compounding JPEG compression…

D
Posted By
danser
Jun 2, 2004
Views
272
Replies
4
Status
Closed
Showdoc and Graham,thanks for clarifying this.

I need to convince a client of this–do you know of anywhere the fallacy of the JPEG recompression myth is "officially documented"?
– is this true of all JPEG tools or just some like Photoshop?

My client has got it into their mind that changing metadata and resaving a JPEG causes a quality loss.

How to Improve Photoshop Performance

Learn how to optimize Photoshop for maximum speed, troubleshoot common issues, and keep your projects organized so that you can work faster than ever before!

GP
Graham_Phillips
Jun 3, 2004
Where did you drag this thread up from? 🙂

Assuming that all JPEG tools do not contain bugs in the compression/extraction algorithm, then an image created using one program should open in another program and be resaved without loss provided that it is resaved at exactly the same quality setting. This is where there could be a problem. If one application provides a Low, Medium or High setting and another application provides a scale from 1 to 10, it might be impossible to marry the two together and use a common setting.

When one does not have the mathematical background to comprehend a problem such as this (and I include myself in this), I’m a great fan of "seeing is believing" as an alternative. Demonstrate the process in front of the client. Take a couple of image styles, e.g. one with areas of smooth changes in tone, such as a photograph, and one with boundaries with very sharp contrast, such as a line art drawing. Save as JPEG, choosing a quality setting. Open the saved JPEG and compare the differences; there are probably many, depending on which setting you chose. Now change the metadata, resave as a new file, using the same quality setting. Close and reopen this file, change the metadata and so on. When you get fed up, compare your last iteration with the first one you saved.
RG
Rene_Garneau
Jun 3, 2004
I compared Image1.jpg with Image5.jpg and could detect no difference, even when zooming in.

Graham,

I did your test, and there is a difference between Image1.jpg and Image5.jpg. (Using PS, same compression). I did "showdoc" test as well and there is a difference.

Therefore, saving a jpeg over and over without any changes (using the same compression) will degrade an image.
GP
Graham_Phillips
Jun 3, 2004
What the !©^?*&%®

Oops.

I owe everyone who’s contributed to or read this thread a big apology. I just repeated the tests on a screen grab of this thread using JPEG quality 3 and saw clear deterioration after at most 5 iterations. Aghast, I then repeated the process using JPEG quality 12. Harder to see, but artefacts are definitely there.

Sorry everyone for misleading you, and thank you Rene for challenging my findings. I’m off to find a corner to stare at.

Graham
R
Ram
Jun 3, 2004
Graham,

Before I abandoned JPEGs alltogether eons ago, I conducted similar tests and I always saw a degradation. On a few particularly difficult film scans, I even saw a degradation the first time I saved the PSD or TIFF image as a JPEG at maximum quality.

Fortunately, for my photographs I never need web images nor do I send out or exchange image files, so abandoning JPEGs completely was easy. Unfortunately, the majority of users don’t have this luxury.

How to Master Sharpening in Photoshop

Give your photos a professional finish with sharpening in Photoshop. Learn to enhance details, create contrast, and prepare your images for print, web, and social media.

Related Discussion Topics

Nice and short text about related topics in discussion sections