[Wien] Supercell calculation dilemma with WIEN2k_12

[Guo-ping Zhang]

Dear Peter and Wien2k users,

(Many thanks for your help for my previous questions!)

I was trying to do a supercell calculation (60 atoms) within 28x28x28 
(Angstrom^3), half of which is vacuum. Here are some of the questions that 
I have. (To be sure, I have searched all the previous 110 emails on this 
topic, but failed to notice any solutions to this, except a nice comment 
on the formation energy by Dr. Stefaan Cottenier.)

(1) Problems with the new WIEN2k.

"x lapw0" complains the insufficient memory like this,

forrtl: severe (41): insufficient virtual memory

This does not happen in the old wien. So my question is whether I can 
change some parameters in SRC_lapw0, so at least x lapw0 runs.

(2) Problems with the old WIEN2k.

Since the new version does not work, I tried to use the older version, 
which worked well, but there is a problem with RKM. Due to the limit set 
by NMATMAX (=10000), the RKM is now truncated to 3.24. To test the vacuum 
thickness effect, I have to increase the thickness, but this further 
reduces RKM to  an even smaller value. Since I am mostly interested in 
unoccupied states several eV above Ef, this casts
doubts on the wien results. (After I compared it with Gaussian results,
I found that APW calculation gives wrong symmetry splitting in those 
unoccupied states, but LAPW is ok).

I really appreciate it if you could give me some suggestions how to get 
the results accurately for such a big system.

If you need additional information,

Many thanks in advance!

Best regards,
Guoping