MSDGW F - Revision history

Csheldon at 09:00, 29 January 2026

2026-01-29T09:00:45Z

← Older revision		Revision as of 09:00, 29 January 2026
Line 26:		Line 26:
	<references/>		<references/>
	----		----
	~~<!--~~ [[Category:INCAR]][[Category:GW]][[Category:Constrained-random-phase_approximation]] ~~-->~~		[[Category:INCAR]][[Category:GW]][[Category:Constrained-random-phase_approximation]]

Csheldon at 08:59, 29 January 2026

2026-01-29T08:59:59Z

← Older revision		Revision as of 08:59, 29 January 2026
Line 3:		Line 3:
	Description: A positive value of {{TAG\|MSDGW_F}} triggers the mixed stochastic-deterministic compression algorithm of Altman and co-workers.{{cite\|altman:prl:2024}}		Description: A positive value of {{TAG\|MSDGW_F}} triggers the mixed stochastic-deterministic compression algorithm of Altman and co-workers.{{cite\|altman:prl:2024}}
	----		----
	{{TAG\|MSDGW_F}} is the constant energy ratio <math>F</math> of the compression algorithm. If set to a positive value, energies beyond the protected space defined by {{TAG\|MSDGW_NP}} are subdivided into energy bins of width <math>\Delta E_i</math> and replaced by other energies <math>E_i</math>, such that <math>F=\Delta E_i/E_i</math>. The original orbitals are replaced by {{TAG\|MSDGW_NXI}} randomly linear combined orbitals. Larger values of {{TAG\|MSDGW_F}} increase the compression level at expense of accuracy. The same holds true for smaller values of {{TAG\|MSDGW_NXI}}.		{{TAG\|MSDGW_F}} is the constant energy ratio <math>F</math> of the compression algorithm. If set to a positive value, energies beyond the protected space defined by {{TAG\|MSDGW_NP}} are subdivided into energy bins of width <math>\Delta E_i</math> and replaced by other energies <math>E_i</math>, such that <math>F=\Delta E_i/E_i</math>. The original orbitals are replaced by {{TAG\|MSDGW_NXI}} randomly linear combined orbitals. Larger values of {{TAG\|MSDGW_F}} increase the compression level at the expense of accuracy. The same holds true for smaller values of {{TAG\|MSDGW_NXI}}.

	This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated ~~thay~~ one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap.{{cite\|altman:prl:2024}}		This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated that one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with a resulting error of only 50 meV or less on the quasi-particle band gap.{{cite\|altman:prl:2024}}
	{{NB\| mind \| ~~Avaliable~~ as of VASP.6.5.2.\|:}}		{{NB\| mind \| Available as of VASP.6.6.0.\|:}}
	{{NB\| warning \| Not recommended for {{TAG\|LRPAFORCE}}{{=}}T.\|:}}		{{NB\| warning \| Not recommended for {{TAG\|LRPAFORCE}}{{=}}T.\|:}}
	==Use cases==		==Use cases==
Line 14:		Line 14:
	bands after compression: 240		bands after compression: 240

	To test different compression settings, it is recommended perform the GW/CRPA calculation in steps and to set {{TAG\|MSDGW_F}} only in the actual GW step. This avoids repeating the expensive exact diagonalization of the Kohn-Sham ~~hamiltonian~~.		To test different compression settings, it is recommended perform the GW/CRPA calculation in steps and to set {{TAG\|MSDGW_F}} only in the actual GW step. This avoids repeating the expensive exact diagonalization of the Kohn-Sham Hamiltonian.

	==Caveats==		==Caveats==

Kaltakm at 14:44, 25 March 2025

2025-03-25T14:44:17Z

← Older revision		Revision as of 14:44, 25 March 2025
Line 7:		Line 7:
	This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap.{{cite\|altman:prl:2024}}		This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap.{{cite\|altman:prl:2024}}
	{{NB\| mind \| Avaliable as of VASP.6.5.2.\|:}}		{{NB\| mind \| Avaliable as of VASP.6.5.2.\|:}}
			{{NB\| warning \| Not recommended for {{TAG\|LRPAFORCE}}{{=}}T.\|:}}
	==Use cases==		==Use cases==
	*Recommended for {{TAG\|ALGO}}=EVG0W[R\|RK]\|CRPA[R\|RK]:		*Recommended for {{TAG\|ALGO}}=EVG0W[R\|RK]\|CRPA[R\|RK]:

Kaltakm: /* Caveats */

2025-03-25T14:41:12Z

Caveats

← Older revision		Revision as of 14:41, 25 March 2025
Line 16:		Line 16:

	==Caveats==		==Caveats==
	Care must be taken for GW/CRPA calculations that include the long-wave limit stored in {{FILE\|WAVEDER}}. After band compression, this limit must be re-calculated by setting {{TAG\|LOPTICS}} in combination with {{TAG\|LPEAD}}.		Care must be taken for GW/CRPA calculations that are performed in steps and include the long-wave limit stored in {{FILE\|WAVEDER}}. After band compression, this limit must be re-calculated by setting {{TAG\|LOPTICS}} in combination with {{TAG\|LPEAD}}.

