bclaff

As Jim mentions there is really no difference up to ISO 500 : And at ISO 640 and above whether you would see a difference would depend on the dynamic range of the scene being photographed.

Yeah, for Photographic Dynamic Range (PDR) the D750 stands out from that pack: Regards,

I enjoy the free information they provide at conferences or on their blog. The fact that I must treat the system as a "black box" is generally fine for me since I'm primarily interested in the ...

In the analysis I perform only the Blue (or Green for older sensors) is clearly different; but never both at the same time.

Thanks, I completed my first pass of last weeks photos. Here are a few more. NIKON D7200 70-180mm f/4.5-5.6D ED AF Zoom Micro-Nikkor@178mm f/5.8 d=1'3" f/32 1/250s ISO 100 w/2 SB-R200 flashes NIKON ...

I'm pretty experienced at close-up photography and wind can be quite a challenge. In a recent outing I found the wind was so strong that insects tolerated my stabilizing flowers by hand perhaps ...

To be clear, at my level of analysis I can only separate channels based on their position in the Color Filter Array (CFA). I don't know for certain the "Blue PDAF pixels" are in fact behind a ...

I really don't know. Just reporting my findings. I'll leave that question to those with a better understanding of how OSPDAF works.

Hmmm. From The State-of-the-Art of Mainstream CMOS Image Sensors : Lines up with my research so far. Sony SLT-A99V (!), ILCE-7, and ILCE-7M2 using Green and ILCA-77M2 (!), ILCE-6000, ILCE-6300, ...

Sorry for the slow response; yeah I find it interesting to see all cameras on a "level" playing field; not that they are all used the same way.

Here are the Photographic Dynamic Ranges (PDRs): Note PDR is a different scale than DxOMark Landscape. You can examine other sensor data like Read Noise at the site as well.

Not surprising ( PDR fom PhotonsToPhotos ): Read Noise from PhotonsToPhotos : The A7ii doesn't have dual conversion gain and so it doesn't have the improvement that kicks in at ISO 640.

I'm not aware of any way to indicate that some pixels in a raw files are "half" intensity as compared to the others.

On review it turns out I had Red and Blue exchanged to the pattern properly colored looks like this: Probably should have caught this earlier as PDAF generally involves the Blue pixels.

There's strong evidence (due to the row noise behavior) that the Blue PDAF pixels are paired. So adjacent masked pixels on a row is what makes sense as a theory. The Green signal level is higher ...

Note minor clarification above underlined in bold. Here's some evidence that the PDAF are masked pixels where adjacent pixels are "combined": Best viewed Original Size Note that the Blue PDAF rows ...

That does seem strange in retrospect. Perhaps I got the channels confused in presenting my findings. I'll try to look into it.

E-M1 Mark II Fixed Pattern Noise (FPN) Results

I don't generally test for signal processing (Noise Reduction (NR)) on a per channel basis. Here's the 2D Fourier Transforms (2D FTs) for the Olympus OM-D E-M1 Mark II and the Pextax K-50: 2D FTs ...

:-)