NUMECA FINE / Open with OpenLabs 9.2
NUMECA FINE / Open with OpenLabs 9.2 | 2.8 Gb
NUMECA International is delighted to announce the official release new version of FINE/Open With OpenLabs is powerful CFD Flow integrated environment. Release 9.2 - is average 10% speed-up for all generated meshes with HEXPRESS/Hybrid.
Our development team worked on different aspects to globally improve the speed of all generated meshes. As a result, a speed-up of 10% can be expected for all cases on average. In addition, particular attention has been put on the automatic local refinement for surfaces to accelerate it even further.
Improvements in the mesh quality for open thin wall configurations have also been tackled.
The distributed parallel version has been upgraded with the capacity to use the enhanced mesh projection method (EXACTMESHING keyword) and the support of SPB and CGNS formats for export.
The computation of Full Non-Matching Boundaries (FNMB) for periodic domains can be performed directly in HEXPRESS/Hybrid: faster time-to-results with a simpler and more integrated workflow.
![NUMECA FINE / Open with OpenLabs 9.2]()
Custom thermodynamic tables (FINE/Open)
Custom thermodynamic table data can be generated with the new version of NIST Reference Fluid Thermodynamic and Transport Properties Database (REFPROP).
REFPROP 10 is integrated into NUMECA’s tables generator TabGen. The benefits of this new version are:
- Availability of 147 pure fluids
- Support for mixtures with up to 20 components
- New equations of state for some fluids
CPU Booster robustness improvements (FINE/Open)
Speed-up your most complex simulations thanks to the improved robustness of CPU-Booster.
Users get more advanced control on CPU-Booster such as the initial constant CFL for the CFL ramping, compressibility correction for supersonic exhaust jets, additional outputs…
![NUMECA FINE / Open with OpenLabs 9.2]()
Other improvements
HEXPRESS/Hybrid
- Capacity to smoothly close viscous layers by hanging nodes between 2 tangent surfaces to ensure an optimal quality
- Capacity to intersect mirror planes and geometries based on the triangulation
- New license check option to wait for a free license seat instead of stopping mesh generation (useful for automatic optimization and job scheduling)
FINE/Open with OpenLabs
- Rigid-body motion with 6 degrees of freedom
- Introduction of the CFL ramping for computations without CPU-Booster
- New best practices guide for premixed combustion
Improvements in the mesh quality for open thin wall configurations have also been tackled.
The distributed parallel version has been upgraded with the capacity to use the enhanced mesh projection method (EXACTMESHING keyword) and the support of SPB and CGNS formats for export.
The computation of Full Non-Matching Boundaries (FNMB) for periodic domains can be performed directly in HEXPRESS/Hybrid: faster time-to-results with a simpler and more integrated workflow.
NUMECA FINE / Open with OpenLabs 9.2
Custom thermodynamic tables (FINE/Open)
Custom thermodynamic table data can be generated with the new version of NIST Reference Fluid Thermodynamic and Transport Properties Database (REFPROP).
REFPROP 10 is integrated into NUMECA’s tables generator TabGen. The benefits of this new version are:
- Availability of 147 pure fluids
- Support for mixtures with up to 20 components
- New equations of state for some fluids
CPU Booster robustness improvements (FINE/Open)
Speed-up your most complex simulations thanks to the improved robustness of CPU-Booster.
Users get more advanced control on CPU-Booster such as the initial constant CFL for the CFL ramping, compressibility correction for supersonic exhaust jets, additional outputs…
NUMECA FINE / Open with OpenLabs 9.2
Other improvements
HEXPRESS/Hybrid
- Capacity to smoothly close viscous layers by hanging nodes between 2 tangent surfaces to ensure an optimal quality
- Capacity to intersect mirror planes and geometries based on the triangulation
- New license check option to wait for a free license seat instead of stopping mesh generation (useful for automatic optimization and job scheduling)
FINE/Open with OpenLabs
- Rigid-body motion with 6 degrees of freedom
- Introduction of the CFL ramping for computations without CPU-Booster
- New best practices guide for premixed combustion
NUMECA FINE / Open with OpenLabs 9.2
The resolution of Computational Fluid Dynamics (CFD) problems involves three main steps:
- spatial discretization of the flow domain,
- flow computation,
- visualization of the results.
