Can I change option for Consolidating Duplicate Fonts?

[Donovan Graen]

I work for a textbook publisher. Our materials are constantly be proofed, updated, and repurposed. Several of the titles are over 1,000 pages. We are experiencing very significant hangups when it comes to "Consolidating Duplicate Fonts" when we replace the initial page of these documents, and then again when we need to do a "Save As." Sometimes this process can take upwards of 45 minutes. This has created a production nightmare without any resolution I can find to the problem. Is there anything we can do to speed up or bypass this process without bloating the file sizes or degrading press-ready compatibility?

[Luke Jennings]

I'm assuming you are talking about PDFs? This is only a guess, but perhaps the fonts in your PDFs are fully embedded, as opposed to embedded subset. Either way will produce a usable PDF, but embedded subset might consolidate duplicate fonts faster. The disadvantage to subsetting is that only the characters used in the document are embedded, so if you want to make type edits using new characters, you are out of luck. You can go to File> Properties> Fonts to see if and how your fonts are embedded. As a test, you could try to optimize your PDF to subset the fonts (PitStop can also do this). How fast is your Mac? perhaps this is a perfect reason to get a new one?

[Donovan Graen]

Thank you for your input, Luke. And, yes, I am talking about PDFs.

In normal situations, your suggestion may be a viable option. However, our printer and the vendor we use to generate/post our digital editions require completely embedded font sets (no subsets). But, I will give it a try, just to test the theory.

Our Student Edition textbooks actually run through this process at a resonable time; taking about 2–3 minutes.

Primarily, the issue happens when updating pages within our Teacher's Edition. The teacher's book pages contain an insert of the student's textbook pages. Those imported graphics for the insets are also PDFs, made from a cropped press-optimized version of the student's pages. It's my thought that is what is causing the trouble. Otherwise, this text in the teacher's book is quite simple. I'm fairly sure the app has to basically read/consolidate each of those particular pages twice. But it still shouldn't take as long as it does.

Typically, we use only four font families: Avant Garde, New Century Schoolbook, Lubalin Graph, and Jacoby. We do use a few others randomly for illustrative purposes at times, but all-in-all, we keep fairly simple in regards to font usage. I really don't believe the issue has anything to do with the quantity of fonts being used.

[Donovan Graen]

@Luke: Oh… forgot to mention… I'm using a 3.33 GHz 6-Core Intel Xeon Mac with 12 GB of RAM.

[Test Screen Name]

Why do you only generate selected pages? Have you found it is quicker? In my view, it's a logical sounding way to work, but often much less efficient and more error prone than making the whole PDF. Depends on your workflow, as I don't think you've said how they are made (i.e. what applications and options).

[Donovan Graen]

@Test Screen: The books i'm referring to typically have 20 body chapters (not including the Table of Contents, Introduction, Resources, Glossary, Index, etc.) We build the books in Quark by individual chapters. Each chapter ranges between 30 to over 100 pages. We generate PDFs seperately for each chapter, and then at the end, marry them into the final pagination. We do it this way for a lot of reasons, as I'm sure you can imagine. Usually our corrections/changes only affect between 3 to 12 pages in each chapter, so it is by far the best way to us to regenerate those few files and reinsert them as we go.

[Luke Jennings]

Well, that should be adequate. As a test, optimize the insert pages, discard objects (all), discard user data (all), clean up (all). now how long does it take to consolidate? Test 2: convert all of the text to outlines on the insert pages. Test 3: Save all of the insert pages as a jpeg or PNG, now how long does it take to consolidate? saving the inserts as hi-res images would probably be perfectly fine, particularly if they are reduced in size. If these ideas don't help, then I would consider using InDesign (book tools) to organize the various PDFs.

[Donovan Graen]

@Luke: I like the optimization option. I will test that. The other suggestions really wouldn't work for us, as it is required that the imported insets MUST be searchable. Converting them to a rasterized format or outlining the fonts would cause the loss of that functionality.

[Luke Jennings]

I would talk to your vendor, embedded subset fonts should not be an issue, unless they are doing type edits to your final files. Were you able to reduce the consolidate times with subsetted fonts? most people consider an embedded subset PDF to have properly embedded fonts, for printing anyway. I'm guessing you are using Distiller to make your PDFs from Quark .ps, you are probably aware you can adjust the Distiller settings to subset the fonts. If you are exporting directly to a PDF from Quark, subsetting is the only option (I believe).

[Test Screen Name]

It sounds as if the PDF files have pathological numbers of fonts. I have seen this e.g. a different copy of the font for each line. You might be able to adjust settings or find a fix within the PDF creation process. How are they made?

[Test Screen Name]

Though on another note... why not just recreate the PDF for each and every change? Is that slower still?

[Donovan Graen]

@ Test Screen Name; We do only generate the pages that require corrections.

The issue occurs when we put those updated pages into the 'whole-book' file of our Teacher's Editions.

[MMP Longwood #1]

I am not sure if this will help you but I also had some issues with a file that had 12 fonts that were repeated and subset in my PDF document about 400x each, as a result the file was large about 132mb but I did not think that our printer could not handle it.

It may not seem cumbersome but boy did our printer choke...

Acrobat also did not have any FAST Viable options for me to consolidate these subsets with just the whole font, So after a few hours messing with files I found a simple elegant solution,

open the pdf in acrobat and save is as a .ps

I then take that file into Distiller and use a custom preset to embed the original fonts and UNCHECK SUBSETS. and the resulting pdf is around 5.1mb

Good to go! Hope this helps others.