MAXMIX - Revision history

Huebsch at 14:54, 17 October 2024

2024-10-17T14:54:28Z

← Older revision		Revision as of 14:54, 17 October 2024
Line 27:		Line 27:
	----		----

	[[Category:INCAR tag~~]][[Category:Electronic minimization~~]][[Category:Density mixing]]		[[Category:INCAR tag]][[Category:Density mixing]]

Ftran at 09:43, 18 October 2023

2023-10-18T09:43:44Z

← Older revision		Revision as of 09:43, 18 October 2023
Line 11:		Line 11:


	~~'''Mind''':~~ {{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.		{{NB\|mind\|{{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.}}

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

Karsai: /* Related tags and articles */

2022-04-08T14:45:30Z

Huebsch at 14:25, 6 April 2022

2022-04-06T14:25:20Z

← Older revision		Revision as of 14:25, 6 April 2022
Line 1:		Line 1:
	{{TAGDEF\|MAXMIX\|[integer]\|-45}}		{{TAGDEF\|MAXMIX\|[integer]\|-45}}

	Description: {{TAG\|MAXMIX}} specifies the maximum number steps stored in Broyden mixer ({{TAG\|IMIX}}=4).		Description: {{TAG\|MAXMIX}} specifies the maximum number of steps stored in the Broyden mixer ({{TAG\|IMIX}}=4).
	----		----
	{{TAG\|MAXMIX}} specifies the maximum number of vectors stored in the Broyden/Pulay mixer, in other words it corresponds to the maximal rank of the approximation of the charge dielectric function build up by the mixer. {{TAG\|MAXMIX}} can be either negative or positive:		{{TAG\|MAXMIX}} specifies the maximum number of vectors stored in the Broyden/Pulay mixer, in other words, it corresponds to the maximal rank of the approximation of the charge-dielectric function build up by the mixer. {{TAG\|MAXMIX}} can be either negative or positive:
	*{{TAG\|MAXMIX}}<0		*{{TAG\|MAXMIX}}<0
	:The mixer is reset after each ionic step or if the number of electronic steps exceeds \|{{TAG\|MAXMIX}}\| (this is the default and similar to the behavior of VASP.4.3 and VASP.3.2).		:The mixer is reset after each ionic step or if the number of electronic steps exceeds \|{{TAG\|MAXMIX}}\| (this is the default and similar to the behavior of VASP.4.3 and VASP.3.2).
	*{{TAG\|MAXMIX}}>0		*{{TAG\|MAXMIX}}>0
	:The charge density mixer is only reset if the storage capabilities are exceeded. The reset is done "smoothly" by removing the five oldest vectors from the iteration history. Therefore, if {{TAG\|MAXMIX}} is positive, the approximation for the charge dielectric function which was obtained in previous ionic steps is "reused" in the current ionic step, and this in turn can reduce the number of electronic steps during relaxations and MD's. Especially for relaxations ~~which~~ start from a good ionic starting guess and for systems with a strong charge sloshing behavior the speedup can be significant. We found that for a 12 Å long box containing 16 Fe atoms the number of electronic iterations decreased from 8 to 2-3 when {{TAG\|MAXMIX}} was set to 40. For a carbon surface the number of iterations decreased from 7 to 3. At the same time the energy stability increased significantly. But be careful, this option increases the memory requirements for the mixer considerably, and thus the option is not recommended for systems ~~were~~ charge sloshing is negligible anyway (like bulk simple metals). The optimal setting for {{TAG\|MAXMIX}} is usually around three times the number of electronic steps required in the first iteration. Too large values for {{TAG\|MAXMIX}} might cause the code to crash (because linear dependencies between input vectors might develop). Too small values for {{TAG\|MAXMIX}} can slow your convergence significantly. For instance, if you need 50 self-~~consisitency~~ cycles, and set {{TAG\|MAXMIX}} to 20, you force the mixer to remove iteration history ~~continously~~, which can cause divergence and at least a slow down of the convergence.		