I'm not sure if this has been asked on the forum before and if it has, I'm sorry. Is there a way to embed only a subset of the fonts used in the document? I have a two page document written in Japanese characters, and when I convert it with the fonts embedded it goes from 183KB to 44.5MB. That is a HUGE difference and I'm not sure what I can do about it.
subset of embedded fonts(10 posts)
Hi, it cannot embed subsets only. The only solution I see is to create it without embedded fonts and ensure that the recipient has those used fonts (or you could send the font separately).
In a future version of doPDF we'll embed by default all used font subsets.
This is a quote of yourself:
So why is it that subsets are not to be/cannot be supported all of a sudden? In my opinion it makes doPDF completely useless, as a 182KB PDF-file with some text becomes a 2.99MB file when fonts are embedded (whereas if I create the same file with subset of fonts embedded in OpenOffice, the file is only 81KB).
No one said that subsets cannot be supported, just that they'll be in a future version. In fact this feature will be included in dopdf 7 which should be released in november or early december.
Right now if you want to reduce the size of the pdf you can choose not to embed the fonts.
How do I embed fonts? I converted some files and the fonts were not embedded with the default settings.
Make sure you have the latest version installed of dopdf, and then in the window that lets you select where to save the pdf file check the option "Embed fonts"
Possibly, my observation/problem fits to this:
I am using doPDF Version 7.3 Build 382 (for free) under win xp and I am also using fonts created by myself.
Today, I found out that for some of my fonts the Euro sign is not embedded correctly in my pdf files created by doPDF. I found out that this problem disappears by changing the Euro sign from a composed glyph into a simple glyph. But several of my fonts have also the Euro sign as a composed glyph but there is no problem there with the Euro sign while creating pdfs via doPDF. Afterwards I found out that composed glyphs from very commonly used fonts such as Times New Roman and Arial have problems with several composed glyphs such as Oslashacute and Aringacute. In my fonts, there is not at all a problem for all composed glyphs, ccaron for example works properly.
Some few months ago I found out that the font Tudor Rose and other fonts from James Kilfiger are not being embedded correctly while creating pdf files via doPDF: They appear in a decodated way. Each letter appears constantly as an other letter. That does not concern extracting a txt file from such pdf files: The txt file shows the original wording.
I appreciate doPDF very much for making non profit print files and therefore I try to contribute my experience here within this forum instead of only publishing additionally font files without any composed glyphs in regard to possible problems with doPDF ;-) .
If useful for you I would contribute more detailed information on these experiences.
Hi wwwupp, thank you for your feedback. Do you think it would be possible to send us via email (support at novapdf.com) a word document with the problematic fonts embedded so we could test the conversion here? If we have the fonts and are able to reproduce the problem here most likely we'll be able to correct it. Thank you.
Hi Softland, thank you for your reply. my zip file should have arrived in your email box. Thank you.
Thank you, we received the email and we're currently doing tests.
You must log in to post.