logic, memory and reconfigurable computing elements. One of these promising startups is ZettaCore, which has built a prototype of a molecular memory designed to replace both SRAM and DRAM kinds of memories. These molecules, which are about 1 nanometer in size, are also self-assembling, meaning that they can be manufactured with existing equipment used in the semiconductor industry. This overview contains more details about the technology and includes a diagram of these molecules in a memory array."
This discussion has been archived. No new comments can be posted.

The Arrival of Very Small Memory

Comments Filter:
  • by Anonymous Coward on Tuesday March 23, 2004 @09:00AM (#8643761)
    Sorry, what were you talking about?
  • by Krik Johnson (764568) on Tuesday March 23, 2004 @09:01AM (#8643763) Homepage
    64 bit computers can have up to 18Tb of RAM, but with motherboard physical limitationss it iss not possible. Even with 4Gb dimms (which are expensive) your lucky to get more than 16Gb out of standard motherboards. With this technology, We will be able to break this barrier, and do wonderful things in small spaces.. I for one, welcome my 18Tb Dimm!
    • by millahtime (710421) on Tuesday March 23, 2004 @09:07AM (#8643796) Homepage Journal
      "18Tb of RAM"

      The problem I would see with this is the addressing of the ram. You couldn't use straight pins to do that high of number for addressing and what speeds would the buss work at. There are other limiting factors on how much ram you can really work with.
      • You couldn't use straight pins to do that high of number for addressing

        No, you have to bend the pins slightly before you insert the DIP.

        Seriously though, what the hell are you talking about. If you use 64 bits for addressing you get 1.84*10^19 addresses. (18 million terabytes, not 18 terabytes).

        The current implementations use 48 bit addressing, which can address something like 250 TB of RAM.
        • but don't you have to divide the bits (by 8) to get the bytes?
          • Re:I'm not sure (Score:2, Informative)

            by PD (9577) *
            In the old days, when the address was put on the bus, you had a bank of 8 memory chips that all read the same address off the same bus. Each individual memory chip would put the bit for that address on its data out line, which represented the 8 bit number at a particular address in memory. That's right, every time you read a byte from memory, each bit came from a different chip. Today, the packaging is different, but the concept is the same.
          • but don't you have to divide the bits (by 8) to get the bytes?

            If memory were bit addressable, you would be correct. However most modern machines are byte addressable. That means that each memory address refers to a full byte.

            It is perfectly possible to build machines that are only word addressable, where a word is 32 bits or 64 bits, or even larger. The advantage is that you can address more memory with a given address size. 32 bit words means address size * 4 bytes, 64 bit means * 8 bytes. The disadvan
      • yeah, like how big an opcode is.
      • by Odin's Raven (145278) on Tuesday March 23, 2004 @12:12PM (#8645500)
        The problem I would see with this is the addressing of the ram. You couldn't use straight pins to do that high of number for addressing...

        Do I detect the foul stench of RDRAM's corpse rising from its grave? (A serial memory bus would certainly help address the pin-count issue.)

        I can just see the future (assuming Rambus waits for "Talk Like a Pirate Day" to pounce):

        Rambus: Avast, ye scurvy memory-lovers, and prepare to hand over all yer sparkling treasure. We be the Pirates of Rambus IP, and we're here to double yer prices, scuttle yer standards committees, and rape yer sheep.

        Flunky: (...whisper whisper whisper...)

        Rambus: Errrr...rape yer RAM.

    • by DrEldarion (114072) on Tuesday March 23, 2004 @09:10AM (#8643809)
      Sheesh, what the hell would you do with 18Tb of RAM in a desktop computer?

      "18Tb of RAM should be enough for everyone!"
      • by TwistedGreen (80055) <twistedgreen AT gmail DOT com> on Tuesday March 23, 2004 @09:17AM (#8643849)
        Mental simulation. Synthetic intelligence. Your computer would be powerful enough to not only do flat speech recognition, but would be able to have its own natural language engine... all processed in real-time.

