Re: [photo-3d] Rating digital cameras

[Mike Kersenbrock <michaelk@xxxxxxxxxxxx>]

Jim Harp wrote:

> Are there other important factors in determining the quality of a digital
> camera besides pixel count?  Does 3.3 megapixels look the same regardless of
> the camera?   (These aren't rhetorical questions;  I'm genuinely curious.)

No there are other differences as well.  There are differences in the CCD's
sensitivity (making the sensors smaller to have more of them, tends to make
them
less sensitive for a lower "ISO" rating and/or more noise), they have
different noise levels, they have different optics in front of the CCD's, they
have very different software that converts that raw data (NOT straight rgb
combination,
the algorithms for their color processing are complex and proprietary), there
are certainly differences in features, differences in package size and 
configuration (I've the obsolete Nikon 950, since replaced by the 990, and
both have a very "different" sort of configuration, and it's nice).  They
have very different focus personalities. Mine (really my
wife's) will focus down to 0.8" and do a zoom from there up to 3x).  How
quick one can take one picture to the next (at full resolution) can vary
widely with processor power, algorithms, and onboard memory.  How much
degradation one gets with jpeg compression also varies from mfg to mfg,
the Nikon ones, for instance, compress particularly well quality-wise.  Not
all use the identical same code.

So yes, the megapixel count is being used in the way you suggest.  One
company is even faking it using software extrapolations to 6 megapixel
output files, with an asterisk on the "6 Megapixels" in ads to indicate
in the fine print that it's the file's pixel count, not the CCD's (and
the CCD's count is another posting in itself that I'll not get into).

Buyer beware, do some homework first!  :-)

Mike K.



> Manufacturers of high tech gear have been known in the past to become
> obsessed with achieving whatever the statistical measure of excellence is
> without regard to the actual quality of the output.  Jim Harp