On 2/20/2019 8:50 AM, Wayne Shumaker wrote:
Moire is when the pattern and the sensor sampling approach the Nyquist rate combined with the fact the color pixels
are not spatially on top of each other. As such slight angle shifts, such as when you re-taped Thomas, may affect the
intensity of the color shift. Also the pixel density of different cameras are sampling differently,
Exactly the point of comparing 16 and 20 MP bodies, with ~50 MP, non-demosaiced
version to show what's really there.
not to mention the degree of filtering in the anti-alias filter in front of the sensor.
No AA/OLPF filter in either camera. There is a definite trade-off of moire suppression and detail resolution in using AA
filters. Here's an example from someone who replaced the filter in a Nikon D200 with flat glass.
<http://www.moosemystic.net/Gallery/tech/AA_Filter/D200_AA.htm>
Nikon choose not to use one in the D810, which increased resolution of detail - and - produced some really awful moire
and demosaicing artifacts with some subjects. <http://www.moosemystic.net/Gallery/tech/X-T1_E-M1_D750/FON.htm> Look esp.
at D750 vs. D810.
There are so many factors, I'm impressed with and applaud your detailed investigation.
Thanks! I can be dogged on occasion, when I think the knowledge gained may
improve the technical aspects of my photography.
A near impossible task to find nirvana.
:-) Nirvana is in the photo, not the camera.
The Siemens star image used in https://blog.kasson.com/the-last-word/adobe-camera-raw-11-2-enhance-details/ would
eliminate differences in pixel density and orientation, to some degree.
I'm not interested in test subjects that I can't relate to actual use. It may be a failure of my imagination, but when I
look at those test images, they don't relate in my mind to how to use the camera and process the results.
Reminds me that film, with randomness in the grain sampling, does not have this problem of uniform sampling. The
randomness of grain in film vs randomness in a scene for digital sensors is the tradeoff.
And yet, it makes, from my perspective, in my straightforward mode, crappier
images. ;-) So what's the point?
In signal processing, the sharper the cutoff of a low pass filter, the more there is overshoot in the step response.
This is what creates the halos with over-sharpening. https://en.wikipedia.org/wiki/Ringing_artifacts A Bessel and
Gaussian filters, for instance, have a much smoother filter roll-off and no overshoot. I spent many years designing
filters where managers never understood that you cannot have smooth time response with sharp filters.
Again, no filters involved in this test. I wonder if you may be confabulating pure sensor pitch, subject frequency,
moire effects with the effects of demosaicing?
If you know the subject matter is going to have moire issues, then perhaps the best approach is to shoot taking that
into account? https://www.nikonusa.com/en/learn-and-explore/a/products-and-innovation/moir%C3%A9-false-color.html I
like Kasson's comment on ACR enhanced detail: "The good news was that it appeared to do no harm. The bad news is that
it didn't do much good."
And that's good news for me. I don't need it, AND it's not useful with my
computers.
I'm not familiar with all the terms. By HR do you mean higher resolution mode using sensor shift technology? Seems
like the best way to compensate for moire, is that what <4,3> E-M5 II HR=>4640 is doing?
Although others are doing versions now, Oly pioneered this with the E-M5 II. Oly shifts the sensor in half pixel pitch
increments to take eight exposures. This accomplishes two intentional things and one nice byproduct:
1. Higher resolution in the combined image. Although limited to tripod an static subjects on that body, it's still very
useful for things like this comparison, lens tests, etc. It reveals that most µ4/3 lenses considerably out resolve the
sensors.
2. No demosaicing of an array of different colors. Each pixel location is sampled with sensels of all three colors. In
this sense, it's analogous to the Foveon sensor.
3. It acts as highly effective noise reduction.
This review of the E-M5 II compares the 40 MP JPEGs of its HR mode to the 36 MP Nikon 810 and then the 64 MP Raw files
to the 51 MP Pentax 645Z. There are some excellent sample subjects and brief, but useful tech piece at the end.
Note that they used the Oly ACR plug-in to process the Raw files, which compromises the HR results. They had to choose
between the clumsy USM used in the plug-in, with the resultant halos, vs. detail resolution. Obviously, Oly were
concerned that simply opening the files in ACR initially showed a very soft image. Unfortunately, they chose a
non-optimal solution. This little test I did shows that using deconvolution, rather than USM, gives much more fine
detail resolution without the halos. <http://www.moosemystic.net/Gallery/tech/Focus_Magic/_B003950fpACRvsPlugin.htm> You
can see that in my first two samples.
Soooo, when reading the review, keep in mind that the Raw HR files in the Pentax comparison are actually a bit better
than what they show. A piece of big news with the E-M1X is faster processor(s) that allow HR shots sans tripod, @
exposure speeds for all eight exposures up to 1/60 sec.
I wonder if sensor shift could be done in some sort of random way to simulate
grain?
Not a clue, and perhaps not necessary, as there are no demosaicing artifacts.
going off the deep end
Have a nice swim!
Detail Oriented Moose
--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|