Willem-Jan Markerink wrote:
Nope, some chips apparently can function without the micro-lenses
(which causes this angle-of-impact-vulnerability in the first place,
TTBOMK).
The Dalsa chip (the one that will be used in the Mamiya ZD medium
format DSLR & digi-back) is one of those....
(although you could deliberate whether 36x48mm is full-format in any
legacy/analog context....;))
I assume that the lower the density of pixel (i.e. a larger sensor size
for a given number of pixel) the easier it will be to achieve a high
fill factor (size of sensitive pixels / total size of "sensor"). With
8+ Mpixel for an APS-C sensor you have very small pixels. I would think
that the necessary spacing between pixels has at least a constant value
for a given manufacturing process. So if you put larger pixels on a
sensor your fill factor will go up.
If you have small fill factors you want to have microlenses to get a
better S/N ratio.
However a high fill factor (i.e. small spaces between sensor pixels)
can cause unwanted effects like blooming as mentioned in the original
text. It would require extra effort to overcome these kind of problems.
Overall it is the old game of tradeoffs price vs. performance.
When I moved from analog to digital EOS 50e (Elan IIe) -> 20D I though
I would be happy with my existing lenses. (I got the EF 17-35 2.8L, the
EF 50 1.8 II, the EF 24-85 and the EF 75-300 IS.
Well, now I think about getting either the EF-S 10-22 or the Tokina
12-24 for landscape and I am debating whether the new 24-105 4.0 L IS
would be nice replacement for my EF 24-85 with less CA.
BTW: For film the new EF 24-105 L IS would be my dream allways on
lense.
I would love to see some EF-S 15-60 (IS) coming out. Well, let's keep
dreaming.
CA is I think the major drawback with some of the old (and even new,
e.g. the Tokina 12-24) lenses.
So all the lenses one bought for film need to be re-evaluated for
digital.
Regards
Thomas
*
****
*******
***********************************************************
* For list instructions, including unsubscribe, see:
* http://www.a1.nl/phomepag/markerink/eos_list.htm
***********************************************************
