[gmx-users] Re: Power4 : Trying to install...

[Marc Baaden]

Hi again,

I can sort of confirm Pim's report. I must admit that the configure/make
went much more smoothly in my case, but the final result is similar. I can
indeed also run the polymer benchmark, but no water simulations.
Depending on how I compile the code (eg qstrict option, -g, optimization
etc) I obtain slightly different error messages, which don't really help
in finding the error.

spoel at xray.bmc.uu.se said:
>> the tpr can be compared using gmxcheck -s1 -s2 or dumped using
>> gmxdump, allowing you to test tpr reading on IBM. 

I checked it. The tpr is ok ! (eg gmxdump on power4 is the same as under
linux) But as I said, the linux one runs fine and the power4 one bombs
out with the Warning: Only triclinic boxes with[..] error !

Can it be that gmxdump works ok, but when the actual data is loaded into
memory there is something fishy ? (I am thinking along 32 vs 64 bits or
so) ..


Maybe someone who has a working gmx3.2.1 mdrun for Power4 wants to send
me a static executable :)) ..

  Marc

-- 
 Dr. Marc Baaden  - Institut de Biologie Physico-Chimique, Paris
 mailto:baaden at smplinux.de      -      http://www.marc-baaden.de
 FAX: +49 697912 39550  -  Tel: +33 15841 5176 ou +33 609 843217