To perform these steps NUMECA has developed three software systems. The first software system, HEXPRESS, is an automated all-hexahedral unstructured grid generation system. HEXPRESS also includes a module HEXPRESS/Hybrid, which allows to mesh complex geometries with an isotropic hybrid mesh including mainly hexahedral cells with tetrahedral, pyramid and prism cells. The second software system, FINE/Open with OpenLabs, is a powerful CFD Flow Integrated Environment dedicated to complex internal and external flows. The third software system, CFView, is a highly interactive Computational Field Visualization system.
These three software systems have been integrated in a unique and user friendly Graphical User Interface (GUI), called FINE/Open with OpenLabs, allowing the solution of complete simulations of 3D internal and external flows from the grid generation to visualization, without any file manipulation, through the concept of a project. Moreover, multi-tasking capabilities are incorporated, allowing the simultaneous treatment of multiple projects.
FINE/Open with OpenLabs fully integrates the concept of multitasking. This means that the user can manage a complete project in the FINE/Open GUI; making the grid using HEXPRESS, running the computation with the FINE/Open solver and visualizing the results with CFView. Furthermore, the user has the possibility to start, stop and control multiple computations.
418M multidomain mesh automatically generated in 4.5 hiyrs with HEXPRESS/Hybrid. Full multiphysics CHT coupling with FINE/Open radiation and heat exchanger macro model.
Numeca International Inc develops computational fluid dynamics (CFD) software for the simulation, design, and optimization of fluid flow and heat transfer worldwide. It offers automated and customized flow integrated environments (FINE), including FINE/Turbo for the simulation of rotating and turbomachinery flows; FINE/Open for the simulation of complex internal and external applications; and FINE/Marine for marine applications.
Product: FINE/Open With OpenLabs
Version: 9.2 *
Supported Architectures: x64
Website Home Page : www.numeca.com
Language: english
System Requirements: PC **
Supported Operating Systems: **
Size: 2.8 Gb
- FINE/Open with OpenLabs v9.2 is packaged with CFView v14.2
- FINE/Open with OpenLabs v9.2 is provided with HEXPRESS v9.2.
- HEXPRESS/Hybrid v9.2 is also included in the installation of FINE/Open with OpenLabs.
- FINE/Open with OpenLabs v9.2 is provided with HEXPRESS v9.2.
- HEXPRESS/Hybrid v9.2 is also included in the installation of FINE/Open with OpenLabs.
Supported platforms and operating systems
NUMECA software is supported in 64 bits on x86_64 processors only (64 bits processors that are compatible with Intel 8086 set of instructions).
Formally tested on
- Windows 8.1 Professional -
- Windows 10 Professional -
Please also notice that the following limitations apply:
- Computations under Windows (except Windows Server) are limited to one PVM daemon running at a time. As a consequence, access to multiple users on a given machine is not allowed,
- Parasolid uses SSE2 instruction for high performance. If your platform is not supporting the SSE2 instruction, the FINE™ GUI cannot be started.
- AMD EPYC and RYZEN processors are slower on Windows 10 compare to Linux.
Multiprocessors calculations
Multiprocessors calculations are allowed on shared and homogeneous distributed platforms. Non-homogeneous configurations are not allowed.
The following limitations additionally apply:
- Multiprocessors calculations are allowed on distributed platforms in local user configurations provided that the same account is used for all machines,
- Domain administrators are not allowed to run multiprocessors calculations.
- MPICH2 library needs Microsoft .net Framework package before installing NUMECA software otherwise execution of mpiexec.exe will lead to a Windows error message.
Hardware requirements
Next to the standard hardware (monitor, keyboard and mouse), some specific recommendations apply in the use of NUMECA software:
- a mouse with scroll wheel is strongly recommended.
- the monitor should support 24-bit color graphics and have a 1280 x 1024 pixel resolution for adequate visualization.
- an Ethernet card should be installed and properly configured.
- NUMECA Software does not support CXFS (Clustered XFS) and IBRIX disk file systems.
