Hey there. I have a raw file from the sony a7cr displaying what I think are PDAF lines. There was strong flaring in the shot due to the Sun being almost in the frame.
The PDAF lines filter under Sensor with Bayer Matrix doesn’t have any effect. Would that be because the camera is too new and the filter needs to recognize it to know where to expect the artifacts? Or is there something I might be doing wrong?
Hi @clockworknapkin! The PDAF lines filter only works on cameras for which the PDAF line pattern is known. The α7CR’s pattern is not known to RawTherapee which explains why the filter has no effect. If you have an image you are able to provide that shows the PDAF lines clearly, we can use that to determine what the pattern is and add it to RawTherapee.
Hey, thank you, that would be great!
I had a quick look at the code, found the pdaf patterns are stored in camconst.json. I was hoping sony a7r4 would be there, since a7cr uses the same sensor. It’s there, but alas no pdaf pattern for it either.
Here’s the raw with the artifact lines visible. They don’t appear on the sides, so I’m not sure if it’ll be sufficient for you to determine the entire pattern across the frame?
DSC01652.ARW (64.5 MB)
Thanks for the image! I will see if enough of the PDAF lines are visible to deduce the pattern.
I just created a pull request containing the PDAF pattern. The pattern I found is not precise. It repeats every 420 pixels, except some rows are different. A better sample image would be needed to confirm if the pattern is larger, but I think the 420 pixel pattern is sufficient given that it only falsely flags a few rows out of every several hundred. The effect on falsely flagged rows should be minute or invisible.
I have just merged the pull request into dev. The PDAF pattern will be in the next release.
