> Can this device be designed for this cost? The key is the 24mm x 36mm
> sensor, and I suspect that right now it is too costly... but consider that
a
> temporary condition, electronics costs fall constantly and quickly.
FYI, I design integrated circuits (microchips) for Motorola. We mainly
build high-speed networking processors, but much of what I say applies to
any silicon device (imaging CCDs or CMOS devices). Not to say that you
should take my opinions as gospel, but I am familiar with many of the
aspects of building chips.
The main reason that full size sensors cost so much is that as the size of a
CCD or CMOS device goes up, the cost of processing that device goes *way *
up, It's not simply a matter of size being directly proportional to cost.
As the size of your sensor increases, the chance of defects greatly
increases. For example, say your chances of a processing defect or defect
in silicon are 1 defect per square mm. If your sensor is a full size
sensor, or 864 mm^2, that presents the opportunity for more than twice as
many defects as the sensor in the D60, which is 343 mm^2. Since the cost of
processing a full wafer of silicon is pretty much constant, reducing the
size of the chip is the only way to reduce the cost of the chip.
There is really no good solution to this problem. The only way to
effectively reduce the cost per device is to make your device as small as
possible, thus insuring that you have reduced the possibility for defects
and you have maximized the number of chips that can be produced on a given
wafer. This is why I'm skeptical that we will ever see a full size sensor
that is cost effective.
< 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 >
|