	== Related tags and articles==		== Related tags and articles==

Kaltakm: Kaltakm moved page Construction:MSDGW F to MSDGW F: Publication

2025-03-25T14:11:00Z

Kaltakm moved page Construction:MSDGW F to MSDGW F: Publication

← Older revision	Revision as of 14:11, 25 March 2025
(No difference)

Kaltakm at 14:08, 25 March 2025

2025-03-25T14:08:56Z

← Older revision		Revision as of 14:08, 25 March 2025
Line 5:		Line 5:
	{{TAG\|MSDGW_F}} is the constant energy ratio <math>F</math> of the compression algorithm. If set to a positive value, energies beyond the protected space defined by {{TAG\|MSDGW_NP}} are subdivided into energy bins of width <math>\Delta E_i</math> and replaced by other energies <math>E_i</math>, such that <math>F=\Delta E_i/E_i</math>. The original orbitals are replaced by {{TAG\|MSDGW_NXI}} randomly linear combined orbitals. Larger values of {{TAG\|MSDGW_F}} increase the compression level at expense of accuracy. The same holds true for smaller values of {{TAG\|MSDGW_NXI}}.		{{TAG\|MSDGW_F}} is the constant energy ratio <math>F</math> of the compression algorithm. If set to a positive value, energies beyond the protected space defined by {{TAG\|MSDGW_NP}} are subdivided into energy bins of width <math>\Delta E_i</math> and replaced by other energies <math>E_i</math>, such that <math>F=\Delta E_i/E_i</math>. The original orbitals are replaced by {{TAG\|MSDGW_NXI}} randomly linear combined orbitals. Larger values of {{TAG\|MSDGW_F}} increase the compression level at expense of accuracy. The same holds true for smaller values of {{TAG\|MSDGW_NXI}}.

	This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated~~{{cite\|altman:prl:2024}}~~ thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap.		This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap.{{cite\|altman:prl:2024}}
	{{NB\| mind \| Avaliable as of VASP.6.5.2.\|:}}		{{NB\| mind \| Avaliable as of VASP.6.5.2.\|:}}
	==Use cases==		==Use cases==

Kaltakm at 14:08, 25 March 2025

2025-03-25T14:08:39Z

← Older revision		Revision as of 14:08, 25 March 2025
Line 5:		Line 5:
	{{TAG\|MSDGW_F}} is the constant energy ratio <math>F</math> of the compression algorithm. If set to a positive value, energies beyond the protected space defined by {{TAG\|MSDGW_NP}} are subdivided into energy bins of width <math>\Delta E_i</math> and replaced by other energies <math>E_i</math>, such that <math>F=\Delta E_i/E_i</math>. The original orbitals are replaced by {{TAG\|MSDGW_NXI}} randomly linear combined orbitals. Larger values of {{TAG\|MSDGW_F}} increase the compression level at expense of accuracy. The same holds true for smaller values of {{TAG\|MSDGW_NXI}}.		{{TAG\|MSDGW_F}} is the constant energy ratio <math>F</math> of the compression algorithm. If set to a positive value, energies beyond the protected space defined by {{TAG\|MSDGW_NP}} are subdivided into energy bins of width <math>\Delta E_i</math> and replaced by other energies <math>E_i</math>, such that <math>F=\Delta E_i/E_i</math>. The original orbitals are replaced by {{TAG\|MSDGW_NXI}} randomly linear combined orbitals. Larger values of {{TAG\|MSDGW_F}} increase the compression level at expense of accuracy. The same holds true for smaller values of {{TAG\|MSDGW_NXI}}.

