Yesterday in my review coverage of the Sony A7R II I referred to the orange peel noise with Sony files, which is still there in the A7R II. It is visible in all the smooth (well, ought to be smooth!) tonal transitions in the 90/2.8 aperture series at ISO 100.
Glenn K, a geophysicist, writes:
The "orange-peel" artifacts around transitions look very similar to ones created around abrupt transitions by the discrete cosine transforms used in JPEG compression... perhaps Sony is using similar compression methods in their "raw" files.
Although probably not an issue for reasonable sized prints, it will play havoc with sharpening algorithms in post as well as upsampling algorithms.
DIGLLOYD: I would say not the “raw files” but rather the “raw data before encoding into the file format”. Indeed it wreaks havoc with sharpening. While I can double or quadruple sharpening on a Nikon D810 file with minimal ill effects, doing so on a Sony A7R or A7R II files looks godawful.
I have to be very careful with sharpening Sony A7R II files, just as I had to with Sony A7R and A7 II and A7 files. The raw files are pre-cooked. Very disappointing to see this practice continue with the A7R II, which has superb sensor quality.
Examples from the A7R (I see no difference in behavior with the A7R II):
- Sony A7R Pixel Quality: Texturing Orange Peel Noise
- Sony A7R Pixel Quality: Orange Peel Noise (Yellow Blue Bike)
The bad news is that I don’t think the effect is related to the Sony 11+7 bit lossy compression scheme. I could be wrong, but I think that Sony is applying excessive image processing to the raw data even before it is encoded into the 11+7 bit format. So even if Sony were to offer a lossless-compressed format, the cooking might still be there.