:The charge density mixer is only reset if the storage capabilities are exceeded. The reset is done "smoothly" by removing the five oldest vectors from the iteration history. Therefore, if {{TAG\|MAXMIX}} is positive, the approximation for the charge dielectric function which was obtained in previous ionic steps is "reused" in the current ionic step, and this, in turn, can reduce the number of electronic steps during relaxations and MD's. Especially for relaxations that start from a good ionic starting guess and for systems with a strong charge sloshing behavior the speedup can be significant. We found that for a 12 Å long box containing 16 Fe atoms the number of electronic iterations decreased from 8 to 2-3 when {{TAG\|MAXMIX}} was set to 40. For a carbon surface the number of iterations decreased from 7 to 3. At the same time, the energy stability increased significantly. But be careful, this option increases the memory requirements for the mixer considerably, and thus the option is not recommended for systems where charge sloshing is negligible anyway (like bulk simple metals). The optimal setting for {{TAG\|MAXMIX}} is usually around three times the number of electronic steps required in the first iteration. Too large values for {{TAG\|MAXMIX}} might cause the code to crash (because linear dependencies between input vectors might develop). Too small values for {{TAG\|MAXMIX}} can slow your convergence significantly. For instance, if you need 50 self-consistency cycles, and set {{TAG\|MAXMIX}} to 20, you force the mixer to remove iteration history continuously, which can cause divergence and at least a slow down of the convergence.
	* Caution: do not set {{TAG\|MAXMIX}}>0 in the following cases. (i) If your initial positions in the {{TAG\|POSCAR}} file are far from the fully relaxed positions, the ions might move considerably during relaxation. In this case, it is not expedient to "reuse" charge mixing information from the previous ionic steps. (ii) During machine learning, the first principles calculations are often bypassed for hundreds or even thousands of ionic steps, and the ions might move considerably between first principles calculations. In ~~this~~ cases using {{TAG\|MAXMIX}} will very often lead to electronic divergence or strange errors during the self-consistency cycle. In general, whenever the column RMS(c) in the {{TAG\|OSZICAR}} files shows a sudden increase in the norm of the charge density residual vector, try to remove the tag {{TAG\|MAXMIX}} from the INCAR file.		* Caution: do not set {{TAG\|MAXMIX}}>0 in the following cases. (i) If your initial positions in the {{TAG\|POSCAR}} file are far from the fully relaxed positions, the ions might move considerably during relaxation. In this case, it is not expedient to "reuse" charge mixing information from the previous ionic steps. (ii) During machine learning, the first-principles calculations are often bypassed for hundreds or even thousands of ionic steps, and the ions might move considerably between first-principles calculations. In these cases using {{TAG\|MAXMIX}} will very often lead to electronic divergence or strange errors during the self-consistency cycle. In general, whenever the column RMS(c) in the {{TAG\|OSZICAR}} files shows a sudden increase in the norm of the charge density residual vector, try to remove the tag {{TAG\|MAXMIX}} from the INCAR file.