        Sweet.
        • by selderrr (523988) on Tuesday March 23, 2004 @09:40AM (#8643974) Journal
          eum, I don't want to disapoint you, but none of these is currently RAM bound. Current connectionst models require far more CPU power than memory to keep all nodes updated. Real-time is a distant future. Even non-realtime AI is currently more stupid then my 3 month old daugther.
          • by Anonymous Coward
            Even non-realtime AI is currently more stupid then my 3 month old daugther

            What did you just call your daughter?
          • Your 3 month old daughter is vastly more intelligent than any hardware humanity has made. The problem is the percieved lack of intelligence due to limited physical capacity. Just wait a few more weeks and that intelligence will be rolling/crawling around and pulling itself up on furniture. After that, you will experience a form of sentience that makes anything like the juvenile musings of Clive Barker and Stephen King laughable.

            Humanity fears biology because it's messy in ways you can't escape...and it's i
          • Dude that's rough, calling your daughter less stupid.
          • Well, connectionist AI is certainly a feasible approach...given enough hardware, and the right set of connections...but I'm not convinced that it's the most efficient approach.

            Still, nobody has anything working yet, so all hats should be in the ring.

            That said, I think that connectionist approaches are practically guaranteed to be more CPU intensive than other approaches, and if more than one approach can work, which I consider aprobability, then a higher-level model might need more memory than CPU (at th
        • Your computer would be powerful enough to not only do flat speech recognition, but would be able to have its own natural language engine... all processed in real-time.

          This isn't so much an issue of memory, as processing power. Some good DSPs would be more helpful here than terabytes of memory.

      • Sheesh, what the hell would you do with 18Tb of RAM in a desktop computer?

        Ever see "Weird Science"? Virtual woman!!!!

        But with that much ram, she'd be even better at remembering all of my fuckups than my real girlfriend.

        LK
      • I'll settle for an 8Tb GeForce >:>, and a 64-Gig RAMDisk on my PDA/Smartphone.

        I'm not greedy.. ;)

      • Run Netscape on Windows?
      • By the tim e I can afford it in a desktop, it will probably be a few TB too little to run Windoze...
      • I ran a huge BBS on an Atari ST a "few" years ago. I had a stunning 520Mb of memory (that's storage memory - SCSI even) for all the files anyone EVER could put there, or something.

        My current desktop computer has 1Gb of RAM. If you had asked me what on earth I would need that much RAM for back then .. well .. .. I don't think I would've answered that I would need it for stuff like Freenet (using ~140Mb right now) or Opera (preview 3 is an extreme memoryhog, using 253Mb right now) etc.

        So, I'm quite sure I'l
    • by silas_moeckel (234313) <silasNO@SPAMdsminc-corp.com> on Tuesday March 23, 2004 @09:23AM (#8643876) Homepage
      I beleive 4GB Dimms are as large as they can go do to limitations in the addressing lines (pins) at 32 (havent checked this might be wrong). So untill a new form factor is released thats what we are stuck with. I would differ on the max expandability most MB's I have seen are running 4 DIMM slots per proc. I beleive this is the max they were designed to handle on there embeded memory controler. I am speaking of the Opterons of course. The PIV's currently have chipsets supporting piles and piles of DIMM slots at least 16 last I saw possibly more (64GB is the current max and I think they did that with 2GB sticks). So with these numbers and 4GB dimms thats 32GB in a 2 way Opteron setup and 64GB on an intel. The nice thing is the 8 way Opterons would be running 128GB max though thats a massive motherboard to support that.

      Overall I dnt see this tech realy reducing the size of the ram on pin count alone more it will reduce the power consumption and profile of the dimms what increasign the potential density of a new replacement for DIMM's.
      • by Anonymous Coward
        DDR supports 1Gb chips. They can be used to make 4GB DIMMs.

