Last Jpeg test

[JZaun]

I can't figure why we get different results. I did this one different.. I took a full size jpeg file and save it as test 1

I opened test 1 in PSE2
I did enhance
auto levels
save as (test 1)
(file exist? overwrite?) I say yes
close file
open test 1 again
Repeated 20 times
100% crop.
save for web
size 800 on long side.

Here is the results of 20 enhancements. Surely I wouldn't change a decent photo 20 times. and if it is not decent I would not keep it. Again I am satisfied that no matter which test is correct the Jpeg compression issue is not a issue for me.

Test 1


Test 20


JZaun

[slejhamer]

Originally Posted by JZaun

Again I am satisfied that no matter which test is correct the Jpeg compression issue is not a issue for me.

Agree!

But whatever you are doing to avoid the degradation, find a way to market it!

[stopbath]

Has the file actually been changed for each new save? It is possible, that the knowing the degredation known to exist in JPEG that a simple check is made on the history of the image. If no change has been made, the the file is just 'touched' (not actually modified, but the file listing has a new time on it.) Same could exist for 'save as' if nothing has been changed, the file could be just copied. This would benefit the user as an unchanged file need not be reprocessed.

Perhaps redo the test, and actually modify the image by placing a white pixel along one edge. Save as a new image. Close it and Load the new image. Modify that, and save as a new image... after you have twenty images, compare. If the image with twenty white pixels has no other artifacts than the image with none, or ten white pixels, you can feel safe. But if at 10 you start seeing problems, then you know your limit is 9 saves for that image. Other images may vary...

Use an image with a variety of colours and textures.

Also what level of quality are you saving at and with what program. Each program saves differently at different levels (there is no standard on how a program saves. What is 'best' quality on one program may not equate to anothers 'best' quality...

[PacAce]

Jerry, here's something you can try to see if you're actually using a buffered copy of the file instead of the save copy.

Instead of saving the jpg file in high quality mode, save it in very low quality mode. Then repeat what you've been doing. If the quality still remains very good, then it probably means that PS is always using a buffered copy of the file you're saving.

If that's the case, the only way to do a realistic test is to either exit out of PS after doing the save and then going back in again or renaming the file you just saved to another new name outside of PS so that when you load the file, PS will have to physically load the file from disk instead of using the buffered one in memory.

[JZaun]

It is my understand that Jpeg is only compressed on save. So if you load a pic, mess with it and save it under a new name the original doesn't change. So again I just don't understand all the fuss about jpeg not retaining quality. There are easy ways to deal with it. For me the extra time and work dealing with raw wasn't worth it. Maybe in time I may get better with PS and will be able to make a raw better than jpeg but I have not found that to be so yet. As with many things, and I can name quite a few, the jpeg issue is no where as big as some make it out to be. We have gone through this with cheap batteries that blow up cameras, (no one has come up and shown that it happened to them yet) to CD's that get eaten by fungus,(I have CD's over 5 years old) to just junk canon lens that no one should buy. Some people just have to take sides and worry the point to death. I posted my test to point out just that and it did If anyone likes useing raw by all means do so! I am satisfied with Jpeg for now!


JZaun