老外的测试结果
Just upgraded the firmware and ran a series of tests (similar to previous ones with Christmas tree lights). I was easily able to duplicate the issue using extreme underexposure around the lights, and with the new firmware, I'm happy to report that I cannot find even a single black dot.
I ran the raw files through DCRAW using document mode (not demosaiced) and examined the area around the lights to look for any anomalies (pixels below the black threshold) in my astronomy image processing software (MaximDL) which shows pixel statistics for selected areas. There were none to be found, whereas in previous tests, there were hundreds.
I also looked for signs of obvious "patching" of black areas, possibly where color data would have been wiped out on the right side or some kind of smearing. I could find none. The bayer mosaic looked symmetrical around the light.
Of course, the JPGs also show no black dots either.
I've taken about 10 pics with a wide swing of conditions. I'll try some more this evening, but I'm confident this already small issue is resolved. If I find otherwise, I'll post here.
|