MathGroup Archive 2003

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Re: Re: v5 for Mac

  • To: mathgroup at smc.vnet.net
  • Subject: [mg43472] Re: [mg43405] Re: Re: v5 for Mac
  • From: Michael Williams <williams at vt.edu>
  • Date: Wed, 17 Sep 2003 08:00:27 -0400 (EDT)
  • Sender: owner-wri-mathgroup at wolfram.com

I've come to a standstill with v 5 on OS X (10.2.6).  After months of
back-and-forth with Wolfram support, they have declared it a personal
problem.  (Can't duplicate but won't tell me what machines they tested
it on.)

Apparently the problem is machine dependent-- it happens with 100%
reliability on our fast dual processor machines, but not on our
laptops. To see the problem (or not), put a series of expressions of
the form (multiple expressions accelerates the crash) into a cell and
evaluate:

ExportString["a","GIF"];
ExportString["a","GIF"];
. . . .

Also crashes Mathematica with other image formats.  I would appreciate
hearing how others fare with this (esp. w.r.t. machine).  Wolfram support
accuses me (literally!!) of having an impure system.  Impure thoughts
maybe, but I replicated this on a machine right out of its carton, with
-only- Mathematica installed.  (Hmm, maybe they meant Mathematica....)

I'm sorry for sending what amounts to a bug report to the list, but
I went through channels with no joy.  Unfortunately, this feature is
important to a series of my applications, hence my frustration.

A few other notes: -  Even if this fails to crash Mathematica (on a PB
for example), it's an order of magnitude (!!) slower than 4.2

- Under Panther, any export call of this type is DOA (dead stall), whatever the machine.

- Nothing like this happens under v 4.2 (even under Panther), but
it is apparent from my observations that Mathematica's converter,
gif.exe/pnm.exe, has a memory leak.  The converters (4.2) are unusably
slow under Panther.

Michael Williams Mathematics Virginia Tech


On  9/16/03,  selwynh at earthlink.net (Selwyn Hollis)  said:

>The first of these --- the (not) rotated text bug --- has not been  
>fixed either.
>
>Another bizarre bug has popped up in 5.0: In text cells, lines are  
>sometimes broken after apostrophes. So a word like  
>can\[CloseCurlyQuote]t will break between \[CloseCurlyQuote] and t.  
>Astonishing. I wonder if that happens in Windows too.
>
>-----
>Selwyn Hollis
>http://www.math.armstrong.edu/faculty/hollis
....


  • Prev by Date: RE: Re: Re: v5 for Mac
  • Next by Date: Re: Finding the Path to/Directory for the current EvaluationNotebook?
  • Previous by thread: RE: Re: Re: v5 for Mac
  • Next by thread: Re: Re: Re: v5 for Mac