25 MP sensor of Sony

External Since: Nov 04, 2007 Posts: 308

Floyd L. Davidson <floyd.RemoveThis@apaflo.com> wrote:
> Bryan Olson <fakeaddress.RemoveThis@nowhere.org> wrote:
>>frederick wrote:

>>The argument that 12 bits is enough challenges the assumption:
>>"presuming that the sensor is capable of recording that depth
>>of data." Consider the signal that will saturate a sensor cell,
>>so that more light will not increase the change. If a signal
>>1/4096 as large is significantly less than the absolute noise,
>>then 12 bits can cover the sensor's dynamic range.

> That is true, but can be misunderstood too. What does
> "absolute noise" mean? You go on to discuss "dominant
> noise", which is a different beast.

>>Roger M. Clark has argued that the dominant noise is "shot
>>noise," which is in the light itself; even a sensor that counts
>>photons perfectly would have this shot noise.

> Shot noise is dominant above some given level of
> illumination. It is not dominant at lower levels.
> Therefore perhaps the most obvious problem with the rest
> of this discussion is the assumption that shot noise in
> some way relates to the "absolute noise" mentioned
> above. It does not.

> The "absolute noise" reference is to the noise when
> there is no light (i.e., no signal). That is read
> noise. It is random.

>>As sensor cell
>>size gets smaller, so does the limiting signal-to-noise ratio.
>>The signal is proportional to the sensor-cell size, while the
>>noise is proportional to the square root of the signal.
>>(Technically, photon arrival is well-modeled as a "Poisson
>>process".)

> The noise in the high signal areas is photo noise. The
> noise in low signal areas is not. That is where dynamic
> range is important.

>>Thus the smaller the cell size, the fewer bits needed to cover
>>its dynamic range. I have not examined the numbers as well as
>>has Roger, but putting out the signal from a 5.94?m sensor cell
>>beyond 12 bits of precision is a bit like saying the Earth will
>>be four and half billion years old next Tuesday.
>>
>>Capturing bright highlights and dark shadows in the same shot
>>is a different issue from the bits needed to cover a cell's
>>dynamic range.

> If the bits needed to cover the cell's dynamic range are
> not there, whatever range of bright to dark the cell can
> capture is *lost*. It *is* exactly the issue. One cell
> may be at it's brightest, and another cell at it's
> darkest... they *both* must be encoded in the digital
> signal.

They can be encoded with any number of bits from 1 upwards. You don't
lose dynamic range with fewer bits, you lose fineness of gradation
steps. With one bit all you get is white and black. With four bits you
get two intermediate shades of grey. With 12 bits you get 4000 shades
of grey. I'm pretty sure I can't distinguish 4,000 different shades of
grey, so I'm sure 12 bits is more than enough for the dynamic range of
my eyes, other things being equal. There's no reason why a 10bit ADC
couldn;t cover 10 times the dynamic range of a 15 bit ADC. However,
it's often technologically convenient to use more bits than necessary
for a variety of reasons, and those reasons change with the
technology.

Given Sony's different implementation of the ADCs their 12 bits might
in fact offer less noise and more dynamic range than the usual 14
bits, just as X's 12MP camera might offer higher resolution images
than Y's 13MP camera. It's not just the numbers.

--
Chris Malcolm cam.RemoveThis@infirmatics.ed.ac.uk DoD #205
IPAB, Informatics, JCMB, King's Buildings, Edinburgh, EH9 3JZ, UK
[http://www.dai.ed.ac.uk/homes/cam/]

External Since: Jan 23, 2008 Posts: 160

Chris Malcolm wrote:
[]
> They can be encoded with any number of bits from 1 upwards. You don't
> lose dynamic range with fewer bits, you lose fineness of gradation
> steps. With one bit all you get is white and black. With four bits you
> get two intermediate shades of grey. With 12 bits you get 4000 shades
> of grey. I'm pretty sure I can't distinguish 4,000 different shades of
> grey, so I'm sure 12 bits is more than enough for the dynamic range of
> my eyes, other things being equal. There's no reason why a 10bit ADC
> couldn;t cover 10 times the dynamic range of a 15 bit ADC. However,
> it's often technologically convenient to use more bits than necessary
> for a variety of reasons, and those reasons change with the
> technology.

Remember the effects of gamma correction, though. 255 shades of
gamma-corrected data could have a linear dynamic range of more than
4000:1. Yes, you are unlikely to distinguish at the high end (3998 vs.
3999), but the gamma-correction seen in JPEGs does make the lower end
important.

Cheers,
David