Larry, I suspect you are making some inadvertent switching in the logic
here.
The part of the in-camera software (that does optimization for JPEG)
is actually "post-processing". I.e. it is what happens after the RAW image
has been acquired by the sensor, and is being converted to JPEG.
I don't think there is any JPEG-targeted optimization of how the [RAW]
image is taken. I.E. If you were to choose recording RAW+JPEG,
then the camera acquires "raw" image, records it as "RAW", and then
converts to JPEG.
Now, the reason why this [post-processing] portion of camera's software
is optimized for JPEGs is that in most cases people who are using JPEGs
are not going to bother with RAW and POSTprocessing outside of the camera
appreciate that automatic ("magic") in-camera processing.
Those geeks and nerds who would bother to use RAW, will do post-processing
outside of the camera anyway. So, why doing that in camera?
But I hear you that it would be nice to have the camera doing a more
careful metering based on the entire image, not just a small portion.
And to do that, essentially, in real time (as opposed to pre-metering,
which is what happening now)
Unfortunately, as far as I understand, today's computing capabilities are
not fast enough (at least for the expected price range of the end product)
to do the real-time metering and on the entire sensor.
I don't even know if the today's technology would be capable of doing
full-sensor pre-metering (in a reasonable time), while maintaining the
portable size of the camera and its battery(!). And that's yet before
the price discussion.
Sorry, I feel I could've explained it better, - but I feel tired tonight.
Cheers,
Igor
Larry Colen Wed, 24 Oct 2018 13:06:02 -0700 wrote:
P. J. Alling wrote on 10/24/18 12:39 PM:
It already does that and if it could everything would be recorded in a
uniform grey.
No, it uses a much smaller number of sensor sites to take a guess at
exposure, rather than looking at everything, setting the raw exposure so
that only at most a certain percentage is at, or close to, the clipping
point.
Similarly, it could also analyze the dynamic range of the scene and find
the ISO that optimizes for that dynamic range, allowing you the fastest
shutter and/or greatest depth of field within the dynamic range of that
scene without clipping the brights or losing the shadow detail to noise.
For landscape photos with DR that exceeds the ability of the sensor, it
could optimize the bracketing to get all of the detail in the scene, to be
compositied later in post processing.
I find it bothersome that the camera software is *only* optimized for
creating jpegs in the camera, rather than collecting the raw images to
give the best photos possible after post processing.
--
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
to UNSUBSCRIBE from the PDML, please visit the link directly above and follow
the directions.
