Re: 24x36 sensors in low end SLRs?

Lourens Smak wrote:
> ...
>> These are two primary directions that lead to the predicted growth of the
>> level
>> of integration: making chips denser and making chips bigger. You seem to
>> neglect
>> the latter for some reason.
>>
>> In case of sensors making chips _denser_ is not really that easy (or even
>> possible) because of noise problems.
>
> You must be joking.... the past 20 years oro so have been spent making
> chips denser and denser (not just imagers, but all kinds of chips) and
> smaller and smaller.

I don't understand what "all kinds of chips" have to do with what I said. I
clearly stated that this sentence specifically applies to imaging sensors and
only imaging sensors.

> Because a wafer is more or less a fixed cost for
> the manufacturer, this means manufacturers earn lots more from the same
> wafer. Everything digital CAN ONLY get smaller, it's economics at the
> simplest level. They will not spend a single dime on research for making
> the same functionality, but in a larger chip; that's idiotic from a
> chip-making point of view.
> ...

"The same functionality, but in a larger chip" - where did you get this from?
Sorry, but that's your own invention. Nobody is talking about "the same
functionality" here. The target, of course, is to put _more_ functionality into
a single chip, by making it either _denser_ or, once again, _bigger_ (or
_both_). The [most] important price-defining factor here is the cost of so
called "packaging", which is the cost of creating the infrastructure that
surrounds the chip - the PCB, the price of mounting, implementing external
connectivity etc. Up to a certain limit, implementing functionality through
packaging is much more expensive than implementing the same functionality in a
chip. For this reason manufacturers are always trying to get rid of as much
packaging as possible by reducing the number of separate chips in a system, i.e.
by squeezing more functionality into individual chips (increase the level of
integration). Both methods of increasing that level - increasing density and
increasing size (both - up to a certain limit) - reduce the final cost of the
system. The economically optimal mix of the two depends on the currently
available technology.

This does not immediately apply to resolution of imaging sensors, though.
Imaging sensors will always have their own specifics and their density will
always be limited by the noise problem. For this reason, increased resolution
will normally require increasing size of the sensor.

Best regards,
Andrey
.

Relevant Pages

  • Re: basic elapsed seconds counter
    ... It requires understanding the device hardware (timers, I/O ports, ... as wiring up 5 small chips. ... modify the functionality, to tweak the design. ... I am also still very interested in wiring up 5 small ...
    (sci.electronics.basics)
  • Re: Some more questions and/or thoughts...
    ... Making big chips is expensive and is going to ... You are aware that a camera is more than just a sensor, ... The materials cost is a constant. ... functionality of one particular chip into a set of two ...
    (rec.photo.digital.slr-systems)
  • Re: [PATCH v2] edac: mpc85xx: Add support for new MPCxxx/Pxxxx EDAC controllers
    ... Anton Vorontsov wrote: ... It appears that the driver has two device ID tables. ... my previous attempt enabled only half of the functionality. ... L2 on the p4080 is quite different from those other chips. ...
    (Linux-Kernel)
  • Re: Higher bit conversion in DSLRs unlikely
    ... in various digital cameras? ... These chips have several million gates and ... and cost around $10 in large quantities. ... The AFE is about 1/100 of the functionality of the chip, ...
    (rec.photo.digital)
  • Re: Honest carb question
    ... good for making chips from what I've heard...dykon radish is supposed ... to make good chips too but peppery ... Radish chips sound good. ...
    (alt.support.diabetes)