        DDR-II makes possible to use 4Gb chips. They can be used to make 16GB DIMMs. However, it will take a few years before manufacturing technology improves enough to make manufacturing of 4Gb chips possible.
    • "640Gb software is all the memory anybody would ever need on a computer." -Gill Bates on small memory
    • I don't know what boards you're looking at for large memory configurations, but 24GB [amdboard.com], 20GB [tyan.com] both use a maximum 2GB sticks for those. (Having more DDR slots than PCI slots is kinda strange looking)

      Now, you could argue that these aren't standard motherboards, but then again, what 64 bit CPU motherboard is? For next year or two, I don't expect to be hitting the 20GB memory limit... ;)

    • 64 bit computers can have up to 18Tb of RAM, but with motherboard physical limitationss it iss not possible.

      yep, yep.. Reminds me of when the MacII first came out. Based on the 68020, it would be, in theory, capable of addressing 2GB of ram. (one bit was used to switch between RAM and I/O space) I did some napkin math and figured that you could camoflage a 2GB memory unit as a desk. The memory would fit in the lid of the desk, with one pillar being a cooling unit, and the other a 16Kilowatt power sup

    • oh, and a 64bit CPU should be able to access 16EXAbytes, not just Terabytes.
      • $ units 2^64 tera

      • * 18,446,744
      (commas mine)
  • Right (Score:4, Interesting)

    by Operating Thetan (754308) on Tuesday March 23, 2004 @09:02AM (#8643768) Journal
    With these nanotech memories, several startup companies are envisioning future chips mixing logic, memory and reconfigurable computing elements

    Do they mention if the CPU and motherboard manufacturing companies care? Technology succeeds because of marketing, not because it's innovative or high quality-witness Betamax,
    • Re:Right (Score:1, Interesting)

      by Anonymous Coward
      What are you nuts? Sorry but you sound like you have no clue what your talking about.

      Stuff succeeds because of MARKETING?

      Technology drives the industry.

      So if your correct then if I can out with a terabyte memory module that cost a third of a DDR ram 4 gig module people will still by the more expansive RAM if I find a pretty enough box and advertise on the SuperBowl?!?

      Marketing is just one part of a company and the products they market, everybody works together to create a successfull product.

      Don't make
    • I imagine that this tech would be f*ing great in Intel's System-On-a-Chip (I forget the name).
    • Technology succeeds because of marketing, not because it's innovative or high quality-witness Betamax

      Well, you could have a point, except we're not really talking about some consumer media here. if this thing can work, it'll all come down to the manufacturing cost. just look at it, small (can it get much smaller ?), fast, non volatile and doesn't require new fabs.
      if on top of that it can get cheap enough, i don't see why it wouldn't appear in all kind of electronic devices, if not all of them.
      well, may
    • Re:Right (Score:3, Insightful)

      by tekunokurato (531385)
      Whether you're a troll like the other guy says or not, there are thousands of companies that employ engineers to develop technology and then licence the technology to companies like motorola, intel, etc. The technology may require some degree of marketing to achieve critical mass usage, but that's irrelevent because if there's enough economic value (which this project practically guarantees, eventually) a larger company will either license it or develop its own.
    • "Do they mention if the CPU and motherboard manufacturing companies care?"

      CPU and motherboard companies probably would not be able to use this until ten years from now. CPU companies may be quietly investigating this already because R&D to production on silicon processes has a long lead time. It won't concern motherboard manufacturers until chips are already sampling.
  • ideal memory (Score:1, Insightful)

    by Anonymous Coward
    this seems to bee ideal, looks like it will need little power for keeping data in memory.. however it might be terribly slow or degrading in time, the article is kind of sloppy on the details of this. anyone?
    • We're talking about molecular physics here - speeds are very fast. As for degradation, as it's molecular, I seriously doubt that's going to be an issue. As they said, the molecules are stable, which is half the battle won :-P

      Of course, I'm not anything even remotely like a molecular physicist.

