[tex-live] running programs from texlive tlc2

George N. White III aa056 at chebucto.ns.ca
Thu Dec 9 14:31:17 CET 2004

On Thu, 9 Dec 2004, Tigran Aivazian wrote:

> On Thu, 9 Dec 2004, David Kastrup wrote:
>> Sebastian Rahtz <sebastian.rahtz at computing-services.oxford.ac.uk> writes:
>>> I'll pass on the chance to read your code, thanks :-} You need a
>>> macro expert, not a dilettante like me.
>> Well, if the incriminated code is from Yannis, it would seem that he
>> already had an expert have a go on it.  That does not mean that the
>> stuff is necessarily done in the best possible way, but it could well
>> be that there is no reasonably obvious way to achieve the same thing
>> more efficiently.
> Yes, it is true that the Hebrew typesetting is extremely "slow" even if I 
> disable completely all my macros for specific layout, i.e. just plain typeset 
> pieces of text (with vowels and accents of course).
> So, due to the complexity of the underlying model (anyone who have read 
> Yannis' paper on Hebrew Bible typesetting knows this) it is probable that 
> there is no easy way to do it faster...

Still, regardless of the code, enormous differences in processing speed 
between linux and Win32 are surprising and suggest some serious hardware 
or configuration problem.

Are these differences apparent for other documents or only Hebrew?

Benchmarks such as you see in magazines aren't much help choosing
systems, but can be useful diagnostic tools when a particular machine
is shown to perform well below expected levels.  I've used them that
way a few times to prove to vendors that their hardware had problems --
usually resulting in swapping components until the system came up to

George N. White III  <aa056 at chebucto.ns.ca>
   Head of St. Margarets Bay, Nova Scotia, Canada

More information about the tex-live mailing list