# Problem with 'feyn' package for 11pt font size

I need to draw several Feynman diagrams and I found the feyn package suitable enough for this purpose. I use MiKTeX to compile the tex-file. The problem is that for some strange reason it does not accept 11 pt font-size? I say strange, because the 10pt and 12pt versions are perfectly executable, and the difference between them and the 11pt-module is just in the values of some arguments (arrow width, loop radius, etc.), which shouldn't cause any compilation problems...

The exact mistake description, which I get, is:

! I can't find file cmr11'.

l.33 input cmr11

;^^I^^I^^I% ...but miss out the generate roman'

Please type another input file name

! Emergency stop.

l.33 input cmr11

;^^I^^I^^I% ...but miss out the generate roman'

Transcript written on feyn11.log.

miktex-maketfm.exe: Windows API error 2: The system cannot find the file specified.
miktex-maketfm.exe: Data: feyn11.tfm
! Font OMS/feyn/m/n/10.95=feyn11 at 10.95pt not loadable: Metric (TFM) file not
found.
relax

l.99 \setbox0=\hbox{$\FeynSpaceChar$}
?


Here is an example with the (working) 12pt font-size:

\documentclass [12pt]{article}
\usepackage{feyn}
\begin{document}
$\Diagram{ fdA & & fuA \\ & g & \\ fuA && fdA }$
\end{document}


• Welcome to TeX.SX! Please help us to help you and add a minimal working example (MWE) that illustrates your problem. It will be much easier for us to reproduce your situation and find out what the issue is when we see compilable code, starting with \documentclass{...} and ending with \end{document}. – Martin Schröder Nov 18 '14 at 16:37
• Please put that document in the question, not in comments – David Carlisle Nov 18 '14 at 16:42
• The example in the above (now deleted?) comment works without error in texlive 2014 – David Carlisle Nov 18 '14 at 16:45

Here is feyn11.mf:

% Feynman diagram font, 11pt display size, release 0.3.3, 2009 October 8.
%
% Copyright 1991, 1994, 2001, Norman Gray <norman@astro.gla.ac.uk>
% Mercurial revision 204fc9a2e560, 2009-10-08 12:56 +0100

font_identifier "FEYN"; font_size 11pt#;

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%
% First the base size

module# := 11pt#;

% phfudge is ... what?!
phfudge# := 0pt#;       % must be, for \Diagram macro

linewidth# := 0.85pt#;      % pen width
thinlinewidth# := 0.55pt#;  % thin pen
bigarrow# := 6.5pt#;        % size of arrows
littlearrow# := 4.5pt#;
blobr# := 7pt#;         % radius of blobs and crosses

on_math_axis := 1;

% We want the roman digits from the corresponding cmr font, and that's
% all.  Therefore, we don't want to obey the generate' command in
% cmr??.mf.  To do this, we read in cmbase, and then set generate to
% gobble its argument.
def gobble primary g=enddef;    % gobbles its argument
input cmbase;           % Read it explicitly, rather than
% implicitly within cmr10
let generate=gobble;
input cmr11;            % ...but miss out the generate roman'

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%

input feyn;

end


There is no cmr11.mf file in the standard distribution of the Computer Modern fonts, so running Metafont on this file will result in the error

! I can't find file cmr11'.

l.33 input cmr11


unless the sauter package is installed too, which is able to supply interpolated versions of the files in the Computer Modern font distribution.

For MiKTeX the package is listed at http://miktex.org/packages/sauter so it can be installed with the wizard. For TeX Live it is

tlmgr install sauter

• This was the problem, indeed. I should point out, though, that the installation of sauter didn't help. Nevertheless, I simply edited the feyn11.mf and feyntext11.mf files and made them load cmr10 (instead of cmr11), which resolved the problem. Thank you for the hint. – NewbieTexnician Nov 18 '14 at 17:36
• @NewbieTexnician I don't know what happens on MiKTeX, sorry. – egreg Nov 18 '14 at 17:37
• @NewbieTexnician: That looks like an error that should be reported to the author (Norman Gray). – Martin Schröder Nov 18 '14 at 23:40
• @MartinSchröder I believe it's deliberate, but of course it assumes cmr11.mf exists or can be generated. Actually, the instruction input cmr11 after let generate=gobble just collects the parameters, which could be obtained in a different way, in this case. – egreg Nov 19 '14 at 0:25