Pixel counts in P&S's should have been held to 5 megapixels

Discussion in 'Digital Photography' started by Rich, Aug 9, 2008.

  1. Rich

    Rich Guest

    Seeing way better highlight retention from a 9 year old, 1.3 megapixel
    P&S compared to a new 10 megapixel model convinced me of that.
    The marketers of this junk should be shot.
    Rich, Aug 9, 2008
    1. Advertisements

  2. Rich

    Scott W Guest

    since the sensors are linear, highlight retention is really a fuction
    of expsoure, over expose any digital and you loose the highlight.
    With the exection of some of the Fuji cameras with different photo
    cell sizes.

    Now the caveat, many cameras have a bit of extra headroom in the raw
    files, and if shooting raw you can better avoid clipping the
    highlights. And the cameras with larger pixels tend to have more
    headroom in the raw files.

    I don't mind that there are P&S cameras a whole lot of pixels, but it
    would be nice if there were a few good ones with a pixels count in the
    3-5 MP range.

    Scott W, Aug 9, 2008
    1. Advertisements

  3. Rich

    SMS Guest

    Buy a used G2 or G3.
    SMS, Aug 9, 2008
  4. Rich

    SMS Guest

    Once the megapixel (and LCD size) race started everyone was compelled to
    enter because so few users understand the physics of camera sensors and
    automatically assume the higher resolution the better. Blame the
    ignorance of the consumer, not the marketers that simply want their
    companies to stay in business.
    SMS, Aug 9, 2008
  5. Rich

    John Sheehy Guest

    Screw your head off, completely, wait 15 seconds for all capacitors to
    discharge, and then hopefully you have already asked someone else to
    screw your head back on. If they do, now consider this:


    The more noise there is in each pixel, relative to saturation, the more
    the mean level has to rise above the clipping point before all pixels are
    clipped. Admittedly, this effect is small in pixels 2 microns and
    larger, but it would be very significant in higher pixel densities.

    Now, get this: while the camera companies were increasing pixel density,
    they were also increasing quantum efficiency, so that by the same
    standard used for giving ISO ratings to DSLRs, compact cameras now have
    base ISOs of 160 to 200. Being that the typical consumer who thinks he
    knows a bit more than the rest is actually ignorant and thinks there's
    something wrong with a camera that doesn't have ISO 100, they call ISO
    200 ISO 100 and meter for 100, leaving one stop less headroom. It also
    leads to less noise, as the camera is "exposing to the right" for you.

    You can not leave modern compacts on 0 EC and expect any kind of extended
    headroom, and this has nothing to do with pixel density.

    John Sheehy, Aug 9, 2008
  6. Rich

    John Sheehy Guest

    That wouldn't help matters at all, in any substantial sense. With
    current technology, you get less read noise and the same shot noise and
    more resolution, at the RAW image level, with higher pixel density.
    There is a flat rate of pixel-level read noise that seems currently
    insurmountable. The pixel-level read noise is about 0.1% of saturation
    at base ISO no matter how large or small the pixel.

    There is an illusion that things are otherwise, because:

    1) people are ignorant and compare cameras of different MP at 100% pixel

    2) people are overly trusting and believe that when they downsize an
    image to fit a window, that the software is doing it correctly, when in
    fact, many programs use horrible downsizing that increases the image-
    level noise more, the bigger the source image is.

    3) because of #1 and #2, and the optical naivety of the public,
    manufacturers are engaged in a noise reduction race which is obliterating
    any of the benefits of higher pixel density.

    John Sheehy, Aug 9, 2008
  7. Rich

    Paul Furman Guest

    Perhaps in a very small way (in the shadows of low ISO shots where it
    doesn't matter much) but the benefits of larger pixels are real in low
    light/high ISO shooting.

    Paul Furman

    all google groups messages filtered due to spam
    Paul Furman, Aug 9, 2008
  8. Rich

    John Sheehy Guest

    Not at all, except that CMOS DSLR cameras with very low read noise at
    high ISOs, such as Canons and the D3 have lower read noise per unit of
    area at high ISOs than the pixels of typical P&S cameras. Even the
    cameras like the D40, which improve on older tech, still can't beat the
    little pixels in high-ISO read noise per unit of area. Only the very
    best CMOS big pixels can surpass the little guys, per unit of area. When
    backlighting moves along, the little guys could possibly have room to do
    the high-ISO CMOS tricks as well. It wouldn't take but a one-stop
    decrease in read noise at ISo 1600 for the little guys to surpass the
    best current CMOS DSLR pixels in every way, per unit of area.

    The Canon 450D, per unit of area, has similar read noise to the 1D3 or
    the D3, at high ISOs, and more resolution, to boot, to compare different
    sizes of similar technology.

    John Sheehy, Aug 9, 2008
  9. Rich

    Rich Guest

    Uh huh. So a pixel with a 4,000 photon capacity is going to provide
    as much DR as one with 20,000? That is what determines how much DR you
    have, it hasn't changed and quantum efficiency has not increased that
    much since microlenses appeared.
    Rich, Aug 9, 2008
  10. Rich

    Scott W Guest

    DSLRs are down below 10 electrons of read noise, given ratio of area
    between them and a P&S, call it 10:1 you would have to have a read
    noise of 1 electron on the P&S to match a rather poor DSLR.

    Scott W, Aug 9, 2008
    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.