MIXPRE: Difference between revisions
No edit summary |
No edit summary |
||
Line 13: | Line 13: | ||
::<math>P\left(G\right)=1+\frac{B^2}{G^2}</math> | ::<math>P\left(G\right)=1+\frac{B^2}{G^2}</math> | ||
:with <math>B</math>={{TAG|BMIX}}. | :with <math>B</math>={{TAG|BMIX}}. | ||
The preconditioning is done only on the total charge density (i.e. up+down component) and not on the magnetization charge density (i.e. up-down component). Up to now we have found that introduction of a metric always improves the convergence speed. The best choice is {{TAG|MIXPRE}}=1 (i.e. the default). | |||
== Related Tags and Sections == | == Related Tags and Sections == |
Revision as of 16:15, 8 February 2011
MIXPRE = 0 | 1 | 2 | 3
Default: MIXPRE = 1
Description: MIXPRE specifies the metric in the Broyden mixing scheme(IMIX=4).
- MIXPRE=0
- No preconditioning, metric=1
- MIXPRE=1
- "Inverse Kerker" metric with automatically determined BMIX (determined in such a way that the variation of the preconditioning weights covers a range of a factor 20)
- MIXPRE=2
- "Inverse Kerker" metric with automatically determined BMIX (determined in such a way that the variation of the preconditioning weights covers a range of a factor 200)
- MIXPRE=3 (implemented for test purposes; not recommended)
The preconditioning is done only on the total charge density (i.e. up+down component) and not on the magnetization charge density (i.e. up-down component). Up to now we have found that introduction of a metric always improves the convergence speed. The best choice is MIXPRE=1 (i.e. the default).
Related Tags and Sections
IMIX, INIMIX, MAXMIX, AMIX, BMIX, AMIX_MAG, BMIX_MAG, AMIN, WC