[Wien] Persisting LOPW and LAPW1 errors

2014-02-05 Thread Muhammad Sajjad
Dear Peter Bala and other wien2k users I am running SCF calculations for Se (structure is attached) with WCGGA and LSDA using Wien2k 12 version. With WCGGA, an error LOPW (plan waves exhausted) appear and I try to solve it as suggested in M. list (like increasing Rkmax,switching to broading

[Wien] Persisting LOPW and LAPW1 errors

2014-02-05 Thread Martin Kroeker
Check your z coordinate. At the very least, you will need more decimal places to accurately describe the special position if it is an irrational fraction (1./6.), but I suspect you also mixed up the crystallographic data between two published structures in different spacegroups. If you want P31

Re: [Wien] Persisting LOPW and LAPW1 errors

2014-02-05 Thread Laurence Marks
As Martin Kroeker said, you have a positional error and x patchsymm thinks that your z-position should be 0.8333 (i.e. 5/6). I strongly suggest that most beginners check their structures using this utility as it frequently finds small errors. This may or may not solve the problem. If it does

Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn?

2014-02-05 Thread Lyudmila Dobysheva
On 05.02.2014 14:29, Muhammad Sajjad wrote: I am running SCF calculations for Se (structure is attached) with WCGGA and LSDA using Wien2k 12 version. With WCGGA, an error LOPW (plan waves exhausted) appear... Then I used LSDA and tis time the error arises in LAPW 1 and its It is rather strange

Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn?

2014-02-05 Thread Laurence Marks
You are right, there seems to be a nn bug. This may explain everything including the LOPW exhausted error. On Wed, Feb 5, 2014 at 8:48 AM, Lyudmila Dobysheva lyuk...@mail.ru wrote: On 05.02.2014 14:29, Muhammad Sajjad wrote: I am running SCF calculations for Se (structure is attached) with

Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn?

2014-02-05 Thread Fecher, Gerhard
for WIEN2k users Betreff: Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn? You are right, there seems to be a nn bug. This may explain everything including the LOPW exhausted error. On Wed, Feb 5, 2014 at 8:48 AM, Lyudmila Dobysheva lyuk...@mail.ru wrote: On 05.02.2014 14:29, Muhammad Sajjad

Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn?

2014-02-05 Thread Laurence Marks
] Persisting LOPW and LAPW1 errors - bug of nn? You are right, there seems to be a nn bug. This may explain everything including the LOPW exhausted error. On Wed, Feb 5, 2014 at 8:48 AM, Lyudmila Dobysheva lyuk...@mail.ru wrote: On 05.02.2014 14:29, Muhammad Sajjad wrote: I am running SCF calculations

Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn?

2014-02-05 Thread Laurence Marks
[l-ma...@northwestern.edu] Gesendet: Mittwoch, 5. Februar 2014 17:00 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn? You are right, there seems to be a nn bug. This may explain everything including the LOPW exhausted error. On Wed, Feb 5

Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn?

2014-02-05 Thread Laurence Marks
] Gesendet: Mittwoch, 5. Februar 2014 17:00 An: A Mailing list for WIEN2k users Betreff: Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn? You are right, there seems to be a nn bug. This may explain everything including the LOPW exhausted error. On Wed, Feb 5, 2014 at 8:48 AM, Lyudmila

Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn?

2014-02-05 Thread Fecher, Gerhard
. Februar 2014 19:22 Cc: A Mailing list for WIEN2k users Betreff: Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn? Line 303 of nn.f can be changed to IF((DIST.LT..001).and.(K.eq.index)) GO TO 110 Note: this will now trap many cases with inappropriate precision with 1/3, 5/6 or similar

Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn?

2014-02-05 Thread Muhammad Sajjad
...@northwestern.edu] Gesendet: Mittwoch, 5. Februar 2014 19:22 Cc: A Mailing list for WIEN2k users Betreff: Re: [Wien] Persisting LOPW and LAPW1 errors - bug of nn? Line 303 of nn.f can be changed to IF((DIST.LT..001).and.(K.eq.index)) GO TO 110 Note: this will now trap many cases