      • yeah... the amount of energy that will be needed to break a covalent bond will be immense... carbon especially has a strong molecular bond to itself at those scales.
  • by Space cowboy (13680) * on Tuesday March 23, 2004 @09:04AM (#8643777) Journal
    Xilinx [xilinx.com] have silicon with embedded PowerPC processors, BlockRam (chunks of pre-generated SRAM) and huge swathes of FPGA cells and interconnect. The chips have other abilities too - built-in 18-bit multipliers and communications channges (10 Gbps/channel, 20 channels!). All very cool stuff. Very expensive too :-(

    I'm sort of surprised there aren't more FPGA-hackers than there appears to be. It's not hard to learn verilog (very similar to C), and despite what most FPGA designers will tell you, as long as you keep your mind focused on 'everything happens in parallel', a decent programmer can produce good FPGA code too. The start kits (300,000 gates, about enough for a hardware JPEG core and maybe a network MAC) are cheap (100 or so), and designing a processor [fpgacpu.org] is a pretty simple operation, and immensely gratifying :-)

    Just my thoughts,

    Simon

  • it seems well and good, but i for one won't be convinced until i see them in the palm of my hand.

    under my electron microscope....
  • by Anonymous Coward
    My memory is small enough, thank you.

    Now... what was I doing?
  • by zazas_mmmm (585262) on Tuesday March 23, 2004 @09:06AM (#8643789)
    Nanotech memory is very exciting, but there's a lot more than the technology itself that determines whether it's the next big thing. So far all I see is a weblog with some basic diagrams of how it works and some serious brochureware at Zettacore.

    Not to state the obvious, but it will take low manufacturing costs, industry willingness, consumer demand, and a whole lot of marketing before this or any other revolutionary changes become de facto standards.

    Better, smaller, faster, is no match for cheaper, more accessible, and well-marketed.

    • I do agree with most of your points, however I don't think you should doubt consumer demand:).
      I see every new program require more memory, more porcessing power. More and more information is processed by computer, which does require memory. Even if we could momentarily reduce or maintain memory needs through optimisation of the programs. In the end we will need better, faster and smaller memory. Wheter that is now or not I don't know but in the end the demande will be there.

      Additionally, if this memory c
  • by sdkramer (411640) <seth@NOSpAM.sethkramer.com> on Tuesday March 23, 2004 @09:06AM (#8643790) Homepage
    Does this mean I'm gonna start getting spam about how HU6E my memory is? I'm starting to get memory envy.
  • Non volatile? (Score:5, Interesting)

    by MrIrwin (761231) on Tuesday March 23, 2004 @09:06AM (#8643791) Journal
    Unlike SRAM, which requires a charged state to be maintained, and DRAMS which reuire continuous refresh, these devices would appear to permanently change a molecular structure.....i.e. they would seem to offer high speed read write non volatile memory.

    This could not only increase RAM but mean we have computing devices with just one big memory pool...no Flash, no Disk, no CD, no DVD.........

    Can I order mine now please?

    • Wow, think of it! 10 TB solid state, ultra fast disks! Whoa. I'm getting dizzy here.
    • Actually, I still see computers having separate memory spaces. However, I think you'll see storage defined as 'portable' or 'integrated' instead of 'volatile' and 'non-volatile'.
    • Considering that computer systems crash, I would expect there to still be a distinction - between the memory that gets wiped on a reset and that which doesn't.

      Imagine if, when you reset your computer after a crash, whatever caused it to crash was still there. Until someone works out a reliable way of automatically recovering crashed programs and OS's, wiping memory will be necessary. And I don't think having to do a full reformat every time Windows crashes would be fun...

  • by davegaramond (632107) on Tuesday March 23, 2004 @09:07AM (#8643799)
    Hm, doesn't sound as good, does it?
  • by jellomizer (103300) on Tuesday March 23, 2004 @09:09AM (#8643804)
    Good old progress making something small and making it smaller then integrated with other parts. This can have impact in a ton of areas including smaller and lighter laptops, PDA, and PCs, perhaps a future where you can mix Xerox's Electronic Paper with this to offer interactive News Papers. As well as a lot of cool stuff. But of corse the will be people who will use it for evil Like a chip that is implanted in Tin Foil that can see where you are. And how you are using tin foil. Or Devices attached to clothing that can all you to be tracked and record everything you see and say. or a Beowulf cluster of these the size of a PC. Oh the horror! Just remember when they start using these chips for evil please remember that you recommend them first!

