Brian Sweeney
Forum Replies Created
-
-
Re: Sony A7R3 shoot compressed or uncompressed?Reply #1 on: September 15, 2019 at 6:31 pm
To see this one would implement the compression algorithm and run it on an uncompressed image and store the difference frame.
The way you calculate the difference frame is important.
https://blog.kasson.com/the-last-word/a-tough-test-of-sony-raw-compression/
What I mean by this: Implement the compression algorithm in software to post-process an uncompressed raw image. Compress the image, decompress it and then subtract from the original uncompressed image. That will show what is being lost by the compression algorithm.
For doing a running sum difference frame on a Mosaic Sensor: do a running difference of elements of the same color on the same row. Store the differences as 8-bit signed values from -127 through +126, reserve 127 as an escape value to store the actual value for when the difference exceeds -127:126. If the encoded difference frame exceeds the original image size, switch to storing an uncompressed frame. Computationally expedient, best to store the original image if the running difference takes too much space. The point response of the sensor+lens is important, tends to work best on oversampled data. Worked 30 years ago.
My Micro-Nikkor AFD 200/4 is one of the sharpest lenses that I own. Online reports state that it is well matched to a 16MPixel sensor. I wonder how many Sony A7RIII images are taken using lenses that outperform the sensor.
Re: Sony A7R3 shoot compressed or uncompressed?Reply #3 on: September 15, 2019 at 1:03 pmhttps://www.ingentaconnect.com/contentone/ist/ei/2016/00002016/00000002/art00020?crawler=true
A paper on how ARW v2 (Alpha Raw) works, and a program available to correct artifacts. Worth reading just for the reference to Blondie. But anyone with the album would realize the parallel lines were B&W.
Re: Sony A7R3 shoot compressed or uncompressed?Reply #4 on: September 15, 2019 at 9:37 amSome web-surfing: One reference that the compressed mode raw of the Sony A7RIII uses 12-bit pixels when shooting burst mode and long exposures. There was a link to a Sony statement about the switch to 12-bit, but Sony seems to have moved or removed it- link was dead. The first step of the Sony compression algorithm on the older series cameras (at least what has been documented through users online) is to reduce the 14-bit raw value to 12-bits. If you require pulling out details from shadows, it is best to have the full 14-bits available. The other effect this has is to lose fine detail in the image: 4 intensity values are binned into one value. This is like making a contour map of the (intensity/ 4). Noise in the image can move pixels near the edge of a contour line to the next bin. To see this one would implement the compression algorithm and run it on an uncompressed image and store the difference frame.
If you like the results with compressed mode- use it. If you cannot see a difference, it works for you. If you do a lot of post-processing, uncompressed raw provides more information to work with. Memory cards are huge these days. You can always shoot uncompressed, and “ZIP” the raw files on your computer to store them when not in use or for archiving. Algorithms such as ZIP will compress an uncompressed raw image without decimating the data.
Re: Forum & Site UI: Bugs, Features & CommentsReply #5 on: September 15, 2019 at 7:21 amhttps://photopxl.com/forums/topic/sony-a7r3-shoot-compressed-or-uncompressed/#post-7835
I did this a third time- and now it took.
I hit the submit button each time, as noted by the edit history.
Odd.
Re: Forum & Site UI: Bugs, Features & CommentsReply #6 on: September 15, 2019 at 7:10 amUsing chrome running under Win7 SP1- made an edit to a post, the added text did not show up but the Post timestamp indicates that I made an edit.
This text was made by Kevin to check to see if an edit appears.
And this is a test edit- which worked.
I went back to the problem post, did another edit- the text did not show up. The edit history was updated.
Re: Sony A7R3 shoot compressed or uncompressed?Reply #7 on: September 15, 2019 at 6:53 amhttps://www.colbybrownphotography.com/getting-the-most-out-of-your-sony-a7r-iii/
According to the above article, and “some quick web-surfing”, Sony offers uncompressed RAW and lossy-compressed RAW. The lossy-compressed RAW can cause problems in certain instances. I’m surprised that Sony does not have “Lossless Compressed”. The original series only offered lossy-compressed RAW, adding uncompressed RAW was done after many users complained. The problem with the lossy-compressed RAW format is that it is “Chaotic”, you don’t know what will be lost as it depends on the structure of the image. With the processing power and buffer storage available, lossless compression is not hard to implement and will not produce the artifacts of the Sony algorithm. With the processing power and buffer memory available, lossless compression schemes are available that avoid the artifacts produced by the Sony algorithm. The running difference algorithm requires very little processing power and buffer space, does well with oversampled data. Given the pixel count of the Sony sensor and resolution of the lenses used on it, this algorithm would do well.
Once you understand how the artifacts manifest, it’s pretty easy to look at a scene and decide if you’ll need lossless or not. 95% or the time I don’t use it.
These algorithms tend to magnify the effects of noise on the image, noise is high-frequency and lossy algorithms do poorly with high frequency.
In most circumstances, noise tends to cover up the artifacts of craw.
Unless Sony offers a Lossless-compressed mode, shoot uncompressed for anything important.
I think that’s a huge overreaction. I don’t hesitate to use craw in many circumstances where quality if paramount.
Jim
Most people do not understand how the artifacts manifest themselves, and don’t have the time to assess if a particular scene will cause the issue. Sony is only attaining a 2:1 compression with a chaotic process. They could use a simple running difference scheme with an escape value to store full-values if the 8-bit difference had been exceeded. They could use Huffman code to store the frame, similar savings. With the processing power and buffer storage available, there are algorithms that could be implemented that do not produce the artifacts of the Sony algorithm. The running difference algorithm would do well given that the sensor is oversampling the resolution of most of the lenses used with it.
