A fun game to play with George Preddy

Discussion in 'Digital Photography' started by Lionel, Jan 19, 2004.

  1. Could it be possible that the illustrations used in the patent paper are
    more schematic (side by side) than realistic? From the illustrations, I
    could not yet derive the way they collect the electrons from different
    penetration depths, but perhaps I should do a more thorough analysis of the

    Bart van der Wolf, Jan 22, 2004
    1. Advertisements

  2. http://www.x3f.info/technotes/x3pixel/pixelpage.html

    Foveon invented the 3 CCD camera. That is, long before they claimed
    radically improved quality, with the introduction of the Pro 10M sensor used
    in the SD9/10.
    George Preddy, Jan 22, 2004
    1. Advertisements

  3. Braindead Preddy, Jan 22, 2004
    George Preddy, Jan 22, 2004
  5. Then the remaining 75% is guessed.
    George Preddy, Jan 22, 2004
  6. Braindead Preddy, Jan 22, 2004
  7. It's the truth, and it's real smart.
    George Preddy, Jan 22, 2004
  8. .... intelligently interpolated form surrounding pixels.
    Braindead Preddy, Jan 22, 2004
  9. Its a lie. They are intelligently interpolated from surround pixels
    producing far better results than capable with the random colour with Foveon
    Braindead Preddy, Jan 22, 2004
  10. I've examined it in detail, it is perfectly executed using concentric wells.
    YIf you can't mentally project 3D spatial concepts very well, don't worry,
    keep trying, you'll eventually get it.
    George Preddy, Jan 22, 2004
  11. So... the most awsome mind in digital imaging gets the last word, Caver Mead
    just ended this debate.
    George Preddy, Jan 22, 2004
  12. ... very badly ..
    ....then you will probably claim the Sd9/10 is 10.2 MP, instead of the
    accurate 3.4 MP, which is the exact size of the output, and exactly the
    number of spatial samples taken by the sensor.
    Braindead Preddy, Jan 22, 2004
  13. .... dishonest ..
    .... by convincing George with his lies.
    Braindead Preddy, Jan 22, 2004
  14. Film measures then records full color. So does Foveon. Bayers measure a
    tiny amount of optical data, then add digital guesses to fill in the
    remaining 75% of what gets recorded.
    George Preddy, Jan 22, 2004
  15. Foveon attempts to sample colour in a stacked silicon sensor which doesn't
    sample as good as Bayer.
    .... more ...
    .... intelligentrly interpolates the full colour values form the surrounding
    Braindead Preddy, Jan 22, 2004
  16. Lionel

    Lionel Guest

    That is a lie.
    Lionel, Jan 22, 2004
  17. Lionel

    YoYo Guest

    Pro for children and new to photography maybe but not for Photographers!!
    YoYo, Jan 22, 2004
  18. Yes, all of them. That's where the "X3" notation was lifted from. Foveon,
    being a famous maker of prism split 3 CCD camera before the Foveon Pro 10M
    (dramatrically increased quality over 3 CCD designs, according to their
    literature), erroneously thought that the "X3" notation would immediately
    ring a bell with informed digital buyers, since 3 sensor cameras are
    advertised as "XXXX x YYYY x 3."

    But their "X3" nomenclature's 3 CCD conotation vastly over estimated the
    sophistication of the average DSLR buyer, it was a monumental marketing
    blunder. No one got it.
    George Preddy, Jan 22, 2004
  19. Note the date, this is long before the Foveon X3 Pro 10M was invented...

    Enter some new technology. Companies like Foveon have been producing
    studio-quality digital cameras using CMOS technology - almost the same wafer
    process that makes your Pentium ICs. In a CMOS sensor, each pixel can be
    separate from the others. One dead pixel can be isolated, and interpolation
    done on that one pixel. It doesn't kill the whole column the way the CCD
    does. Also, the drive electronics are much simpler than a CCD. In CMOS, you
    can also make a separate amplifier for each pixel, which allows better
    tradeoffs of sensitivity, dynamic range, blooming control, etc. Foveon's
    cameras are enormously expensive, using separate CCDs for red blue & green
    and making truly phenomenal pictures, but the technology is working
    downstream. Many PC cams now use CMOS sensors, since the CMOS sensor can be
    built in a fab that was designed for older-generation processors and digital
    logic. (New processors are below 0.15 micron feature sizes - CMOS sensors
    use 0.35 to 0.5 micron devices)

    There are still a lot of bugs to work out in CMOS-based sensors for
    "mass-market" quality cameras, but those bugs are disappearing quickly.
    Using a CMOS device, the defect rate of the process will become much less
    important - instead of 1-2, or sometimes no good sensors per wafer,
    manufacturers should be able to get several. A large format sensor would
    still cost a few hundred dollars, but would be achievable, if the consumer
    market existed to buy them.

    (Note: That last statement turned out to be an under statement. Foveon
    chips are very expensive nowadays. NSC said they could build and market
    digital cameras for less than the cost of film disposables. The 5M probably
    runs about $25.)
    George Preddy, Jan 22, 2004
  20. It can be, as long as you use the same definition for what you are
    comparing. Using the same standard, Foveon will always beat Bayer by at
    least (10.3/6.0) in B&W, and (13.7/6.0) if color is involved.
    George Preddy, Jan 22, 2004
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.