I came across the following post in the dpreview Olympus SLR forum. I
thought I'd post it here to see what you technical ones think of it.
Richard
**************************************************************************************************************************************
The imaging CCD used in the 4/3rds dSLRs is smaller than the APS and
"full-frame" sensors used by competing manufacturers. Any technological
advances for improving signal to noise ratio or dynamic range of the 4/3rds
size chip can also be applied to the larger sensors used by the competition.
Therefore, 4/3rds dSLRs will always lag behind the competition in image
quality. Of course, this argument applies only to 1-chip cameras. If 2 or 3
imaging sensors are used in the dSLR - it becomes a whole new ballgame.
The 4/3rds standard does not specify an imaging sensor of any particular
type or size, but only the diameter of the image circle. Indeed, the name
4/3rds refers to the size of a vidicon tube that was used in early video
cameras. With a vidicon tube being a tubular, cylindrical object, one could
compare it to the type of condenser lens and dichroic prism block used in
multi-chip dSLRs. For an illustration, examine the optical path diagram for
the Minolta RD-3000, which is a 2-chip dSLR design from 5 years ago:
http://www.steves-digicams.com/rd3000.html
Now, if the question is "what is the ideal size (diameter) of the condenser
lens and dichroic prism for this type of optical path?" then "4/3rds" is a
relevant and descriptive answer.
The 4/3rds standard has been criticized for demanding a higher degree of
telecentricity than is necessary, as illustrated by the success of Nikon's
DX lenses with their APS sized sensor. However, if one considers an optical
path containing the prisms and dichroic mirrors needed to split the image
for 2 or 3 imaging sensors - then the need for near telecentric lenses
becomes obvious. One of the reasons the Minolta 2 and 3 chip dSLRs were not
successful was because they used film camera lenses, instead of lenses
designed for the purpose (like 4/3rds).
This may help explain why Sigma and other manufacturers have not put 4/3rds
lens mounts on their current lens designs. While they may provide
satisfactory performance with low-resolution, 1-chip 4/3rds cameras - they
will be inadequate for the high-resolution, 2 and 3 chip 4/3rds cameras of
the future. Lenses purchased for the 4/3rds cameras of today must be
compatible and useful with the 4/3rds cameras of tomorrow.
The 4/3rds approach to high resolution is a 3 sensor array. Using 3 smaller
chips to achieve 21 megapixels is much more practical than using just 1
larger chip, such as the Canon "full-frame" 35mm film size, or a "medium
format" size chip like in the new Mamiya ZD.
A 3-chip dSLR is 10 year old technology, as illustrated by the Minolta
RD-175:
http://www.epi-centre.com/reports/9605cs.html
I am convinced that a perfected version of the Minolta RD-175 is what
Olympus had in mind when they invented the 4/3rds concept. The RD-175 did
not flourish because it used imaging chips which were too small, and legacy
film lenses with inappropriate focal lengths, not telecentric, and had image
circles which were too large.
One problem with the Minolta RD-175 was that the internal optics could
gather dust - which was almost impossible to clean. This is why Olympus
invented the sonic dust shaker system, which will undoubtedly be applied to
the optical path components in future 4/3rds cameras. With a single chip
camera, sonic dust removal is a minor convenience. With a multi-chip camera,
it is essential - and Olympus has already got it.
I believe the design of the Olympus E-3, a 3-chip camera of approximately 21
megapixels, was begun before the Olympus E10 was produced. When the 4/3rds
standard was fist announced in 2002, a high resolution 3-chip dSLR was
(secretly) the ultimate goal. From the beginning Kodak, Fuji, Matsushita,
perhaps Foveon and perhaps others have been developing prototype sensor
chips for the Olympus E-3 dSLR.
Now, what would stop Canon, Nikon, and the others from developing a 3-chip
dSLR of their own - to compete against the Olympus E-3? Nothing - except
they would have to develop a whole new line of telecentric lenses to be used
with their own cameras. So in this regard, Olympus is about 5 years ahead of
the competition - and probably holds some patents that will be difficult to
get around.
Regarding the E-series dSLRs, Olympus has taken some steps which appear
puzzling. This message is nothing more than my speculation on how the 4/3rds
standard came to be designed as it was, and what the future may bring.
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|