	This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated{{cite\|altman:prl:2024}} thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap~~. Comparable results can be achieved for Constrained Random Phase Approximation (cRPA) calculations~~.		This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated{{cite\|altman:prl:2024}} thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap.
	{{NB\| mind \| Avaliable as of VASP.6.5.2.\|:}}		{{NB\| mind \| Avaliable as of VASP.6.5.2.\|:}}
	==Use cases==		==Use cases==

Kaltakm at 14:07, 25 March 2025

2025-03-25T14:07:20Z

← Older revision		Revision as of 14:07, 25 March 2025
Line 6:		Line 6:

	This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated{{cite\|altman:prl:2024}} thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap. Comparable results can be achieved for Constrained Random Phase Approximation (cRPA) calculations.		This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated{{cite\|altman:prl:2024}} thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap. Comparable results can be achieved for Constrained Random Phase Approximation (cRPA) calculations.
	{{NB\| mind \| Avaliable as of VASP ~~version >~~ 6.5.1.\|:}}		{{NB\| mind \| Avaliable as of VASP.6.5.2.\|:}}
	==Use cases==		==Use cases==
	*Recommended for {{TAG\|ALGO}}=EVG0W[R\|RK]\|CRPA[R\|RK]:		*Recommended for {{TAG\|ALGO}}=EVG0W[R\|RK]\|CRPA[R\|RK]:

Kaltakm at 14:05, 25 March 2025

2025-03-25T14:05:57Z

← Older revision		Revision as of 14:05, 25 March 2025
Line 5:		Line 5:
	{{TAG\|MSDGW_F}} is the constant energy ratio <math>F</math> of the compression algorithm. If set to a positive value, energies beyond the protected space defined by {{TAG\|MSDGW_NP}} are subdivided into energy bins of width <math>\Delta E_i</math> and replaced by other energies <math>E_i</math>, such that <math>F=\Delta E_i/E_i</math>. The original orbitals are replaced by {{TAG\|MSDGW_NXI}} randomly linear combined orbitals. Larger values of {{TAG\|MSDGW_F}} increase the compression level at expense of accuracy. The same holds true for smaller values of {{TAG\|MSDGW_NXI}}.		{{TAG\|MSDGW_F}} is the constant energy ratio <math>F</math> of the compression algorithm. If set to a positive value, energies beyond the protected space defined by {{TAG\|MSDGW_NP}} are subdivided into energy bins of width <math>\Delta E_i</math> and replaced by other energies <math>E_i</math>, such that <math>F=\Delta E_i/E_i</math>. The original orbitals are replaced by {{TAG\|MSDGW_NXI}} randomly linear combined orbitals. Larger values of {{TAG\|MSDGW_F}} increase the compression level at expense of accuracy. The same holds true for smaller values of {{TAG\|MSDGW_NXI}}.

	This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated{{cite\|altman:prl:2024}} thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap. Comparable results can be achieved for Constrained Random Phase Approximation (~~CRPA~~) calculations.		This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated{{cite\|altman:prl:2024}} thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap. Comparable results can be achieved for Constrained Random Phase Approximation (cRPA) calculations.
	{{NB\| mind \| Avaliable as of VASP version > 6.5.1.\|:}}		{{NB\| mind \| Avaliable as of VASP version > 6.5.1.\|:}}
	==Use cases==		==Use cases==

Kaltakm at 14:04, 25 March 2025

2025-03-25T14:04:37Z

← Older revision		Revision as of 14:04, 25 March 2025
Line 6:		Line 6:

	This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated{{cite\|altman:prl:2024}} thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap. Comparable results can be achieved for Constrained Random Phase Approximation (CRPA) calculations.		This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated{{cite\|altman:prl:2024}} thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap. Comparable results can be achieved for Constrained Random Phase Approximation (CRPA) calculations.
	{{NB\| mind \| Avaliable as of VASP version > 6.5.0.\|:}}		{{NB\| mind \| Avaliable as of VASP version > 6.5.1.\|:}}
	==Use cases==		==Use cases==
	*Recommended for {{TAG\|ALGO}}=EVG0W[R\|RK]\|CRPA[R\|RK]:		*Recommended for {{TAG\|ALGO}}=EVG0W[R\|RK]\|CRPA[R\|RK]:

← Older revision		Revision as of 14:44, 25 March 2025
Line 7:		Line 7:
	This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap.{{cite\|altman:prl:2024}}		This compression algorithm has been developed to reduce the large number of virtual states required for the calculation of the screened interaction in GW calculations. It has been demonstrated thay one can reduce the virtual manifold by more than 50 per cent and speed up the GW step by a factor of 2 or more with an resulting error of only 50 meV or less on the quasi particle band gap.{{cite\|altman:prl:2024}}
	{{NB\| mind \| Avaliable as of VASP.6.5.2.\|:}}		{{NB\| mind \| Avaliable as of VASP.6.5.2.\|:}}
			{{NB\| warning \| Not recommended for {{TAG\|LRPAFORCE}}{{=}}T.\|:}}
	==Use cases==		==Use cases==
	*Recommended for {{TAG\|ALGO}}=EVG0W[R\|RK]\|CRPA[R\|RK]:		*Recommended for {{TAG\|ALGO}}=EVG0W[R\|RK]\|CRPA[R\|RK]: