What is lost by converting a PDF 7.0 file to a 5.0 file

[improve]

I have some very large Color PDF files (one is actually over 1 Gig in size) and when I print them to a 5.0 file they are about 5% of the original file size. Can anyone tell me what it is that I am losing when I make that conversion? Visually, it appears to be no different on the screen.

 

[GordonIt]

What do you mean by "I print them to a 5.0" ?
What also do you imply when you say they are about 5% of the original size?
Because; I wonder if you compare a PDF 1.6 (Acrobat 7) with a PDF 1.4 (Acrobat 5) or if you care about compression issues.

5% of original size most probably means that your 5.0 printer have the Compression option turned on and your original is not compressed,
AND AND AND!!! maybe also that the 5.0 printer setting allow to resample image to a lower resolution if it's over a specific tuning.


Then not only is the compression impacting your original data (Jpeg is irreversible in 99.9% of the case) but maybe also that your images are of lowered quality.


However a very large image is that big that most probably this is a "scientific" answer but the processing is OK for final handling. Just use latest version of Reader and check at 6400% on "hard edges" like a dark to light transition, DO you see some "spreaded ghost pixels" in the light area?


Also to mention a major difference between 7.0 and 5.0 (which doesn't apply here as you are using 5.0) Is the Jpeg2000 compression which only exist in 7.0

This compression is far more efficient than Jpeg (around 400% More efficient!!)
BUT your file can only be read by PDF 1.6 compliant product (like latest version of Acrobat Reader for 7.0 rev)

Hope it shade some light.
Bye

 

[improve]

Gordon, thanks for the interest in my issue.

I am using Adobe Acrobat Professional 7.0 (not 5.0). I am only saving it to 5.0 because it seemed to have drastically reduced my filesize. If I can do this without leaving 7.0, much better solution, even if it can only be read by 1.6 compianct product. I can always save-as if I need a document to be compatible with an earlier product.

When I say print to 5.0, I mean print to Adobe PDF (instead of my laser printer) and set the properties to Compatibility of Adobe Acrobat 5.0 (PDF 1.4), Object Level Compression to Tags Only and Resolution to 600dpi.

I'm not very knowledgeable about these settings but when just playing around to see how to shrink these very large files, I discovered that a 500 Meg file when printed in this capacity comes out to be a 50 Meg file (thus the 5% assertion).

Strangely, they are set to Black and White as well on the Paper Quality tab, but it did not actually remove the color, only changed it slightly.

I would happily do this to all of my large files if I knew that I was not losing anything important in the metadata. Not sure why you reference jpg in your email? What does jpg have to do with these pdf files? I don't think that I am saving to jpg since I can still add comments and OCR the text and highlight the text. Am I misunderstanding?

Is there a way for me to check or turn on the compression for the 7.0 file and see what difference this makes instead of going to 5.0?

 

[GordonIt]

If you are using 7.0,
Once installed, there could be 2 printer created (on my machine) depending upon option during install

"Adobe PDF"
and
"Adobe Distiller"

In fact what's behind the scene is exactly the same Adobe PDF engine, BUT the way you can setup or edit setting are very different, so I first need to know which one you are using.

The "Adobe PDF" on my machine, once you enter
Printer->Adobe PDF-> Properties -> Printing preferences (or more directly with right click on printer -> Printing preferences)
Have a Tab called "Adobe PDF Settings" which contains 4 drop down menu
"Default Setting"
"Adobe PDF Security"
"Adobe PDF Output Folder"
"Adobe PDF Page Size"
and some check button.


The "Adobe Distiller" printer, fro the same tab show a different menu, mainly a unique dropdown menu called "Conversion Settings" and just under a button called "edit conversion settings".

That could be a way to distinguish between the two.


I suspect you use "Adobe PDF"

Anyway once you enter the Editing menu for Preferences you might have a dropdown box called "Compatibility", then choose PDF 1.6 (Acrobat 7)
( I suspect you exactly see where is this tuning based on your answer)
This allow every kind of compression to be allowed but could make your file requiring Acrobat 7 reader…


Then on the left pan click on "Images" you will see the tunning for Images compressions

You have 3 different areas, one for each color model.
You have a "Compression" dropdown menu, then choose "Automatic Jpeg2000", probably the best choice in your case.
Save and try it with your files.
You can experiment changing the "Image Quality" choice.
With Jpeg2000 Medium is higher than with the traditional Jpeg so you can safely decrease it.

