Ok, the reason for this failure is that your data is quite unique.
You have hit a minimisation space more complex than the algorithm can
handle! No one has encountered this type of convoluted space before,
ever in the 8 years of relax development! Not a bad effort. If you
are curious to see what this looks like, you can use the map.py sample
script together with OpenDX to visualise the space. You may need to
map out model m4 and then look at where te is zero to see this model
m3 space.
As for a fix I'll have to put some error checking code into minfx -
this is where the failure occurred. It might also be worth
investigating your data and errors as I have a feeling this is because
the R2 errors are far too small. As the Rex is assumed to be
quadratic with field strength (fast exchange), if the R2 values are
noisy and the R2 error too small this will really warp the space.
Regards,
Edward
On Fri, Oct 24, 2008 at 9:52 PM, Edward d Auvergne
<NO-REPLY.INVALID-ADDRESS@xxxxxxx> wrote:
Update of bug #12499 (project relax):
Status: None => Confirmed
Assigned to: None => bugman
Operating System: Mac OS X => All systems
_______________________________________________________
Reply to this item at:
<http://gna.org/bugs/?12499>
_______________________________________________
Message sent via/by Gna!
http://gna.org/
Re: [bug #12499] DIFF_MODEL = 'prolate' optimization failure