	'''Mind''': {{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.		'''Mind''': {{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.

	== Related ~~Tags~~ and ~~Sections~~ ==		== Related tags and articles ==
	{{TAG\|IMIX}},		{{TAG\|IMIX}},
	{{TAG\|INIMIX}},		{{TAG\|INIMIX}},
Line 27:		Line 27:
	----		----

	[[Category:INCAR]][[Category:Electronic ~~Minimization~~]][[Category:Density ~~Mixing~~]]		[[Category:INCAR]][[Category:Electronic minimization]][[Category:Density mixing]]

Kresse: Description when not to use MAXMIX added

2022-03-15T14:35:20Z

Description when not to use MAXMIX added

← Older revision		Revision as of 14:35, 15 March 2022
Line 7:		Line 7:
	:The mixer is reset after each ionic step or if the number of electronic steps exceeds \|{{TAG\|MAXMIX}}\| (this is the default and similar to the behavior of VASP.4.3 and VASP.3.2).		:The mixer is reset after each ionic step or if the number of electronic steps exceeds \|{{TAG\|MAXMIX}}\| (this is the default and similar to the behavior of VASP.4.3 and VASP.3.2).
	*{{TAG\|MAXMIX}}>0		*{{TAG\|MAXMIX}}>0
	:The charge density mixer is only reset if the storage capabilities are exceeded. The reset is done "smoothly" by removing the five oldest vectors from the iteration history. Therefore, if {{TAG\|MAXMIX}} is positive, the approximation for the charge dielectric function which was obtained in previous ionic steps is "reused" in the current ionic step, and this in turn can reduce the number of electronic steps during relaxations and MD's. Especially for relaxations which start from a good ionic starting guess and for systems with a strong charge sloshing behavior the speedup can be significant. We found that for a 12 Å long box containing 16 Fe atoms the number of electronic iterations decreased from 8 to 2-3 when {{TAG\|MAXMIX}} was set to 40. For a carbon surface the number of iterations decreased from 7 to 3. At the same time the energy stability increased significantly. But be careful, this option increases the memory requirements for the mixer considerably, and thus the option is not recommended for systems were charge sloshing is negligible anyway (like bulk simple metals). The optimal setting for {{TAG\|MAXMIX}} is usually around three times the number of electronic steps required in the first iteration. Too large values for {{TAG\|MAXMIX}} might cause the code to crash (because linear dependencies between input vectors might develop).		:The charge density mixer is only reset if the storage capabilities are exceeded. The reset is done "smoothly" by removing the five oldest vectors from the iteration history. Therefore, if {{TAG\|MAXMIX}} is positive, the approximation for the charge dielectric function which was obtained in previous ionic steps is "reused" in the current ionic step, and this in turn can reduce the number of electronic steps during relaxations and MD's. Especially for relaxations which start from a good ionic starting guess and for systems with a strong charge sloshing behavior the speedup can be significant. We found that for a 12 Å long box containing 16 Fe atoms the number of electronic iterations decreased from 8 to 2-3 when {{TAG\|MAXMIX}} was set to 40. For a carbon surface the number of iterations decreased from 7 to 3. At the same time the energy stability increased significantly. But be careful, this option increases the memory requirements for the mixer considerably, and thus the option is not recommended for systems were charge sloshing is negligible anyway (like bulk simple metals). The optimal setting for {{TAG\|MAXMIX}} is usually around three times the number of electronic steps required in the first iteration. Too large values for {{TAG\|MAXMIX}} might cause the code to crash (because linear dependencies between input vectors might develop). Too small values for {{TAG\|MAXMIX}} can slow your convergence significantly. For instance, if you need 50 self-consisitency cycles, and set {{TAG\|MAXMIX}} to 20, you force the mixer to remove iteration history continously, which can cause divergence and at least a slow down of the convergence.
			* Caution: do not set {{TAG\|MAXMIX}}>0 in the following cases. (i) If your initial positions in the {{TAG\|POSCAR}} file are far from the fully relaxed positions, the ions might move considerably during relaxation. In this case, it is not expedient to "reuse" charge mixing information from the previous ionic steps. (ii) During machine learning, the first principles calculations are often bypassed for hundreds or even thousands of ionic steps, and the ions might move considerably between first principles calculations. In this cases using {{TAG\|MAXMIX}} will very often lead to electronic divergence or strange errors during the self-consistency cycle. In general, whenever the column RMS(c) in the {{TAG\|OSZICAR}} files shows a sudden increase in the norm of the charge density residual vector, try to remove the tag {{TAG\|MAXMIX}} from the INCAR file.


	'''Mind''': {{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.		'''Mind''': {{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.

Schlipf at 09:14, 10 May 2019

2019-05-10T09:14:53Z

← Older revision		Revision as of 09:14, 10 May 2019
Line 25:		Line 25:
	----		----

	[[Category:INCAR]][[Category:Electronic Minimization]][[Category:Mixing]]		[[Category:INCAR]][[Category:Electronic Minimization]][[Category:Density Mixing]]

Karsai at 10:24, 19 April 2019

2019-04-19T10:24:28Z

← Older revision		Revision as of 10:24, 19 April 2019
Line 24:		Line 24:
	{{sc\|MAXMIX\|Examples\|Examples that use this tag}}		{{sc\|MAXMIX\|Examples\|Examples that use this tag}}
	----		----
	~~[[The_VASP_Manual\|Contents]]~~

	[[Category:INCAR]][[Category:Electronic Minimization]][[Category:Mixing]]		[[Category:INCAR]][[Category:Electronic Minimization]][[Category:Mixing]]

Karsai at 11:13, 7 March 2019

2019-03-07T11:13:16Z

← Older revision		Revision as of 11:13, 7 March 2019
Line 26:		Line 26:
	[[The_VASP_Manual\|Contents]]		[[The_VASP_Manual\|Contents]]

