Future of Computing

Posted:
in Future Apple Hardware edited January 2014
So I was reading on Slashdot today:



Quote:

ThinSkin writes "Photon processors that transmit data via light, not electrons, are slated to enter production in mid-2006, ExtremeTech reports. Headed by a UCLA professor and a Nobel Prize winner, startup Luxtera claims that its optical modulator clocks in at 10-GHz, tens times that of Intel's optical modulator researchers talked about last year. Since the optical module exists as its own entity, it will require a standard CMOS processes to integrate the optical waveguides. Luxtera has worked closely with Freescale Semiconductor to develop this technology."



Then I thought for a second ... wait ... Freescale ... Intel behind on technology ... could this be a big break for Apple technology. Then I read on the link to the oritinal artical that:



Quote:

Freescale has taped out several engineering samples of the optical technology, including a chip, one side of which includes the optical interface built in. The sample chip use a 130-nm SOI process, the same technology used to fabricate the G4 microprocessor. Part of Luxtera's job has been to develop silicon libraries, the files used to design the photonic chips in the same way other libraries serve as the blueprint for making more conventional semiconductors.



Sounds very interesting. Any thoughts?

Comments

  • Reply 1 of 6
    hirohiro Posts: 2,663member
    Give it an interested wait and see. It's easy to make small numbers of components run at those speeds, but much harder when the logical complexity of a modern CPU is thrown in.
  • Reply 2 of 6
    thttht Posts: 3,929member
    If the wiring or interconnects, as opposed to the transistors, in a processor is optical rather than electrical, I could see processors being clocked a little bit higher (20%?) since part of what holds back or limits processor clock speeds is wire or signal delay.



    With the first implementation of this technology being a chip-to-chip bus, it could also provide better processor performance by providing a path for higher memory bandwidth. But buses aren't necessarily the bottleneck here either. Memory chip performance looks to be lagging everything else for the forseeable future.
  • Reply 3 of 6
    hirohiro Posts: 2,663member
    Well optical signalling through fiber-optic media is roughly 2/3 the speed of light in a vacuum, which is slower than the propagation of electrons through copper wire. I don't really see how channeled optical interconnects can make "wire" delays shorter unless you went medialess. And that adds a whole host of problems to complex circuits.
  • Reply 4 of 6
    thttht Posts: 3,929member
    Quote:

    Originally posted by Hiro

    Well optical signalling through fiber-optic media is roughly 2/3 the speed of light in a vacuum, which is slower than the propagation of electrons through copper wire.



    Really? I will have to check that out.
  • Reply 5 of 6
    cubistcubist Posts: 954member
    Optical interconnects should also reduce noise and crosstalk, possibly making higher speeds possible even if there is greater delay.



    But Ben's got a point, it is interesting that it's our old friends at Freescale who are doing the work...
  • Reply 6 of 6
    hirohiro Posts: 2,663member
    Quote:

    Originally posted by THT

    Really? I will have to check that out.



    Yep. The density and light limit-speed differences in various media are responsible for refraction. Think prisims, the actual speed differences even vary with wavelength. So it's actually not quite so surprising. The 2/3 number is not hard and fast, pretty much a back of the envelope starting point used for fiber-optic network designs that gets specifically hammered out once actual media and transmission wavelength decisions are made.
Sign In or Register to comment.