    Some times there is truth in sarcasm, other times there isn't hmmm.
    • by Morosoph (693565)
      Like a chip that is implanted in Tin Foil that can see where you are.

      Noooo! My tin foil hat might be chipped!

  • by revolvement (742502) on Tuesday March 23, 2004 @09:12AM (#8643818)
    No one will ever need less than 640mm of memory
  • by stecoop (759508) on Tuesday March 23, 2004 @09:13AM (#8643824) Journal
    Very cool but memory chips aren't really gigantic. I would be more interested in speed or parallel memory access.
  • ... ST Microelectronics already supply devices that mix programmable logic, memory and IO from their Programmable System Device [st.com] range. But there is something of a reluctance for commercial designs to incorporate them because they're single source components. Why risk being unable to make your product in the future because you've used a specialised component in your system which has gone obsolete - especially when there's a plethora of available direct drop-in replacements for a discrete solution (EG separ
  • by mkro (644055) on Tuesday March 23, 2004 @09:16AM (#8643843)
    These molecules, which are about 1 nanometer in size, are also selfassembling
    and at night, after you turn off the PC and go to bed, they swarm out of your computer, heading for your pillow to EAT YOUR BRAIN.
    • You turn your computer off?
    • Michael Chrichton be damned. "Self-assembling" simply means that the molecules arrange themselves in a useful pattern when they fall out of solution. It used to be called "chrystallization", but the marketing people must have decided that it would be cute to invent a new word for making salt. This has absolutely nothing to do with Eric Drexler's assemblers or nanotechnology (at least as he originally defined it). And one final point: even Drexler's assemblers are only machines. THEY ARE NOT ALIVE!!!! Damn i
      • by achurch (201270)

        And one final point: even Drexler's assemblers are only machines. THEY ARE NOT ALIVE!!!! Damn it! They will not eat your brain any more than your feature-filled VCR will.

        We'd have a much more intelligent populace if it wasn't for the brain-eating features on modern TVs and VCRs . . .

  • by femto (459605) on Tuesday March 23, 2004 @09:18AM (#8643852) Homepage
    From their "details about the technology."
    The company said it has designed molecules with eight states, potentially offering a 4-bit-per-cell density.

    I hope their research is better than their PR. Or maybe their technology really is unique!

  • I wonder... (Score:3, Insightful)

    by rkoot (557181) on Tuesday March 23, 2004 @09:20AM (#8643860)
    whether these new technologies could change the way a modern computer works.
    I mean, if the chips become so much smaller, it's easy to see the capacity of i.e. Ram chips will reach levels unimaginable now.
    But how are these bits gonna be addressed ? you need *lots* of pins, and how to connect those pins to the logical layer ?
    I guess motherboards, processors and such need to be radically redesigned to be able to use this new technology.
    How long would it take before mainstream mobo's use other (like i.e. photons instead of electrons) than conventional techniques ?

    just curious

    r.

    • Re:I wonder... (Score:4, Insightful)

      by millahtime (710421) on Tuesday March 23, 2004 @09:32AM (#8643924) Homepage Journal
      "whether these new technologies could change the way a modern computer works."

      Nanotech sure will change the way a computer works. If you can have atoms doing the work you have gates doing now you can fit a lot more on a chip. They can manipulate gates at the molecular level now, the problem to be solved is between that tiny world and our big interfaces.
      • Having atoms do the work Billionaires are doing now... manipulating billionaires at the molecular level...

        This newfangled nanotech thing is even better than I thought! to bad it won't solve the problem between our big interfaces and Billionaires but I guess that's just the next step...
    • Re:I wonder... (Score:2, Interesting)

      by MrIrwin (761231)
      I envisage just swarthes of tiny black boxes interlinked by a grid of channelised synchronous serial links.

