I have seen numerous people with the NAMELIST read error, (NAMELIST's were always flaky on my compiler!), I have a different crash, an EOF during a read on unit code 20.

I don't know if it is significant, but I always find coincidence suspicious. The wu has been running for 199:48:33 and has 399:40:43 to go, the % indicator shows excatly 33.33% complete. It does seem persistant :(

<img src="http://www.adrianxw.dk/personalsite/oddsandsods/visfortcrash.jpg">

hmmm, that's a "smallexecs" error (in our post-phase processing). the fort.20 only happens at one point, I believe. I suppose you've tried exiting and coming back in?

&gt; hmmm, that's a "smallexecs" error (in our post-phase processing). the fort.20
&gt; only happens at one point, I believe. I suppose you've tried exiting and
&gt; coming back in?
&gt;

Yup, closed and reopened, exactly the same error. Rebooted machine, same. Was completely persistant.

I detatched and re-atatched the project, and got a new download. Now, however, my stats page shows I have 2 machines which is wrong. I raised that a short while ago in another thread when I noticed it.

Nice to see Fortran still going strong. I worked with IV/66/77 from 1980 until about 1996.

I think Fortran will be haunting us forever! ;-)
I'm sorry you had to reboot, I really wanted to try and see what file it was and see if we could "bypass" it. Unfortunately I'm now away for a week at a conference and then a holiday in northern Germany. And I'd hate to crash the houses of "Team Nordlichter" when I get to Bremen and just start working instead of going to the pub with them!