	[[Category:INCAR]][[Category:Mixing]]		[[Category:INCAR]][[Category:Electronic Minimization]][[Category:Mixing]]

Karsai at 10:37, 20 March 2017

2017-03-20T10:37:53Z

← Older revision		Revision as of 10:37, 20 March 2017
Line 22:		Line 22:
	{{TAG\|WC}}		{{TAG\|WC}}

	~~== Example Calculations using this Tag ==~~		{{sc\|MAXMIX\|Examples\|Examples that use this tag}}
	{{~~TAG~~\|~~liquid Si}}, {{TAG~~\|~~vibrational frequencies of CO on Ni 111 surface~~}}
	----		----
	[[The_VASP_Manual\|Contents]]		[[The_VASP_Manual\|Contents]]

	[[Category:INCAR]][[Category:Mixing]]		[[Category:INCAR]][[Category:Mixing]]

Karsai at 15:31, 15 February 2017

2017-02-15T15:31:59Z

← Older revision		Revision as of 15:31, 15 February 2017
Line 21:		Line 21:
	{{TAG\|MIXPRE}},		{{TAG\|MIXPRE}},
	{{TAG\|WC}}		{{TAG\|WC}}

			== Example Calculations using this Tag ==
			{{TAG\|liquid Si}}, {{TAG\|vibrational frequencies of CO on Ni 111 surface}}
	----		----
	[[The_VASP_Manual\|Contents]]		[[The_VASP_Manual\|Contents]]

	[[Category:INCAR]][[Category:Mixing]]		[[Category:INCAR]][[Category:Mixing]]

← Older revision		Revision as of 14:25, 6 April 2022
Line 1:		Line 1:
	{{TAGDEF\|MAXMIX\|[integer]\|-45}}		{{TAGDEF\|MAXMIX\|[integer]\|-45}}

