I am not “re-compressing” any images. Perhaps you are thinking that I am comparing the JPEG images; the one with the other?
I am placing the source TIFF in the base layer and a JPEG conversion in the next layer up and then setting the mode to Difference. That is comparing the JPEG with the source TIFF, NOT another JPEG.
Then, I repeat the above with the same source TIFF but with different JPEG parameters. Please see my post that you objected to - where the two sets of JPEG parameters are clearly stated.
The value of the levels in the two Difference-between-TIFF-and-JPEG plots is showing the degree of deviation of the JPEG conversion from the source TIFF.
So does this 8x8 encoding method make my statement that a jpg doesn’t record each pixel correct or incorrect. I have always thought JPG worked in blocks of 64 pixels rather than individual pixels and tries to decode them back to the original image. Doing this once or twice is not a major problem but after multiple edits the errors compound and artefacts become more noticeable. That is why I use tiff or png instead of archival storage and subsequent edits.
I have to admit to being more concerned about the top of my head looking like a barcode nowadays than the graph on the levels command
I like the gimp ‘we will release the next edition when it’s ready’ view I was told a long time ago that somebody said something similar in the Rolleiflex factory to their line manager
16bit and 8bit edit using same settings in tone curve and levels in gimp:
quick and lazy edit few days ago from OOC jpg in mint and gimp
It’s a contentious area where’s working with raw or 16 bit all the time seems to be gaining ground
I can remember when not so long ago camera manufacturers seemed to caution against
Having worked for a long time though not all the time in a fine art background I couldn’t see why people get so upset
I think retaining highlight detail is where working with raw or 16 bit scores or problematic images that need a lot of work on them
It was good to see somebody wanting advice on how to get that overworked look
It’s helpful when people challenge preconceptions: dangerous to creativity being an expert; though some people obviously are in some areas
I think the answer is to experiment and make your own decision for your own style I go for a sort of flow chart approach where different images go down a different route
terrible thought: might a phone be the best solution in rare occaisions
AI will be able to recreate digital clones of ourselves in the future, if we take high resolution 32 bit depth pictures of ourselves, IMI it could probably be much better for the reconstruction process in many years.
(That is if we are not vaporized by a nuclear war before we reach such technology.)
They say, take a lot of pictures and videos of your loves ones and you will see them after they passed away with digital reconstruction. It’s weird a bit like when people have their pets stuffed and kept in their homes.
