Re: 25 Reasons to Aviod the SD-10 (was 15 Reasons to Aviod the SD-10)

Discussion in 'Digital Cameras' started by Paul Howland, May 9, 2004.

  1. Paul Howland

    Lionel Guest

    Kibo informs me that (Georgette Preddy)
    stated that:
    Why can't you show us any of your photos, Preddiot?
     
    Lionel, May 22, 2004
    #81
    1. Advertisements

  2. ....

    Laurence:

    To my mind it is more accurate to say that the Foveon sensor measures
    3.42 million items. Now, admittedly the Foveon chip uses a different kind
    of sensor than the Bayer sensor, but whether a sensor uses 1, 3, or
    2,000,000
    individual pieces, it is still only creating data for one point in the image
    plane.
    If the foveon sensor is really 3 times the information of the Bayer sensor,
    then my imaginary 2,000,000 receptors per pixel sensor ought to be so
    incredible that it blows our minds! In reality, my hypothetical sensor
    would
    produce virtually no additional improvement over the 1 or 3 receptor
    sensors.

    So, what we really have with the Foveon sensor is a 3.42 MP sensor where
    each sensor is, in theory, more accurate than the sensors in a Bayer array.
    In practice, the Foveon sensor does give a higher *Chroma* resolution than
    the Bayer sensor, but much of that advantage is given back due to less
    accurate Chroma samples.

    The trade off is simply: a chip with more, simpler, but less accurate
    sensors,
    or a chip with fewer, more complex, slightly more accurate sensors. As long
    as the Bayer chips can be produced with significantly more sensors than the
    Foveon, it looks as though the larger number of less accurate sensors is the
    better solution.
     
    Dan Wojciechowski, May 22, 2004
    #82
    1. Advertisements

  3. Paul Howland

    E. Magnuson Guest

    I have a monochrome only Miltary Policeman? Missing Persons? Member of
    Parliment?

    Why would any of this bother me?
     
    E. Magnuson, May 22, 2004
    #83
  4. Paul Howland

    Skip M Guest

    <snip>

    Might I point out that the Canon 1D is a 4mp camera with a 1.3 crop factor,
    so it has fewer pixels spread out over a bigger area? It ain't exactly
    apples to apples. Not to mention a few things were sacrificed on the altar
    of frame rate.
    And if what you say about the "clean up" of the 300D vs. the 10D is true,
    why isn't the 10D significantly noisier than the 300D, since they use the
    same sensor?
     
    Skip M, May 23, 2004
    #84
  5. Paul Howland

    Lionel Guest

    Kibo informs me that (Laurence Matson) stated that:
    Which Canon sensor, relative to what, & what's your cite for it?
    As a matter of fact, the 10D is generally held up as being the lowest
    noise DSLR on the market.
    Again, which sensor, & relative to what?
    Which sensor, & what comparison method?
    Got a cite, or any other sort of evidence for that claim?
    Yes. I have a 10D, & I normally shoot RAW. The only time I shoot JPEG is
    when I'm paid to do it that way.
    Again, got any evidence for that, or do you expect us to just take your
    word for it?
    Why not? If you have two otherwise identical sensors, but one has larger
    photosites than the other, the latter will be more sensitive, & less
    noisy.
    Will they? Why's that?
    No kidding...
    Steven's claim was: "The experts unanimously agree that the Canon CMOS
    sensors are the best on the planet". I was under the impression that he
    was referring to expert opinions on the quality of the image produced by
    those sensors. And yes, every review I've ever read that compares image
    quality between DSLR image sensors has said that Canons best are the
    best there is.
    Just out of curiousity, have you used a Canon DSLR more recent than the
    1D? The reason I ask is because the 1D is getting a bit long in the
    tooth. Canons newer DSLRs are much better than the 1D, in just about
    every respect.
     
    Lionel, May 23, 2004
    #85
  6. Paul Howland

    Lionel Guest

    It's 3 now: 1Ds, 10D & 1DmkII. ;)
     
    Lionel, May 23, 2004
    #86
  7. Paul Howland

    Lionel Guest

    Kibo informs me that (Laurence Matson) stated that:
    George is a pathological liar. He makes up technical details, links to
    photos & lies about how they were taken, lies about what kind of
    equipment he has, lies about being a professional photographer, lies
    about his name, makes up fake identities (hence our suspicion of you),
    etc. Even his name is fake. If George told me the sky was blue, I'd look
    out the window before believing him.
    By all means, please fell free to point out any misinformation you feel
    that I'm delivering. If you can show that I'm doing so, I'll be more
    than happy to acknowledge my errors.
    He's not explaining, he's preaching. Everyone involved in this
    'discussion' has at least a reasonable laypersons knowledge of the
    technology involved, & many of us are quite expert. I myself have
    designed a digital imaging system, & there are at least two other
    regulars in this group who're more expert than I am on this topic.
     
