Skip to Content.
Sympa Menu

forum - Re: [abinit-forum] abinit result stability.

forum@abinit.org

Subject: The ABINIT Users Mailing List ( CLOSED )

List archive

Re: [abinit-forum] abinit result stability.


Chronological Thread 
  • From: Alain Jacques <ajs.jacques@gmail.com>
  • To: forum@abinit.org
  • Subject: Re: [abinit-forum] abinit result stability.
  • Date: Fri, 15 Aug 2008 15:51:54 +0200
  • Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:user-agent:mime-version:to:subject:references :in-reply-to:content-type:content-transfer-encoding:from; b=QPqvn54Cv3h1anBniDDoNS51e9pwgdVJugzbxILvKoyszW4g76uNsBFlWnIskcgbu6 1jcux7KmCUiu9PDE+yKReoQw/BQK1LQsMyyQrvwTN3fq+gm9NVEXQiuSCJPYpWWDggf/ Dxw3WqDgTNepgdR/+PWK2fIGd+XhiXuKPQkdk=

Thanks Samuel for your efforts ; I would be very interested to have a feedback about failed tests. I don't know if this is well documented but, at the end of the test run, you will find a "report" and a "summaryfldiff" file in each test subdirectory.


Samuel Sarholz wrote:
Hi Alain,

the minimal tests ran fine, except with the Intel Compiler with -O3. I get core dumps as well with some tests.

I will run the accuracy tests and see whether they report any problems.

best regards,
Samuel

Alain Jacques wrote:
Hello Samuel,

Some aggressive optimization options may produce unstable code and therefore abinit comes with an extensive set of test programs. Can you give some information about how your different builds behave with the "minimal" or "accuracy" test suite?

Kind regards,

Alain


Samuel Sarholz wrote:
Hi,

we are having trouble with abinit (abinis) results. They seem very unstable and differ a lot depending on the compiler used.

I used the 5.5.4 abinit version and 3 different compilers.
Intel 10.1 (10.1.015)
GNU 4.3.1
Sun Compiler (latest express release Sun Ceres Fortran 95 8.3 Linux_i386 2008/07/10)







Archive powered by MHonArc 2.6.16.

Top of Page