Memory and disk space requirements
The Random Access Memory (RAM) required to run NUMECA software depends upon several factors, in particular including the number of grid points inserted in the grid, the physical models selected (turbulence, adaptation,…), the introduction of non-matching boundary conditions and the selected calculation mode (mixed or double precision).
Similarly, a minimum disk space is required on the hard disk to allow the storage of the project files. The size of these files largely depends upon the number of grid points generated.
More product specific details
HEXPRESS
The use of minimum 1 GB RAM is strongly recommended. As a general guideline, about 0.5 to 0.7 GB RAM are required to run a one million nodes project. The suggested swap space should be equivalent to at least 3 times the RAM installed.
As a general guideline, a minimum of 300 MB is required to store the project files (provided that the geometry is not defined by Parasolid™ or CATIA files) for a one million nodes project.
HEXPRESS/Hybrid
The Random Access Memory (RAM) required to run HEXPRESS™/Hybrid depends upon several factors. In particular, the complexity of geometries and the settings of the CONF file. As a general guideline, the Random Access Memory (RAM) required to run HEXPRESS™/Hybrid are:
- for mesh without viscous layers - between 500 MB and 700 MB RAM per million of cells
- for mesh with viscous layers - between 700 MB and 900 MB RAM per million of cells
The use of minimum 4 GB RAM is mandatory, but the installation of 8 GB is recommended. The suggested swap space should be equivalent to at least three times the RAM installed.
About 1 GB is required to store a two million nodes mesh in SPH format.
OpenFINE/Open with OpenLabs
The use of minimum 1 GB RAM is mandatory, but the installation of 2 GB is recommended. As a general guideline, about 1.5 GB RAM is generally required to run a one million nodes project in double precision mode. The suggested swap space should be equivalent to at least 3 times the RAM installed.
Mesh partitioning for parallel computations requires about 1 GB memory per million cells. When using serial partitioner, the user must ensure that the memory on the head node is sufficient, because the partitioning is performed only on the head node in case of serial partitioning.
As a general guideline, a minimum 1.4 GB is required to store the mesh and solution files (one file, no full non-matching connections, solution stored after one iteration) when running a one million nodes project.
CFView
The use of minimum 1 GB RAM is strongly recommended. As a general guideline, about 250 MB RAM are required to visualize a one million nodes project. The suggested swap space should be equivalent to at least 3 times the RAM installed.
No specific disk space is required on the hard disk for CFView except if images or data files are saved or if the meridional average is created in CFView (turbomachinery application).
Graphics requirements
NVidia graphics cards are fully supported, installed together with the latest drivers.
NUMECA software makes use of the advantages of the available Graphics Card. Some trouble may however come up with the default driver OPENGL since it may not be supported by the user host computer and machine display. In order to set the system compatible with a more portable driver, typically MSW driver is used under Windows instead. Note however that the proposed drivers are usually less efficient in terms of graphical speed than the default driver.
Compilation requirements for OpenLabs
Customization of the CFD models using OpenLabs requires the possibility for compilation.
On Windows 64 bit, Microsoft Visual C++ Build Tools 2015 must be installed: make sure to select the component Windows 10 SDK 10.0.10240 during the installation.
NUMECA software is supported in 64 bits on x86_64 processors only (64 bits processors that are compatible with Intel 8086 set of instructions).
Formally tested on
- Windows 8.1 Professional -
- Windows 10 Professional -
Please also notice that the following limitations apply:
- Computations under Windows (except Windows Server) are limited to one PVM daemon running at a time. As a consequence, access to multiple users on a given machine is not allowed,
- Parasolid uses SSE2 instruction for high performance. If your platform is not supporting the SSE2 instruction, the FINE™ GUI cannot be started.
- AMD EPYC and RYZEN processors are slower on Windows 10 compare to Linux.
Multiprocessors calculations
Multiprocessors calculations are allowed on shared and homogeneous distributed platforms. Non-homogeneous configurations are not allowed.
The following limitations additionally apply:
- Multiprocessors calculations are allowed on distributed platforms in local user configurations provided that the same account is used for all machines,
- Domain administrators are not allowed to run multiprocessors calculations.