      I am not a visionary, BTW, this is more or less how big digital switches in the telecoms industry works. We are just talking about scaling down from board level to chip level.

      IMHO, the biggest headache to overcome in the chip industry will not be how to package and interconnect, but how to incorporate "outside world" buffers on the edge of these devices which are powerfull enougth to pump the data,

  • Surgical implanets to repair lost memory cells in the human brain. Was there not a film about this a few years ago called Johnny something?
    • Surgical implanets to repair lost memory cells in the human brain. Was there not a film about this a few years ago called Johnny something?

      Y'know, for the life of me, I can't remember...
    • Re:NExt step (Score:2, Informative)

      by carm$y$ (532675)
      Johnny something

      Mnemonic. Johnny Mnenonic. Tough word, isn't it? :)
      And it wasn't about lost memory cells, it was about selling storage space in your
      enhanced brain...
  • by digrieze (519725) on Tuesday March 23, 2004 @09:36AM (#8643948)
    Except for embedded devices like cell phones and pdas, this won't change much. The memory density may go up, and since the chips are thinner the heat problem may improve, but the size of system chips won't change.

    The reason is simple, human fingers and hands aren't going to shrink. SDRAM cards are about as small as most people can handle comfortably. SDRAM chips for CPUs work very well not at holding chips but at being easy to install and make positive contact with a large number of contacts on a relatively small edge. The design factors for these things are many, the chips they carry are only a single one of them.

    I suppose someday it'll be theoretically possible to put that monster gamer machine in a thinline dress watch, but as they found with the "databank" watches the limitations are the input/output devices average people can comfortably work with, not electronic capabilities.

    • SDRAM cards are about as small as most people can handle comfortably.

      I'd be quite satisfied popping 24TB of RAM into a machine using a part the size of a slim watch battery. I think my grubby paws could handle that.

    • The reason is simple, human fingers and hands aren't going to shrink. SDRAM cards are about as small as most people can handle comfortably. SDRAM chips for CPUs work very well not at holding chips but at being easy to install and make positive contact with a large number of contacts on a relatively small edge. The design factors for these things are many, the chips they carry are only a single one of them.

      I think you're missing something... SDRAM is not designed to be handled on a day-to-day basis (or ev
  • by G4from128k (686170) on Tuesday March 23, 2004 @09:36AM (#8643951)
    The biggest challenge to this type of tech is creating complex large-scale patterns. Its one thing to create a fully regular "crystal" of 1-bit memmory cells, its another to create the highly irregular, specific, chip-spanning structures of a CPU. If we are going to make complex nanocircuits, we need a way to ensure that the right bit gets connected to the left bit.

    I wonder if a better process would be to adapt the proteosynthesis process for creating micro-polypeptide clusters that are circuit elements with highly specific binding sites for self assembly. A DNA sequence would encode an mRNA sequence that is passed to a ribsome-like micro-factory. An alphabet of tRNA units would carry heavily modified amino-acids and provide both the electrical and structural of properties of the polypeptide. Different polypetides might make transistors, autonomous clock circuits, chemical-to-electrical battery subunits, wires, tees, etc.

    Part of the DNA sequence would encode binding sites that are highly specific. Each electrical component would have a unique code on each terminal that only binds with the component that it connects to in the circuit. By labelling all the terminii of the components with these specific binging patterns, you the potential for self-assembly. To make a complex circuit, you make separate batches of each component, then mix the batches together and they self-assemble into the circuits. Thus, a soup of appropriately labeled transistors and wires would self-assemble into a soup of full-adder circuits.

    The use of larger-scale binding sites would enable hierarchical self-assembly of self-assembled micro-components (e.g., a soup of 1-bit full-adder circuits might self-assemble into a 8-bit full-adders, or 8-bit full-adders might bind to a gated accumulator registers, etc.)

