Re: Any in-camera HDRs available?

Peter <peter@xxxxxxxxxxxxxxxx> wrote:

How *do* you decide the pixel is becoming over-saturated
without looking into it (i.e. reading the pixel out)?

There *has* to be a way!

Sure --- there also *has* to be a way to measure both the
position and the speed of, say, an electron.

Yeah, me know sum physics or sumpthin. Me knows that aint possible

But "There *has* to be a way!"

Maybe by detecting photon/electron spill in
adjacent area, hmm?

So you basically want to put 4 "empty" pixels around each
active pixel? Gonna be very good for fill factor and full
well capacity!

Oh, and you'd need tons of expensive circuity, since you cannot
normally read out random pixels with CCDs (CMOS doesn't have
that effect).

Isnt there a microlens above each pixel?

Usually, these days.

If so there should be some empty
space around each pixed already.

No, the microlens is *not* glued to the pixel on the space
around it.

But there usually is some 'empty space' there, yes --- the
very same empty space that stops the electrons from spilling
into adjacent pixels *before* the well is full. Oh, and some
electronics, for "electronic shutter" and stuff.

Sure it would be expensive! We are not penny-mongers here!

You show approx. 130.000 USD and I'll show you a highly sensitive,
backlit (80-90% QE), high quality[1], low defects 6.1x6.1cm sensor,
16.8MPix, large (15µm)² pixel size (100% fillgrade, too!) ---
and if you take one and a half minutes reading out the sensor
(at -40°C), you also get less than 5 electrons read noise. If you
are in a hurry, you can read the sensor out in a good 4 seconds,
but read noise is 10-12 electrons then (at -60°C).

But are you really sure you want to play in that liga?

Should I take out a patent right now? ;)

Patenting stupid ideas is patented. I'd sue you.

LOL! :)

-Wolfgang

[1] no more than 200 dark (<50% response) or hot (>10e-/s
@ -60°C) pixels, no more than 5 broken columns (e.g. >10
pixels in a row) and no more than 2 of them adjacent, no
more than 25 dark/hot spaces where the dark/hot pixels
cluster within 3 rows distance.
.

Relevant Pages

  • Re: Smear
    ... All explanations I found say that smear happens during the vertical ... at every step of vertical transfer the pixels next to the bright ... excess electrons spill along the transport direction both up and down. ... A bright spot on a ccd can also cause electrons to be ...
    (sci.optics)
  • Re: Why not make the sensor larger?
    ... off the sensor at speeds faster than the speed of light? ... you need to move those electrons a couple of centimeters, ... read out voltages from individual pixels (the charge in each individual ...
    (rec.photo.digital)
  • Re: Why not make the sensor larger?
    ... The Janesick article does not state that readout times for individual CMOS pixels are huge. ... To get 96 db of dynamic range, ... enough photons: 65535*(read noise in electrons). ... So if read noise matched the current lowest read noise ...
    (rec.photo.digital)
  • Re: Talking to digital camera novices about choosing a camera.
    ... Lens quality ... 2a) Full well capacity in electrons ... Full Press Shutter lag ... misinterpreted as number of pixels. ...
    (rec.photo.digital)
  • Re: Resolution question
    ... That really is an insane number of pixels per mm. ... (The limit on lens performance is diffraction. ... sensor, giving you larger pixels. ... People with more sense than money who own cameras like the 1Dsmk2 or 5D will ...
    (rec.photo.digital)