- MPICH2 library needs Microsoft .net Framework package before installing NUMECA software otherwise execution of mpiexec.exe will lead to a Windows error message.
Hardware requirements
Next to the standard hardware (monitor, keyboard and mouse), some specific recommendations apply in the use of NUMECA software:
- a mouse with scroll wheel is strongly recommended.
- the monitor should support 24-bit color graphics and have a 1280 x 1024 pixel resolution for adequate visualization.
- an Ethernet card should be installed and properly configured.
- NUMECA Software does not support CXFS (Clustered XFS) and IBRIX disk file systems.
Memory and disk space requirements
The Random Access Memory (RAM) required to run NUMECA software depends upon several factors, in particular including the number of grid points inserted in the grid, the physical models selected (turbulence, adaptation,…), the introduction of non-matching boundary conditions and the selected calculation mode (mixed or double precision).
Similarly, a minimum disk space is required on the hard disk to allow the storage of the project files. The size of these files largely depends upon the number of grid points generated.
More product specific details
HEXPRESS
The use of minimum 1 GB RAM is strongly recommended. As a general guideline, about 0.5 to 0.7 GB RAM are required to run a one million nodes project. The suggested swap space should be equivalent to at least 3 times the RAM installed.
As a general guideline, a minimum of 300 MB is required to store the project files (provided that the geometry is not defined by Parasolid™ or CATIA files) for a one million nodes project.
HEXPRESS/Hybrid
The Random Access Memory (RAM) required to run HEXPRESS™/Hybrid depends upon several factors. In particular, the complexity of geometries and the settings of the CONF file. As a general guideline, the Random Access Memory (RAM) required to run HEXPRESS™/Hybrid are:
- for mesh without viscous layers - between 500 MB and 700 MB RAM per million of cells
- for mesh with viscous layers - between 700 MB and 900 MB RAM per million of cells
The use of minimum 4 GB RAM is mandatory, but the installation of 8 GB is recommended. The suggested swap space should be equivalent to at least three times the RAM installed.
About 1 GB is required to store a two million nodes mesh in SPH format.
OpenFINE/Open with OpenLabs
The use of minimum 1 GB RAM is mandatory, but the installation of 2 GB is recommended. As a general guideline, about 1.5 GB RAM is generally required to run a one million nodes project in double precision mode. The suggested swap space should be equivalent to at least 3 times the RAM installed.
Mesh partitioning for parallel computations requires about 1 GB memory per million cells. When using serial partitioner, the user must ensure that the memory on the head node is sufficient, because the partitioning is performed only on the head node in case of serial partitioning.
As a general guideline, a minimum 1.4 GB is required to store the mesh and solution files (one file, no full non-matching connections, solution stored after one iteration) when running a one million nodes project.
CFView
The use of minimum 1 GB RAM is strongly recommended. As a general guideline, about 250 MB RAM are required to visualize a one million nodes project. The suggested swap space should be equivalent to at least 3 times the RAM installed.
No specific disk space is required on the hard disk for CFView except if images or data files are saved or if the meridional average is created in CFView (turbomachinery application).
Graphics requirements
NVidia graphics cards are fully supported, installed together with the latest drivers.
NUMECA software makes use of the advantages of the available Graphics Card. Some trouble may however come up with the default driver OPENGL since it may not be supported by the user host computer and machine display. In order to set the system compatible with a more portable driver, typically MSW driver is used under Windows instead. Note however that the proposed drivers are usually less efficient in terms of graphical speed than the default driver.
Compilation requirements for OpenLabs
Customization of the CFD models using OpenLabs requires the possibility for compilation.
On Windows 64 bit, Microsoft Visual C++ Build Tools 2015 must be installed: make sure to select the component Windows 10 SDK 10.0.10240 during the installation.
Please visit my blog
Added by 3% of the overall size of the archive of information for the restoration
No mirrors please
![NUMECA FINE / Open with OpenLabs 9.2]()
Added by 3% of the overall size of the archive of information for the restoration
No mirrors please
NUMECA FINE / Open with OpenLabs 9.2
