I will be contacting support, although from past experience they aren’t too responsive.
The point of this whole thread is in the subject heading, and clearly shown in my post above yours, with RAW and DOP’s provided for anyone who cares to try and duplicate on there system. Your example above also does not disprove what I’m saying or seeing, or disprove that there is a clear difference in HQ preview vs magnifier. I can also select a region with the magnifier that looks like the preview, does that mean it holds for the whole image ? No. Which brings back how ridiculous having the magnifier the only way to see the other Prime noise removals before export.
As for why I’m doing this, it is to test noise removal in real world case of long exposure for astrophotography, and see how well DxO handles these hot pixels. These hot pixels become clearly visible in the actual light frames.
My old Canon cameras have the means to capture a dark frame, which is then factored into the resulting image. Downside being the extra that is needed. Does your camera provide such services?
I believe for that purpose exists software that can do dark frame operations with raw files converted to DNG (not demosaicked) files and then resulting DNG after whatever pixel suppression executed can be fed to DxO PL ( with extra operation to supply makernotes from original actual image raw using exiftool if raw to DNG converter did not supply that )
Yes it does, but I wanted to test if DxO is good enough to do it, that saves me half the time, as the Sony takes a dark frame the same exposure time as the shot, so in my case 30 sec. That would be awesome if DxO does. In the same way you can do lens corrections out of camera as well using DxO. Thats why this whole thread came to life.
How can DPL distinguish a star from a hot pixel? As of today, it can’t, but maybe you can reap some benefit from combining a highly denoised image and its original…assuming that NR will swallow the stars rather than the hot pixels.
PL should be able to do so or at least extremely close to it as PL processes the raw bayer before demosaic step.
A stuck or dead sub pixel of the matrix should be possible to detect and allowed to be suppressed by the hot pixel algorithm control slide.
It is for me on my D3S and D3X - that said when I there are two stuck sub pixels next to each other PL is unable to cancel them out.