	Description: {{TAG\|MAXMIX}} specifies the maximum number steps stored in Broyden mixer ({{TAG\|IMIX}}=4).		Description: {{TAG\|MAXMIX}} specifies the maximum number of steps stored in the Broyden mixer ({{TAG\|IMIX}}=4).
	----		----
	{{TAG\|MAXMIX}} specifies the maximum number of vectors stored in the Broyden/Pulay mixer, in other words it corresponds to the maximal rank of the approximation of the charge dielectric function build up by the mixer. {{TAG\|MAXMIX}} can be either negative or positive:		{{TAG\|MAXMIX}} specifies the maximum number of vectors stored in the Broyden/Pulay mixer, in other words, it corresponds to the maximal rank of the approximation of the charge-dielectric function build up by the mixer. {{TAG\|MAXMIX}} can be either negative or positive:
	*{{TAG\|MAXMIX}}<0		*{{TAG\|MAXMIX}}<0
	:The mixer is reset after each ionic step or if the number of electronic steps exceeds \|{{TAG\|MAXMIX}}\| (this is the default and similar to the behavior of VASP.4.3 and VASP.3.2).		:The mixer is reset after each ionic step or if the number of electronic steps exceeds \|{{TAG\|MAXMIX}}\| (this is the default and similar to the behavior of VASP.4.3 and VASP.3.2).
	*{{TAG\|MAXMIX}}>0		*{{TAG\|MAXMIX}}>0
	:The charge density mixer is only reset if the storage capabilities are exceeded. The reset is done "smoothly" by removing the five oldest vectors from the iteration history. Therefore, if {{TAG\|MAXMIX}} is positive, the approximation for the charge dielectric function which was obtained in previous ionic steps is "reused" in the current ionic step, and this in turn can reduce the number of electronic steps during relaxations and MD's. Especially for relaxations ~~which~~ start from a good ionic starting guess and for systems with a strong charge sloshing behavior the speedup can be significant. We found that for a 12 Å long box containing 16 Fe atoms the number of electronic iterations decreased from 8 to 2-3 when {{TAG\|MAXMIX}} was set to 40. For a carbon surface the number of iterations decreased from 7 to 3. At the same time the energy stability increased significantly. But be careful, this option increases the memory requirements for the mixer considerably, and thus the option is not recommended for systems ~~were~~ charge sloshing is negligible anyway (like bulk simple metals). The optimal setting for {{TAG\|MAXMIX}} is usually around three times the number of electronic steps required in the first iteration. Too large values for {{TAG\|MAXMIX}} might cause the code to crash (because linear dependencies between input vectors might develop). Too small values for {{TAG\|MAXMIX}} can slow your convergence significantly. For instance, if you need 50 self-~~consisitency~~ cycles, and set {{TAG\|MAXMIX}} to 20, you force the mixer to remove iteration history ~~continously~~, which can cause divergence and at least a slow down of the convergence.		:The charge density mixer is only reset if the storage capabilities are exceeded. The reset is done "smoothly" by removing the five oldest vectors from the iteration history. Therefore, if {{TAG\|MAXMIX}} is positive, the approximation for the charge dielectric function which was obtained in previous ionic steps is "reused" in the current ionic step, and this, in turn, can reduce the number of electronic steps during relaxations and MD's. Especially for relaxations that start from a good ionic starting guess and for systems with a strong charge sloshing behavior the speedup can be significant. We found that for a 12 Å long box containing 16 Fe atoms the number of electronic iterations decreased from 8 to 2-3 when {{TAG\|MAXMIX}} was set to 40. For a carbon surface the number of iterations decreased from 7 to 3. At the same time, the energy stability increased significantly. But be careful, this option increases the memory requirements for the mixer considerably, and thus the option is not recommended for systems where charge sloshing is negligible anyway (like bulk simple metals). The optimal setting for {{TAG\|MAXMIX}} is usually around three times the number of electronic steps required in the first iteration. Too large values for {{TAG\|MAXMIX}} might cause the code to crash (because linear dependencies between input vectors might develop). Too small values for {{TAG\|MAXMIX}} can slow your convergence significantly. For instance, if you need 50 self-consistency cycles, and set {{TAG\|MAXMIX}} to 20, you force the mixer to remove iteration history continuously, which can cause divergence and at least a slow down of the convergence.
	* Caution: do not set {{TAG\|MAXMIX}}>0 in the following cases. (i) If your initial positions in the {{TAG\|POSCAR}} file are far from the fully relaxed positions, the ions might move considerably during relaxation. In this case, it is not expedient to "reuse" charge mixing information from the previous ionic steps. (ii) During machine learning, the first principles calculations are often bypassed for hundreds or even thousands of ionic steps, and the ions might move considerably between first principles calculations. In ~~this~~ cases using {{TAG\|MAXMIX}} will very often lead to electronic divergence or strange errors during the self-consistency cycle. In general, whenever the column RMS(c) in the {{TAG\|OSZICAR}} files shows a sudden increase in the norm of the charge density residual vector, try to remove the tag {{TAG\|MAXMIX}} from the INCAR file.		* Caution: do not set {{TAG\|MAXMIX}}>0 in the following cases. (i) If your initial positions in the {{TAG\|POSCAR}} file are far from the fully relaxed positions, the ions might move considerably during relaxation. In this case, it is not expedient to "reuse" charge mixing information from the previous ionic steps. (ii) During machine learning, the first-principles calculations are often bypassed for hundreds or even thousands of ionic steps, and the ions might move considerably between first-principles calculations. In these cases using {{TAG\|MAXMIX}} will very often lead to electronic divergence or strange errors during the self-consistency cycle. In general, whenever the column RMS(c) in the {{TAG\|OSZICAR}} files shows a sudden increase in the norm of the charge density residual vector, try to remove the tag {{TAG\|MAXMIX}} from the INCAR file.


	'''Mind''': {{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.		'''Mind''': {{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.

	== Related ~~Tags~~ and ~~Sections~~ ==		== Related tags and articles ==
	{{TAG\|IMIX}},		{{TAG\|IMIX}},
	{{TAG\|INIMIX}},		{{TAG\|INIMIX}},
Line 27:		Line 27:
	----		----

	[[Category:INCAR]][[Category:Electronic ~~Minimization~~]][[Category:Density ~~Mixing~~]]		[[Category:INCAR]][[Category:Electronic minimization]][[Category:Density mixing]]