Also of very high importance if you want smaller file you can change the threshold of resolution above which image is 'DownSampled' (means reduce resolution) to a lower res.
I use "Bicubic DownSampling" (best quality) to 400 dpi (tuning on the right)
"for images above" I set 600 dpi.
You can play with that. Don't go under 300 it's not good and not over 600 it's overkill unless you have a specific reasons.

You might achieve not only 5.0 kind of result but even smaller for higher quality.

Regarding jpg:
Jpeg is a kind of compression mathematically called DCT (Discreet Cosine Transform)
Jpeg2000 is another kind of compression using a mathematical model called Wavelet (far more clever, and curiously as old as DCT…!)

And Jpeg stand for Joint Picture Expert Group, a standardization organization

And Jpeg defined as well a File format, as a data format.
PDF use the Jpeg Data format (same internally as a Jpg file use) but enclose it within PDF object.
So you could be using Jpeg compression, as in JPG file, inside PDF file without even knowing it. Most probably the case with you smaller 5.0 file.
Generally Jpeg create some artifact around 'hard edges" like a Black to White, you see some slightly darker pixels in the White area.


I don't fully understand your concern with MetaData.
PDF Metadata as Adobe call it and use it, have nearly nothing to do with Image size or compression.
And it is not sure that generating a new PDF will preserve your MetaData anyway.

 

[improve]

I have no idea what has changed but now when I try print to "Adobe PDF" (your assumption was correct that I am not using Distiller), it now creates the following message:

%%[Page: 1]%%
%%[ Error: ioerror; OffendingCommand: imageDistiller; ErrorInfo: DCTDecodeFilter JPEG marker 0xAF disallowed in Huffcoded region ]%%

Stack:
1.0
-save-


%%[ Flushing: rest of job (to end-of-file) will be ignored ]%%
%%[ Warning: PostScript error. No PDF file produced. ] %%

Any idea what is happening?

 

[GordonIt]

Looks like a corrupted data in a Jpeg section of the file.


But I am completely mixed up here.

Are you printing one of your original big file ?
or is it a 'converted' file ?
Is Original big file in Acrobat 5 PDF ?
Is the one you print here in 5.0 or 7.0 ?

Are you printing to a Acrobat Distiller 7.0 or a 5.0 ?
it looks like you have both on the same machine.

How do you proceed to print ?
Open file with acrobat reader then print ?



First thing I would do, is be sure that an existing file can still be processed the way you used to.



I can't think of any joboptions in Acrobat 7.0 that can be source of this problems.
But you must better be using a consistent revision of Acrobat along your process, or use an upward compatible schem, 5.0 pdf-> Reader 5.0 or 7.0 -> 7.0 pdf done from print, -> ONLY 7.0 tools (unless very careful to select 1.3 compatibility in printing options

 

[improve]

I use Acrobat Pro 7 ONLY. I am using it on two computers. The files that I am working with were scanned in by the local UPS store (yes, they have a scanning service). They scan into Acrobat 7 as well. They seem to have either never compressed them or used 100,000 dpi or something. I am trying to get them to a reasonable size without losing anything within reason (metadata, 600 dpi resolution, compatibility with most readers (probably 5.0 and greater)).

I print from the normal print icon or file print menu item and then select Adobe PDF as the printer. Turns out, I do not even have a Distiller print option.

I believe that the file that I printed that generated the message above was a "preconverted" file 7.0 file, though it is possible that it had already been converted.

Last night, I noticed that some of my files may have become corrupted, perhaps during a batch OCRing. When I open them, some have portions of pages blanced out (not sure how to describe but part of page is random colors like US televisions sometimes have when there is used in the emergency broadcast system). Other times, when I come upon a certain page, it presents a dialogue that says that there is "insufficnet data for an image". This is on half meg files even. This may have been the case on the files producing the error in my previous entry. My computer has 512MB RAM.

 

[GordonIt]

OK more clear now.
I think that you found ou the answer with corrupted files.

I don't see any reasons your ram is too small. Worst it take hours to open it.
Check also available space on the disk containing Acrobat,
and on disk containing "Documents and Settings" (generally the same and is C you need to leave 2X at least of your file size. Looks like 4G of free space.

Available space (or unavailable space) can be a reason for some problems.
Acrbat do clean up when exiting so it will give it back.

The file size you mention is rather unusual, looks also like this is a multipage PDF, can't you ask them to do multiple single page or decrease number of pages/PDF ?

Have you investigate Image to vector conversion, (not to text, just to vector that 100% respect original contents) solution ?

 

[improve]

I'll be in touch in a few days after I've tried to recover from the corrupted files.