• Best Wishes to all for a Wonderful, Joyous & Beautiful Holiday Season, and a Joyful New Year!

Image Compression

Easy one. What is the general consensus on Zip versus JPEG compression in the Adobe PDF Export Presets? I've always done ZIP, but I'm reading a lot about JPEG at maximum quality being adequate for 175lpi. The same file with the images uncompressed (test) was 381mb, with Zip it was 91mb and Jpeg was 19mb. I'd say better safe than sorry and go lossless, but the IT department wants to save every byte they can. Ha anyone seen a noticeable difference in image quality between Zip and JPEG?
 
If you are in the business of publishing, regular JPG compression should be the last thing on your list regardless what IT people want to save. I don't think I would ever trade quality vs hard drive/server space. 1TB drive can be bought for less than $100 these days. What are they thinking?

Going back to lossy vs lossless, it's a yes and no question. The difference should be noticeable enough to any image professionals. It's often "no" if you are talking about average consumers. I, for one, could always tell when looking at those crappy PDF/X1 magazine ads...printing on toilet paper just make them jump out even more to my eyes.
 
Last edited:
I have looked alot at images, and would say that for a final PDF to go with Maximum JPG is fine.
(for an add you would not be doing a final since the add is usually placed in a layout program)
If you are planning on doing any post processing like sharpening however then I would have it in lossless, since artifacts would be enhanced at sharpening.
Most of the time you see poor quaility stuff the place quality is lost is other places. There are so many ways to destroy a picture. I'd say CMYK workflows with curve adjustments after converting to CMYK are a bigger cause of bad images. Many images are just plain bad because there are more people with less knowledge as content creators.
 
If you do NOT have a requirement of PDF/X-1a or PDF/X-3, then I would recommend going with standard PDF 1.5 or 1.6 and using JPEG2000 compression. It offers a lossless mode (for those that want no loss) that still gets GREAT compression ratios compared to the classic methods.

Leonard
 
I use JPEG max quality as default image compression in 99% of the PDFX1a I send to publications (and I send a lot). 2400 dpi proofing result and printed results are excellent. The only exception is with documents containing images that are CMYK but have an empty channel like nothing on the cyan channel. I will then use zip compression. I had an ad once that only had magenta, yellow and black information but nothing in the cyan channel that, once turned into PDFX1a with JPEG, showed some very light cyan zones in the cyan plate of the PDF. Changed the compression to zip and the resulting PDFX1a was just like the original image channel wise.
 
Leonard, I would prefer X compliance just because of our customer base. That way they can preflight their own PDFs through my Acrobat Preflight profile. Is there a preflight profile available for 1.5 or 1.6? I am starting to lean toward X-4 though.
 
With jpg ther is always the risk of unexpected result due to the compression capability, with zip you don't have to think or check - you know.
Edwin
 
That's what I'm saying Matt. If I were the designer or were accepting only native files then 1.5 or 1.6 would be the way to go. Since I the only way I can make sure I get half way decent files is to require they are X-compliant after running the designers run them through the preflight profile I have supplied to them. I figure if I give them my .csf, .kpf, .idpp and .joboptions (Export Preset), then I have a fighting chance of getting something rippable and printable.
 
For any book projects I've always used ZIP, for the same reasons that Tech stated.

Interesting bit about JPEG2000, though.
 
Well if you really want to control the conversion process then you have to look at deploying Enfocus Instant PDF for every client, or Enfocus SWITCH with FlightCheck Online, or some other local PDF creation and delivery process.
 
We don't compress anything. Just another variable to worry about, and in print there is enough already. If I had to choose then it would be zip for the lossless.

Good luck
 
Leonard, I would prefer X compliance just because of our customer base. That way they can preflight their own PDFs through my Acrobat Preflight profile. Is there a preflight profile available for 1.5 or 1.6? I am starting to lean toward X-4 though.

PDF/X-4 fully supports JPEG2000 and would be an excellent choice!

Leonard
 
First problem: the basic issue with JPEG compression is that each compression - then each saving - destroys the picture a little bit more...

... so ONE compression in maximum quality will never be noticable, even by expert eyes or using great magnification... but some successive compressions, applied to the pictures at different steps of the job by different people, will become visible even at max quality...

... and when you work on JPEG pictures there is no way to know exactly:

• how many times the pictures have been saved in JPEG mode before you get them??? (and exporting your PDF with JPEG compression will add one more!!!)