    Lionel, May 23, 2004
    #87
  8. Paul Howland

    Lionel Guest

    Kibo informs me that (Laurence Matson) stated that:
    Dozens of people have corrected the Preddiot's numbers literally
    hundreds of times. He goes with whichever set of numbers sounds best
    from post to post.
     
    Lionel, May 23, 2004
    #88
  9. Paul Howland

    Lionel Guest

    Kibo informs me that (Laurence Matson) stated that:
    It's not accepted by anyone but Foveon's marketing department & some of
    their users.
    A pixel represents a particular location on a grid. The colour depth is
    totally irrelevent in this context. The sensor could be monochrome, &
    it'd still be 3.4MP. It could be some future Foveon sensor with 18
    distinct photosite layers at each pixel, but the pixel resolution would
    still be exactly the same.
    Yes, but those 10.2 million items aren't pixels. They're R,G & B values
    for ~3.4 million pixels.
    Well you could, but you'd be incorrect.
     
    Lionel, May 23, 2004
    #89
  10. Paul Howland

    Lionel Guest

    Kibo informs me that (Laurence Matson) stated that:
    Heh.
    I wonder how many ex-Sigma users there are at the equivalent Canon fora?
     
    Lionel, May 23, 2004
    #90
  11. Paul Howland

    Lionel Guest

    Kibo informs me that stated that:
    Good luck trying to get George to admit that it's true, though. Maybe
    Laurence will succeed where we've all failed? Either way, I'm betting
    that watching him trying to pound some facts into the Preddiot's head
    will be pretty entertaining. ;)
     
    Lionel, May 23, 2004
    #91
  12. My witty ironic remarks were obviously completely lost on you... ;)

    -JP
     
    Jukka-Pekka Suominen, May 23, 2004
    #92
  13. Bart van der Wolf, May 23, 2004
    #93
  14. Paul Howland

    Lionel Guest

    Kibo informs me that (Georgette Preddy)
    stated that:
    ...you drooling with envy.

    Why do you lie all the time, Preddiot?
     
    Lionel, May 23, 2004
    #94
  15. Paul Howland

    Lionel Guest

    Kibo informs me that (Georgette Preddy)
    stated that:
    ....people as stupid as you - Sigma might have turned a profit on ther
    SD9s, instead of having to use them as landfill.

    And we're still waiting to see those photos you lied about.
     
    Lionel, May 23, 2004
    #95
  16. but since the Foveon doesn't actually sample the primary colours, but
    rather a gross approximation of them (requiring heavy
    post-processing)...

    GK
     
    grant kinsley, May 23, 2004
    #96
  17. Prove that statement with a reliable cite.

    GK
     
    grant kinsley, May 23, 2004
    #97
  18. Paul Howland

    Lionel Guest

    Kibo informs me that (Georgette Preddy)
    stated that:
    You're full of shit, Preddiot.
     
    Lionel, May 23, 2004
    #98
  19. Paul Howland

    Bill Funk Guest

    It's rather amazing that you continue withthis fantasy.
    Yes, the CCD/CMOS sensors are monochrome; so are the Foveon sensors,
    BTW.
    What each technology does is take that monochrome data, and, using
    algorithims, interpret that data to apply color.
    The idea that Foveon chips somehow magically pull *color* off the
    silicon is just wrong; it pulls *data* off in the form of electrical
    potentials, which have no color. They *represent* color. The firmware
    in the camera's electronics interprets those potentials, and applies
    color data to them, and then processes that data as a color photo
    file.
     
    Bill Funk, May 23, 2004
    #99
  20. Paul Howland

    Bill Funk Guest

    Let's start from a basic understasnding of a pixel.
    The word pixel is a contraction of Picture Element. Using Google (with
    a search criteria of "define: pixel", without quotes), you'll see what
    the accepted xefinition(s) of 'pixel' is.
    It's not, as you seem to be claiming, a point where data is pulled off
    the sensor; it is, instead, the smallest element of a digital image.
    The data that comes off *one* data point of a Foveon sensor is only a
    *part* of the smallest part of a digital image; IOW, the pixel, inthe
    case of the Foveon sensor, is actually made up of *three* data source
    outputs.
    This means that each pixel from a Foveon sensor has *three* data
    outputs used in it.
    This can not be streched into saying that each optical sensor position
    (which has three data taps) is actually three pixels, since the data
    from each data tap point is only 1/3 of the information needed to make
    up a pixel.
    (As a close analogy, see how the different points of color (dots) in a
    computer monitor are not, in themselves, pixels; rather, it takes at
    least three of those dots to comprise one pixel. Thus, one monitor
    with X number of dots does not have 1/3 the pixels of another monitor
    with 3X dots, if both are used that the same resolution.)

    What is "generally accepted" cuts little ice with me. Reality counts.
    Really? You've evidently not been reading what I write, then, because
    that's exactly what I've been saying.
    Gee, thanks.
    However, as I will continue to say (because it's true), the Sigma SDx
    cameras are *not* 10.2MP cameras, any more than my 3030Z is a 6MP
    camera, regardless of how many pixels I can interpolate the output
    into.
     
    Bill Funk, May 23, 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.