Re: [Bug-gnubg] Crashes of a notebook while analysing mat-files

[Joern Thyssen]

On Mon, Aug 25, 2003 at 08:59:46AM -0400, address@hidden wrote
> 
> Unfortunately I have a very specific problem with my notebook 
> (AMD Athlon XP 1800+) when analysing mat-files: it very often
> crashes!
> These crashes occur only on my notebook (NOT on 4 desktops
> where I have installed GNUBG).
> 
> And furthermore they occur only during match-analysis (NEVER
> while playing with the program).

Do you see any message after the crash?

It does look like a hardware failure during CPU intensive operations.  A
German former collegue of mine at the University of Auckland also had a
laptop that crashed under CPU intensive operations (e.g., a linux kernel
compile).  In his case the fan wasn't working properly, and the laptop
simply crashed when it got too hot.  I can't remember the brand of his
laptop -- but it was a rather cheap one; not a "known" brand.

Does this happen during rollouts too? 

> I have to say that I noticed these crashes also when ANALYSING
> chess games with my notebook using the world champion program
> Shredder 7 (which runs absolutely stable on my desktops, and never
> failed while PLAYING on my notebook) !

Again, this indicates hardware problem during CPU intensive operations.

> So I believe it might be a problem with my special notebook or
> did you already hear any crash-reports from any other laptop-users ?

I only have a laptop and have never experienced crashes that wasn't due
to bugs in gnubg. My laptop is a high-quality IBM laptop (disclaimer: I
work for IBM, so I may be a bit biased in my view :-)).

> I already wrote to the support-team of the producer of my notebook.
> If you can read german, I add the text as an attachment (targa1r.htm).
> Here a short report what they said: "We have installed GNUBG on 
> one of our notebooks (which runs fine). With GNUBG it was sufficient
> to click two times on the second dice and the program crashed.
> We strongly believe that the program is not properly programmed
> respectively it might be a compatibility problem of the software with
> the chip-set. We recommend that you contact the programmers of
> the software."

Err, I'm not with you here... click two times on the second dice? Can you
describe in details exactly what the did?

Jørn

pgpo8AWRlMTEi.pgp
Description: PGP signature