Re: [pretest] regression in pdftex

Am Sun, 15 Feb 2026 16:17:53 +0000 schrieb Clason, Christian
([email protected]):

>> <test.log><test.pdf><test.tex>
> 
> The log files are pretty much identical, but the generated PDF differs in how the text is stored. Here’s the first few bytes of the content
> 
> 2025:
> /F32 9.9626 Tf 148.712 707.125 Td [(Lorem)-477(i)1(psum)-477(dolor)-476(sit)
> 
> 2026:
> /F32 9.9626 Tf 148.712 707.125 Td [(L)]TJ 6.227 0 Td [(o)]TJ 4.981 0 Td [(r)
> 
> Looks like every character is encoded individually rather than in chunks?
> 

Yes every glyph is places individually and that clearly explains the
increased size. But on windows I do not see that. Does it happens
with all fonts? E.g. also with \usepackage{lmodern} or
\usepackage{tgtermes}? 



-- 
Ulrike Fischer 
http://www.troubleshooting-tex.de/