Reduce file size of extracted PDF pages

[betsyp@ema-eda.com]

I am trying to use Acrobac DC Pro. I have a 100 page PDF document and I am trying to extract one or two pages at a time and save them as a separate PDF files. However, each single page I break out is 1-3 megabytes. How can I reduce the file size to a few kb? The broken out pages are 20 times larger than the original PDF!!

Betsy

[Email removed by moderator for your protection.  This is a public web forum.]

[try67]

That's not unheard of. This can happen because of a large, fully-embedded font being used in the pages, for example, or a graphic image that is shared across all pages in the original file. You can try optimizing the file, but it's not a magical solution. At a certain point if you want to reduce file size you have to reduce the file's quality, or make changes to it (like replace custom fonts with built-in ones).

[betsyp@ema-eda.com]

How can I change the font? What font types are compatible with the creation of a PDF?

Betsy

[Luke Jennings3]

From Acrobat, go to File> Save as other> Optimized PDF. I would first try selecting all of the Discard Objects, Discard User Data & Clean up options (unless there is something you need to keep) and if that PDF is still too large, you can include fonts (never unembed a font, but subsetting is OK) and if it's still too large you can downsample images.

[andrewo31439192]

Hi there i am having the same issue:

Original PDF:
44 Pages of text only layout
File Size = 232Kb

Extracted PDFs: 
After extracting all 44 Pages as 44  separate PDF files:
Total combined File weight of all 44 extracted pages is now 2,3mb
'Audit Space Usage' from extrated page:
Content steams -  3.2%
Fonts - 66.48%
Document Overhead - 29,97 %
Extended Graphics States - 0,35%

Wondering where the bloat is coming from on the individual extracted pages and if there is an automated script that can be used to address this at scale before or after extraction. 

I have attempted Fix above without result. All discard and clean up options checked and extract page saved. no gains in reduced file size.
 
  

[try67]

The issue is most likely due to embedded fonts. Each individual file must have them too, which can cause their file-size to "bloat". This can't be solved with a script.

[andrewo31439192]

Thanks so much for the prompt reply and taking the time to provide feedback.

Given your feedback and Audit Space Usage I definitely feel like we narrowing in on the issue here.
Screenshots from the Document properties also seem to support this. 

Original PDF:

Extracted File Sample:


Original PDF is coming from a Google slides download. 
What methodologies would you suggest for removing embedded fonts to reduce the extracted file sizes.
(Apart from reducing the quantity of faces used.)
Either at time of generation, or in Acrobat before extraction to indivudal files.

The 'bloat' seemingly occuring with the need to embed these fonts into each extracted file. 
 

[Luke Jennings3]

I'm guessing the original PDF contained one set of fonts for all 44 pages, when extracted, each page may have the same set of fonts (font size x 44).

If you recombine the 44 pages, what is the size? Is 2.3 mb an issue?

[andrewo31439192]

Thank you Luke - you and @try67 seem to be aligned on this issue. Thanks for the feedback.
Definitely makes sense that the embedded fonts are the culprit here.
For the usecase if the individual file sizes can be optimised then that would be beneficial.

Is there a solution to reduce the need for embedded fonts? 
non of the fonts that show up in 'Document Properties' show above reflect in the PDF Optimiser Panel



Would using different fonts in the origin material (Google slides Docs) be a solution? 

[andrewo31439192]

@Luke Jennings3 @try67 fonts are definitely the culprit here. thanks for your assistance. 
Reducing the quantity of fonts used in the layout has reduced extrated file size !! A win

To rephrase this question for an optimal result. 

Is it possible to have 0 embeded fonts in a PDF without converting the type to shapes
(which increases file size and negates the excercise).
Is there a particualar font that I could use in the software of origin to achieve this ? 

Ideally the below screen would have 0 items in it without having to convert to shapes.

[try67]

Yes, by using fonts that come embedded with the application, such as: Times-Roman, Helvetica, Courier, Symbol and ZapfDingbats (and their bold/italic variants: Times-Bold, Helvetica-Oblique, Courier-BoldOblique, etc.).

[Luke Jennings3]

I don't think so. As long as the PDF contains live type, the fonts used will appear in the font list. You can unembed the fonts by saving as an optimized PDF, which may make the PDF a bit smaller, but I suggest you do not do that, as unexpected changes to the copy are likely to occur when viewing and/or printing the PDF, even if all appears well on your computer. Another way to eliminate fonts from the list is to save the PDF as an image (bitmap), although the size would likely increase.

[andrewo31439192]

Thanks for runnings this down with me guys. Legendary Status. An interesting problem.

Insights here have already allowed us to reduce the file load significantly. Thank you!
A large archive to process so even small gains in final file size make for significant performance improvements. 
We are not concerned with print, only display. See use-case description at end.
See learnings so far at the end. 

@try67 the fonts you mention here the base fonts for Adobe Reader/Acrobat ? 
Would using the baseline fonts you mentioned @try67  resolve this?  
I have attempted to change the face to helvetica in acrobat pro and saved out
helvetica still shown in in the fonts list indicating some sort of embedded font file.
Is there a method to force the use of these default fonts thus potentially removing the the associated font file?

@Luke Jennings3 Seems that your assumption may be correct as the screen shots below seem to indicate that live type = some sort of embedded font file which must be duplicated for each file instance.
Hence a single file with 44 pages is smaller than the sum of all 44 pages extracted as individual files. 

 
Fonts in the font list do not appear in the optimzed PDF menu as embedded.




Workflow discription:
1. Origin -  Google Slides. (client generated) 
2. Download as PDF
3. Extract All pages as separate files
4. Upload Separate Page Files to a display program (this program is sensitive to total import weight) 


Learnings so far:
The less faces or subsets you use the smaller the end file size.
Use of Raster or conversion to shapes increases file size
beware of automated functions such as bullet point lists, these added additional font files to the list.

[betsyp@ema-eda.com]

Thank you for your quick help. My PDF files are now more than 20 times smaller. Thank you so very much.