    I doubt this technology would let you create a 64-bit processor - the binding-site combinatorics get too ugly. But it might let you create RAM, RFID circuits, or small CPUs (e.g., the Intel 8080 only needs 6000 transistors)

    BTW, my post is a modified dup of a previous post of mine [slashdot.org], but I thought it might be relevant.
    • Doesn't it occur to anyone that making a super fast self assembling potentially AI cabable organic based computer might be a bad idea? I can just see those zombie PCs running around now - "Need more brains - must eat brains..."

      The only thing that will save us is if they're running windows.
    • Maybe it would be better program DNA to grow an organic computer instead. Of course, that would require the reverse engineering of how DNA encodes its data. Once understood, I'm sure DNA compilers wouldn't be too far off in the distant future. So rather then turning source code into binary, you could turn source code into DNA. From there, you grow whatever it is you programmed.

      And yes, the applications from doing this virtually unlimited.

  • How is small memory a new thing? I had small memory on my VIC-20.

    Obligatory 80s microcomputer fanboy reference: anyone else remember the adverts for the Dragon 32 and its "massive 32Kb memory"? The VIC's 5Kb is the smallest amount I've had to work with, but only because I managed to avoid the ZX-81.

    It certainly makes you think about browsers whose publicity material describes them as having a "small footprint", which then turns out to mean no more than ten megabytes. Or two thousand VIC-20s, if you wa

  • Months from now Extremetech gets this memory to test, but in a tragic testing mishap they set it to half of the actual rated MHz in the motherboard bios, and don't bother to run CPUID to verify the memory speed making the same mistake they have done with their latest couple of Athlon 64 FX tests.
  • This could mean a lot to industry. These chips can be constructed cheaper once mass production starts, can hold more bits, which means that machines with very large memories can be built cheaper than what is currently available today, and more memory can be next to CPU for caching reasons.

    What I like is that games and simulations can become more complicated because of the increased memory.

    Has anyone heard if it is faster or what the power consumption is going to be? I would assume that the power c
  • It's not about smaller, it's all about bigger.

    Making a megabyte of SRAM have a smaller footprint won't change much in the current world of microelectronics.

    Making a megabyte-sized SMD hold a gigabyte, however...
  • Bandwidth? (Score:3, Insightful)

    by palad1 (571416) on Tuesday March 23, 2004 @10:54AM (#8644659)

    I could not grep any 'Bandwidth' occurence in the article.

    What's the use of having a 4mm 18Tb chip if the bandwith still is 1Mb/s?

    As they used to say: Never underestimate the bandwith of a truck full of tape drives

  • "...are also self-assembling, meaning that they can be manufactured with existing equipment used in the semiconductor industry."

    OK, where are there existing semiconductor plants using nano-tech self-assembly techniques? That's an odd statement, implying the current UV light and mask etching equipment could just as easily do nano self-assembly.

    • > OK, where are there existing semiconductor plants
      > using nano-tech self-assembly techniques?
      > That's an odd statement, implying the current UV
      > light and mask etching equipment could just as
      > easily do nano self-assembly.

      First of all, "self-assembly" is not "nano-assembly", it is just chrystallization. The process is chemical in nature and would require similar equipment to that of circuit board etching. Second, mask etching is still required to draw the address wires on the silicone subs
  • If this stuff is in standard use when Longhorn comes out, MS is going to really have to make some changes to it if they want it to be as big a memory hog as previous versions of windows were.
  • Heck, I've had "very small memory" myself since about... um, since... wait a minute...

    What was the question again?
  • Its interesting that nobody mentioned reconfigurable memory elements in this thread. They offer, IMHO, one of the most exciting potential advances in hardware.

    Basically, the idea is that you put a bunch (from a few dozen to thousands) of very simple arithmatic units integrated right into memory. Inside memory, there is an enormous amount of bandwidth available at the sense amps --- several terabytes per second on current memory chips. These processors could all work in parallel, unfettered by memory bandwi
  • Now I can be a bear of little brain, and not suffer.

In a consumer society there are inevitably two kinds of slaves: the prisoners of addiction and the prisoners of envy.

Working...