The megapixel race heats up again

Discussion in 'Digital SLR' started by Roy Smith, Jun 20, 2006.

  1. Roy Smith

    Roy Smith Guest

    Roy Smith, Jun 20, 2006
    1. Advertisements

  2. Mick Anderson, Jun 20, 2006
    1. Advertisements

  3. Roy Smith

    Arthur Small Guest

    I don't believe that the average photographer, or even most pros will be
    able to afford this.
    If it was developed for the government you can guess how much it cost.
    Arthur Small, Jun 20, 2006
  4. Roy Smith

    tomm42 Guest

    Right now we are working in factors of 10 with each jump in sensor
    size. Last I heard the costs were something like this:
    6mp APS sensor $100
    full 35mm sensor $1000
    39mp mf sensor $10000
    This sensor is 4x4 inches, about 8x area of silicon of the 39mp sensor,
    so are we looking at $100000 or $500000, just shows where we will be in
    20 years, really want one for my 4x5 though.


    PS, prices are estimates are what was floated around on dpreview about
    6 months ago.
    tomm42, Jun 20, 2006
  5. Roy Smith

    Jeff Rife Guest

    Roy Smith () wrote in
    A 12 megapixel APS-C camera has a *much* more impressive sensor than
    the one in this link.

    The 4" x 4" sensor is only 10560 x 10560, which gives you a whopping 4
    megapixels in a APS-C area. Heck, even at full 35mm sensor size, you
    only get 9 megapixels.
    Jeff Rife, Jun 20, 2006
  6. Thomas T. Veldhouse, Jun 20, 2006
  7. NO pro can afford it. It is built for a telescope as a one shot. I am sure
    the chip probably cost in the hundreds of thousdand dollars.
    Thomas T. Veldhouse, Jun 20, 2006
  8. Roy Smith

    ian Guest

    several times more than it should have and 4x the original estimate.
    ian, Jun 20, 2006
  9. Roy Smith

    ian Guest

    By that reckoning what about 8mp compact point and shoot cameras? I should
    think the surrounding mechanisms that exist are what limited the pixel
    density. I should think noise levels need to be extremely low for serious
    astrometric stuff.
    ian, Jun 20, 2006
  10. Roy Smith

    Jeff Rife Guest

    ian () wrote in
    Any 4-6MP APS-C sensor will have pretty much the same noise level,
    assuming similar technologies. Now, if the referenced sensor *is* a
    lot better about noise, then I'd like to see that same technology on
    a 8-10MP 35mm sensor...that would be news.
    Jeff Rife, Jun 20, 2006
  11. Roy Smith

    John Francis Guest

    Ah, but it's really easy to cool it - just make sure it's not in
    direct sunlight and stick a couple of cooling fins on the back.
    John Francis, Jun 20, 2006
  12. Roy Smith

    ttdaomd Guest

    ttdaomd, Jun 20, 2006
  13. Roy Smith

    Prometheus Guest

    Not very good resolution, but the larger elements mean less noise, but,
    and perhaps you missed the point, it is not an APS-C sensor, it is a low
    noise high resolution sensor.
    Prometheus, Jun 20, 2006
  14. Roy Smith

    Prometheus Guest

    The limiting noise factor in sensors is the uncertainty in the arrival
    of photons (quanta), if you make the individual elements four times the
    area they detect four times as many photons and since the uncertainty is
    a square law the signal to noise ratio is doubled. Current sensors are
    about 30% efficient, doubling that to 60% would only give a noise
    improvement of 1.4x.
    Prometheus, Jun 20, 2006
  15. Roy Smith

    ttdaomd Guest

    You are right but I am sure these sensors are cooled and have much
    better quantum efficiency and much less noise than the ones we use for
    general photography.

    ttdaomd, Jun 20, 2006
  16. Roy Smith

    Jeff Rife Guest

    Prometheus ([email protected]) wrote in
    Where in the article does it say the sensor is "low noise"? If you
    assume that it is low noise because of the lower pixel density, then
    an array of 10MP 35mm-sized sensors would equal the "high resolution"
    of this unit with about the same noise.

    The only truly impressive thing about this sensor is that they didn't
    chop the chip into multiple units, and I assume there are no stuck
    Jeff Rife, Jun 20, 2006
  17. Roy Smith

    ttdaomd Guest

    You must be an engineer! I have no idea how and exactly where this
    chip is going to be used. On earth, it would obviously not be
    effective at dropping the temperature of the chip sufficiently below
    ambient temperatures needed to improve noise. Maybe if it were just
    floating in space it would be enough.

    ttdaomd, Jun 20, 2006
  18. Roy Smith

    Prometheus Guest

    They probably are cooled. Even with cooling and near 100% efficiency
    there will be an uncertainty in how many photons arrive at each well
    from a uniform source in any given time frame.

    libration.html#qe> Note carefully the unrealistic assumption that every
    photon gives rise to a pixel detection.

    Take a look at
    <> for a
    real CCD telescope.

    Prometheus, Jun 20, 2006
  19. Roy Smith

    John Francis Guest

    I assume it's for use in space - nowadays a lot of astrophotography
    is done from outside the earth's atmosphere, and that is a locale
    where digital photography offers many advantages. All you need is
    a solar panel for power, and a radio transmitter to send the images
    back to earth (both of which are probably already present) - no need
    for bulky one-use media kept in light-tight enclosures, nor for any
    mechanical transport that can operate in a vacuum.
    John Francis, Jun 20, 2006
  20. Roy Smith

    Prometheus Guest

    I assume that it is low noise, in fact given that it is for astronomical
    use it WILL be low noise.
    I am assuming that the designers designed it for lower noise, the
    possibly large area of the individual elements is one factor.
    And how would you deliver the image from the telescope to a
    non-contiguous array without degrading the performance.

    Why should they want to brake it? Do you think a shop window is
    impressive because the manufacturers cut it in to little pieces?
    Or the error rate is below the threshold set in the design specs.

    P.S. To place your own order for one of these see
    Prometheus, Jun 20, 2006
    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.