← Older revision		Revision as of 14:35, 15 March 2022
Line 7:		Line 7:
	:The mixer is reset after each ionic step or if the number of electronic steps exceeds \|{{TAG\|MAXMIX}}\| (this is the default and similar to the behavior of VASP.4.3 and VASP.3.2).		:The mixer is reset after each ionic step or if the number of electronic steps exceeds \|{{TAG\|MAXMIX}}\| (this is the default and similar to the behavior of VASP.4.3 and VASP.3.2).
	*{{TAG\|MAXMIX}}>0		*{{TAG\|MAXMIX}}>0
	:The charge density mixer is only reset if the storage capabilities are exceeded. The reset is done "smoothly" by removing the five oldest vectors from the iteration history. Therefore, if {{TAG\|MAXMIX}} is positive, the approximation for the charge dielectric function which was obtained in previous ionic steps is "reused" in the current ionic step, and this in turn can reduce the number of electronic steps during relaxations and MD's. Especially for relaxations which start from a good ionic starting guess and for systems with a strong charge sloshing behavior the speedup can be significant. We found that for a 12 Å long box containing 16 Fe atoms the number of electronic iterations decreased from 8 to 2-3 when {{TAG\|MAXMIX}} was set to 40. For a carbon surface the number of iterations decreased from 7 to 3. At the same time the energy stability increased significantly. But be careful, this option increases the memory requirements for the mixer considerably, and thus the option is not recommended for systems were charge sloshing is negligible anyway (like bulk simple metals). The optimal setting for {{TAG\|MAXMIX}} is usually around three times the number of electronic steps required in the first iteration. Too large values for {{TAG\|MAXMIX}} might cause the code to crash (because linear dependencies between input vectors might develop).		:The charge density mixer is only reset if the storage capabilities are exceeded. The reset is done "smoothly" by removing the five oldest vectors from the iteration history. Therefore, if {{TAG\|MAXMIX}} is positive, the approximation for the charge dielectric function which was obtained in previous ionic steps is "reused" in the current ionic step, and this in turn can reduce the number of electronic steps during relaxations and MD's. Especially for relaxations which start from a good ionic starting guess and for systems with a strong charge sloshing behavior the speedup can be significant. We found that for a 12 Å long box containing 16 Fe atoms the number of electronic iterations decreased from 8 to 2-3 when {{TAG\|MAXMIX}} was set to 40. For a carbon surface the number of iterations decreased from 7 to 3. At the same time the energy stability increased significantly. But be careful, this option increases the memory requirements for the mixer considerably, and thus the option is not recommended for systems were charge sloshing is negligible anyway (like bulk simple metals). The optimal setting for {{TAG\|MAXMIX}} is usually around three times the number of electronic steps required in the first iteration. Too large values for {{TAG\|MAXMIX}} might cause the code to crash (because linear dependencies between input vectors might develop). Too small values for {{TAG\|MAXMIX}} can slow your convergence significantly. For instance, if you need 50 self-consisitency cycles, and set {{TAG\|MAXMIX}} to 20, you force the mixer to remove iteration history continously, which can cause divergence and at least a slow down of the convergence.
			* Caution: do not set {{TAG\|MAXMIX}}>0 in the following cases. (i) If your initial positions in the {{TAG\|POSCAR}} file are far from the fully relaxed positions, the ions might move considerably during relaxation. In this case, it is not expedient to "reuse" charge mixing information from the previous ionic steps. (ii) During machine learning, the first principles calculations are often bypassed for hundreds or even thousands of ionic steps, and the ions might move considerably between first principles calculations. In this cases using {{TAG\|MAXMIX}} will very often lead to electronic divergence or strange errors during the self-consistency cycle. In general, whenever the column RMS(c) in the {{TAG\|OSZICAR}} files shows a sudden increase in the norm of the charge density residual vector, try to remove the tag {{TAG\|MAXMIX}} from the INCAR file.


	'''Mind''': {{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.		'''Mind''': {{TAG\|MAXMIX}} is only available in VASP.4.4 and newer versions, and it is strongly recommended to use this option for molecular dynamics and relaxations.

← Older revision		Revision as of 14:45, 8 April 2022
Line 27:		Line 27:
	----		----

	[[Category:INCAR]][[Category:Electronic minimization]][[Category:Density mixing]]		[[Category:INCAR tag]][[Category:Electronic minimization]][[Category:Density mixing]]