• how many times the pictures will be re-saved in JPEG mode after you send your job (native files or PDF)... OK, normally, a final ready-to-image PDF is at the end of the process, just before imaging and there should not be any more saving... but are you absolutely sure that your job is absolutely perfect, and that the pre-press will not need to open your pictures in your PDF for some technical reasons and re-save it once or twice?

So, when I have to make a PDF from a customer native file or from one of my file using customer's pictures, I never use JPEG compression, I either use ZIP compression or no compression.


Second problem: Quark and Adobe have two opposite ways to handle the JPEG compression settings in their sofwares... Adobe uses the quality, Quark uses the efficiency of the compression... the "maximum" setting of one is equivalent to the "minimum" of the other: it's then easy to mistake (even for aware users) and using ZIP compression is the best way to be sure to avoid any confusion in any cases.
 
While I agree with claude72 in theory, in practice I'm not convinced multiple JPEG compressions of an image is an issue.

I took a low resolution version of one of my photos and did a small test.

1-2.jpg


Because it's low res image any jpeg artifacts should be quite visible:

However after saving it 10 times (i.e. save as, reopen, save as, reopen, etc) using the "High 9" setting in PShop rather than the "Maximum 12 setting" a total of 10 times, the only way I could see a difference was to put the 10th version on a layer above the original and use the layer "Difference" mode to determine which pixels had changed.
Here's the strongest area of difference from the original (on the left) and the jpegged 10 times (on the right) identified by PShop:

Sidebyside.jpg


Download and the images and compare for yourself - the differences are completely insignificant. With a higher res image the differences would even be less visible - especially once the image has been halftone screened since any artifacts would be too small for the screening to resolve.

So, while I only use jpeg compression for web images - I doubt that there is an issue with jpeg in print application unless, as claude72 points out the compression type selected is max compression with minimum quality.

best, gordon p
 
I'm sure I read somewhere that Photoshop Save As doesn't degrade jpeg unless you've edited some pixels. I'm interested if there was a difference between saves 9 and 10. it could be that all the difference was generated at the 1st save.
 
I did the following test using Photoshop CS4, opened a CMYK tiff image, resaved it as JPEG quality 10 and named it "original jpeg", creating a master I would compare to. Duplicated the "original jpeg" on my desktop and named this new copy "jpeg resaved". Did the following 6 times: open "jpeg resaved" resave as JPEG quality 10 on top and close. Compared both images in Photoshop, there is definitely some areas where artefacts are stonger compared to the original JPEG. While some might think no one will ever notice, one thing is certain: each new generation of JPEG resave affects the image integrity. Attached are 2 snapshot where you start to see artefacts on the resaved version.
 

Attachments

  • original jpeg.jpg
    original jpeg.jpg
    5.7 KB · Views: 227
  • resaved_6times_ jpeg.jpg
    resaved_6times_ jpeg.jpg
    5.8 KB · Views: 228
[SNIP] Compared both images in Photoshop, there is definitely some areas where artefacts are stonger compared to the original JPEG. While some might think no one will ever notice, one thing is certain: each new generation of JPEG resave affects the image integrity.

Agreed, but if after all that you need to look at individual pixels in order to see the microscopic changes - then, IMHO, any loss in image integrity will not show up in the press work in real world production.

best, gordon p
 
Someone mentioned looking at real-world useage.

Back when the industry was transitioning from the DDAP standards over to PDF/X and the default compression in PDFs changed from ZIP to JPG we ran a few test pages of the same PDF compressed with ZIP compression and JPG (max quality). The starting image was a 300 dpi scan saved as a Tiff placed in Quark so the only lossy compression being applied was the final JPGing as the PDF was made.

People couldn't reliably tell the difference between the ZIP proof and the JPG proof. It was literally a 50-50 split.

When we created our standard PDF joboptions we gave people the choice to use either type of compression. Customers overwhelmingly chose the JPG settings for the better file size, especially when sending files in over the Internet, since they could be about ten times smaller.

During a recent update I recently the ZIP settings from our website, since they really weren't getting used. Our internal Prinergy workflow still uses ZIP compression, but any PDF that comes from an external source is probably JPG compressed. And we're printing with a 20 micron FM screen.

I generally tell clients to avoid re-saving in the JPG format over and over, but saving them once or twice with max quality JPG compression isn't going to adversely affect image reproduction in real-world printing, even with finer screens.

Shawn
 

PressWise

A 30-day Fix for Managed Chaos

As any print professional knows, printing can be managed chaos. Software that solves multiple problems and provides measurable and monetizable value has a direct impact on the bottom-line.

“We reduced order entry costs by about 40%.” Significant savings in a shop that turns about 500 jobs a month.


Learn how…….

   
Back
Top