In article , Peter Leyssens <Peter.Leyssens@xxxxxxxxx> writes
>>From: "John A. Prosper" <prosper@xxxxxxxxxxxxxxxxx>
>>
>>|Let's
>>|also hope that it's future-proof, i.e. you can change the CCD so that
>>|when the 16Kx16K CCD in full 24-bit, with hardware grain-simulation
>>|comes out, you can fit it in the OM-5 too. :-)
>>
>>Well, if they want to make a pro digital camera, you would
>>have to shoot for 36-bit color (12 bits apiece for red,
>>green, and blue). After all, there are 36-bit color slide
>>scanners.
>
>I meant *full* 24 bit. If you find yourself a 36-bit color slide
>scanner with 12 *effective* bits, you'll be paying the kind of money you
>think of when buying a house (with a swimming pool, that is). Some
>people I know work in space industry and they have a *full* 16-bit CCD
>(B/W of course). They bought it for $50.000, but that was only because
>there was a complete line defective.
>
>In commercial CCDs, the lowest n bits are always garbage.
>
Actually, Peter, CCD's are ANALOGUE devices - the output needs to be fed
to an ADC before it can be read by computer. The ultimate limit of
signal to noise for a CCD (or any other light sensor) is the photon
noise limit. Most visible CCD's have a storage capacity at each pixel
of around 10^6 photo-generated carriers, and since the arrival of one
photon is totally uncorrelated with the arrival of another at the pixel
(ie. it is a random process) the noise level on this is approximately
the square root of the carrier capacity - around1000 - as is the signal
to noise ratio. To prevent saturation, CCD's are normally operated
closer to 700f saturation, so the SNR is actually lower than this in
practice. So the limit for a typical high performance CCD detector is
around 10 bits.
Some specialist devices have higher capacities - I have designed with
infra-red detectors with storage capacities in excess of 30 million
carriers at each pixel but then, off course, the pixels can be MUCH
larger due to the longer wavelengths involved.
Astronomical units - as Lee will doubless confirm - achieve higher
dynamica ranges by reading the CCD out numerous times, converting the
signal to digital form and accumulating the result. This increases the
effective photon noise limit by a factor of the square root of the
number of integrations, but pretty soon this becomes dominated by other
effects - particularly dark current noise etc. Whilst such multiple
integrations are fine in the relatively stable astronomic field, they
would not be suitable for normal photography since the subject movement
would corrupt the image.
So, until someone invents a higher dielectric optically sensitive
material than silicon for CCD's to be manufactured from, the practical
resolution limit for digital designs is less than 10 bits, and since
silicon technology has been dominant for the past 50 years - don't hold
your breath waiting for that alternative!
--
Kennedy
Yes, Socrates himself is particularly missed;
A lovely little thinker, but a bugger when he's pissed.
Python Philosophers (replace 'nospam' with 'kennedym' when replying)
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|