[OS X TeX] Re: The microtype mystery

[Martin Berggren]

But the really interesting problem for us outsiders is why this happens...

On Aug 20, 2010, at 17:02 , George Gratzer wrote:

> I just woke up that my problem is solved!
> 
>> Interestingly, the amssymb package---which loads the amsfonts package---appears to have all the symbols that latexsym contains so there is no need for the duplication and the speed difference is obvious. Am I wrong here?
> 
> First, Herb you are wrong. My recollection is that latexsym defines about a dozen symbols
> that amsfonts does not. It seems, however, that I do not need any one of them. So I commented out
> 
> \RequirePackage{latexsym}
> 
> in the sty file and here are the results:
> 
> without microtype:
> 
> real	0m2.604s
> user	0m2.534s
> sys	0m0.067s
> 
> with microtype:
> 
> real	0m7.172s
> user	0m7.103s
> sys	0m0.062s
> 
> PERFECT!
> 
> I am really grateful for your help Martin and Herb. Ido not know how you came up
> with the idea that latexsym could be a factor. To me it was only a definition
> of a dozen commands...
> 
> I will report soon what these times are like with the new iMac.
> 
> GG
> 
> 
> 
> 
> 
> 
> 
> On 2010-08-20, at 7:41 AM, Herbert Schulz wrote:
> 
>> 
>> On Aug 20, 2010, at 4:56 AM, Martin Costabel wrote:
>> 
>>> Josep Maria Font wrote:
>>> []
>>>> Again, a very large increase factor! The preamble of the document is:
>>>> \documentclass[a4paper]{article}
>>>> \usepackage{fixltx2e}
>>>> \usepackage{amssymb,latexsym}                         \usepackage{amsmath,amsthm}
>>>> \usepackage{bm}                 \usepackage{enumerate}
>>>> \usepackage{mathtools}
>>>> \usepackage{url}
>>>> I will e-mail the files privately to Robert. 
>>> 
>>> Here is another data point, obtained by playing with your preamble.
>>> 
>>> From my own experiment, I would suspect the interaction of microtype with the latexsym or similar packages:
>>> 
>>> I took some random article of mine (46 pages; if others want to repeat this experiment, the latex sources, two files, are publicly available here: <http://arxiv.org/format/1002.1772v1>). Then I inserted after the \documentclass line
>>> 
>>> \usepackage{microtype}
>>> 
>>> and then \usepackage{latexsym}, or alternatively just the following two lines from latexsym.sty:
>>> 
>>> \DeclareSymbolFont{lasy}{U}{lasy}{m}{n}
>>> \SetSymbolFont{lasy}{bold}{U}{lasy}{b}{n}
>>> 
>>> These two lines do absolutely nothing visible, as far as I can tell. But here are the timings from
>>> 
>>> time pdflatex CoDaNi_2010_art1
>>> 
>>> on my MbookPro core2duo:
>>> 
>>> Original file:
>>> user	0m1.835s
>>> 
>>> With just the two lines from latexsym.sty:
>>> user	0m1.909s
>>> 
>>> With \usepackage{microtype} alone:
>>> user	0m8.591s
>>> 
>>> With \usepackage{microtype} plus the two lines from latexsym.sty:
>>> user	0m49.910s
>>> 
>>> This is a factor of roughly 5 for microtype alone, and of 25 for microtype+latexsym. The extra time is spent in raw computing cpu time, no system calls or disk activity.
>>> 
>>> -- 
>>> Martin
>>> 
>> 
>> Howdy,
>> 
>> Interestingly, the amssymb package---which loads the amsfonts package---appears to have all the symbols that latexsym contains so there is no need for the duplication and the speed difference is obvious. Am I wrong here?
>> 
>> Good Luck,
>> 
>> Herb Schulz
>> (herbs at wideopenwest dot com)
>> 
>> 
>> 
>> ----------- Please Consult the Following Before Posting -----------
>> TeX FAQ: http://www.tex.ac.uk/faq
>> List Reminders and Etiquette: http://email.esm.psu.edu/mac-tex/
>> List Archive: http://tug.org/pipermail/macostex-archives/
>> TeX on Mac OS X Website: http://mactex-wiki.tug.org/
>> List Info: http://email.esm.psu.edu/mailman/listinfo/macosx-tex
>> 
> 
> ----------- Please Consult the Following Before Posting -----------
> TeX FAQ: http://www.tex.ac.uk/faq
> List Reminders and Etiquette: http://email.esm.psu.edu/mac-tex/
> List Archive: http://tug.org/pipermail/macostex-archives/
> TeX on Mac OS X Website: http://mactex-wiki.tug.org/
> List Info: http://email.esm.psu.edu/mailman/listinfo/macosx-tex
> 

Martin Berggren

-------------------------------------------------------------------
Department of Computing Science, Umeå Universitet
S-901 87 Umeå, Sweden. http://www.cs.umu.se/~martinb,
Martin.Berggren at cs.umu.se, Ph: +46-90-786 6307, +46-70-732 8111 (cell)