William Sommerwerck wrote:
>In Gary Schloss's review of the PMA show, he mentions:
>
>> IMHO, the C-2000Z is "much ado about very little". Yes, it has more
>> pixels, but on a smaller CCD than the D-600/620, so the resolution
>> gain is insignificant.
>
>Is that what you meant to write? More pixels usually provide higher
>resolution regardless of the chip size.
... and Dave Haynie jammed all night, and by sunrise added insult to
injury: :-) :-)
>Yeah, I almost responded on this one, too. the actual size of the
>imaging device doesn't matter in the least [...] the only thing that
>really matters is the true resolution (color and pixels).
Oh boy! Each time I choose brevity over clarity, I come to regret it.
(Oh boy #2! Court jesters must work harder to earn any professional
respect :-))
Disclaimer: I am by no means an expert on CCD/CMOS imaging technology,
so pls take the following with a grain of salt. Also, my apologies to
those Zuiks who couldn't care less for the following discussion, and
consider it OT.
First, virtually all digital still-image cameras utilize progressive
scan CCD chips with square pixels. "Progressive scan" means that the
output of each individual pixel can be accessed separately. There are
a few important facts to remember:
(a) CCD is in essence an analog device, i.e. its output has to be
digitized; by their nature, both the CCD itself as well as the
quantization process produce quite a bit of noise;
(b) the output signal of a CCD element depends on its physical size,
i.e. larger pixels result in higher signal => higher S/N ratios
=> wider dynamic range => better picture contrast;
(c) more densely packed pixels increase inter-pixel noise => lower
S/N ratios => lower picture contrast;
(d) CCD's are manufactured like semiconductors, i.e. larger size
rectangular chips result in lesser wafer utilization and lower
yield => higher per unit cost.
Conclusion: for a given number of pixels, a larger size CCD chip
will capture a higher quality picture, albeit at a higher cost.
Second, there is the issue of optics in front of the CCD element.
Arguably, the larger the CCD chip size, the easier and more cost-
effective it is to devise the front end optics with adequate APO
correction and resolving power. Hence, I would surmise, though
at this point -- without any proof, that the 9.2-28mm F2.8-3.9
lens on the D-600L camera that has a 2/3 inch CCD, is as good or
better than the 5.4-16.2mm F2.8-4.4 lens on the D-400 camera that
has a 1/3 inch CCD.
To summarize: the number of pixels in an imaging sensor is important.
But, it's definitely not the only criterion in assessing the averall
digital camera performance. And so, I stand by the above statement
from my PMA report.
Note that the high-end digital wonderbricks from Ni/Ca/Mi/Fuji/Kodak
use the 2/3 inch CCD chip (1.4 Mpix, probably same as in D-600/620L),
although a 2.3 Mpix 1/2 inch CCD chip has been available for quite
a while. However, Fuji has just released a new digicam with a 1/1.7
inch rectangular CCD chip that has 1800x1200 pix (raw 2.3 Mpix). This
may be the basis for the next generation of wonderbrick digicams.
(Note that a CCD is responsive to light's intensity, not its
wavelength (color!), so special filtering techniques are needed to
produce a color picture, but this issue is beyond the scope of our
discussion).
Same guys also had problems with my following observation:
>> With all its push into the digital domain, Olympus is yet to dazzle
>> the world with a truly high-end, professional "flagship" digital SLR.
William Sommerwerck wrote:
>In fairness to Olympus, I have yet to see a "dazzling" pro SLR from _any_
>company. As long as the resolution of such products doesn't even remotely
>approach that of silver-based media, I refuse to be "dazzled."
... and Dave Haynie added:
>If you're into that kind of thing, the fairly monsterous Kodak/Canon,
>Kodak/Nikon, and Fuji/Nikon wonderbrick conversions do this, if you
>have $5K-$15K to spend. Olympus isn't tempted by this market.
Well, the purpose of those expensive exercises is more a marketing
tool than anything else. Sadly, Olympus took the same road in the past,
foregoing the expense of developing a pro-level OM body (in the style
of Nikon F and Canon F-1). I think it was a mistake then, and now
history may repeat itself.
Of course, Olympus has very few options left, even if it wanted to,
since it really has no SLR body suitable for a "wonderbrick conversion",
and the IS-3 doesn't really cut it. Besides, I believe the IS-3 has
been discontinued.
Cheers,
/Gary Schloss.
Studio City, CA
schloss@xxxxxxxxxxxxxx
< 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 >
|