MicroStation 2024 CellStation Edition?

As mentioned elsewhere, BIM is heavily dependent on using components provided primarily by manufacturers. This means being able to convert RFA content into Mstn equivalents with as much fidelity and retained functionality as possible. Mstn's RFA Interpreter is a pioneering tool... but I think the results have been more 'zombie'  than production-ready, lately. Overdue for an overhaul... using ODA BIMrv libraries?

Some of the problems:

  1. Missing geometry. Not sure but I suspect that this might have to do with troubles with reverse engineering Revit tools.
  2. RFA Cell is a monolithic 'zombie' element that can not be edited without dropping and losing all the attached BIM / Item info.
  3. Properties Panel does not display the nested elements.
  4. No way to use tools like FACET MODIFY EDGEDISPLAYTOOL to clean up geometry problems
  5. Everything lands on the Default level. I suspect that this due to Mstn's Parametric Cell (Shared Cells) limitations.
  6. No support for Revit's coarse medium fine LOD settings. All LOD versions come in and land on the Default level, making the results impossible to use. No Mstn Element Template support?
  7. No support for Revit's 2d plan, section etc geometry. High hopes that this will be corrected soon now that Named Presentations has been released.
  8. No support for Revit's Visibility state parameter. Again, all versions of the geometry come in and land on the Default level, making the results impossible to use.
  9. Unreliable support for Revit's parametrics. This will probably never be comprehensive, but we need to understand better what the RFA-I's limitations are. RFA-I previously targetted OBD parametric functionality. Even this seems to getting worse. This needs to be picked up again with Mstn's Constraints and Parametric Solids functionality.
  10. No support for Revit materials, textures etc for rendering. Visualisation is a very important side product of BIM.
  11. ??

No one is really going to want to use Mstn if every component has be to remade from scratch. OBD is infamous for having very bad OOTB datasets, based on decades old content, often US origin that has been 'metrified', based on old and multiple tools accumulated over the years. This has been overlooked for a long time as a lot of the corresponding functionality like Placement Points, Perforators, Named Presentations etc in Mstn just wasn't available. There is a huge pent up demand for updating all that old content to leverage the new MSCE Update16.1,2,3 etc functionality. OBD for example desparately needs to be able to include modern content in its OOTB datasets to be competitive. The converted BIMstore RFA content on ComponentCenter is no better. Hopefully, the localisation teams have been consulted by platform and the relevant conversion tools are being written.

  • Making from scratch: due to the nature of the content and their sources, I think that Mstn will need to change to be accommodate more interop workflows.

    Finding, viewing and evaluating, converting, transformating, separating and merging and (re)assembling both geometric and nongraphic info, (re)parameterisation smart objects and content will be a large and recurring workflow... that will need some big changes.

    For buildings, most of the incoming info will be in multiple formats (RFA - Revit Families, IFC, 2d and 3d DWG. FBX (mainly for rendering), SKP (sketchup because its popular) IGS (content converted from MCAD). All of which have their good and bad points... and incomplete if you want a fully functional BIM object that has the nongraphic BIM attribute info, well-formed 3d geometry, materials / textures for rendering and the 2d plan / section proxy geometry for drawing extraction.

    For every component to be imported, you end up with a folder full of versions of what you want in different formats, and other files. They all have to viewed, evaluated and edited before being assembled to make the 'BIM-ready' component Cell for use in Mstn.

    The current RFA-Intrepreter or importer needs to be more global and address other file formats and info sources. In many ways, it is only one part of the Parametric Content Modeler that Bentley had in mind a decade or so ago. Need to resuscitate this vision. Mstn must be able to appropriate content efficiently, it does not have the market share or ecosystem to make content come to it...

  • Appropriating content efficiently: as usual, Bentley already has most of the tech available... just not weaponised... anywhere? Whenever I hear 'everbody tells us our tech is the best. We are a great tech provider' I cringe. It sounds like Bentley has resigned itself to never being able to go the last mile to give its users tools that are a joy to use.

    1. File format support: arguably already best in class with RealDWG, direct IFC, Rhino etc referencing.
    2. Comprehensive geometry gamut - solids, surfaces, meshes, rasters, point clouds, reality meshes etc. Only subdivision surfaces and pressed metal missing?
    3. MCAD level constraints and solid modeling tools
    4. Built in rendering and materials system
    5. Built in support for drawing extraction (Named Presentations, Dynamic / Building Views, Hypermodeling etc)
    6. BIM attribute info support via Item Types and Items, Imodel transformer etc
    7. Property-based annotation tools and scheduling
    8. Structural section catalog and schema support via Prostructures
    9. Support for environmental modeling via OpenBuildings Energy Simulator schemas
    10. Long standing CAD drafting tools for both 2d and 3d modeling
    11. Best in class file referencing tools (soon to be joined by a Edit Cell in place function?)
    12. Best in class for handling large models
    13. Vector pdf support on the horizon?
    14. ??

    The vast majority of BIM content online is really basic point-based 'non-parametric' objects like sanitary ware, plumbing fittings, furniture etc. As mentioned above, the content you get is almost always never ready to be used without some processing, even for Revit users. You will find that the object in mind is good enough for one stage of work but not another. Just fixing the RFA importer will not really be enough.

    Linear and plane-based  'parametric objects' that can morph based on size, length etc parameters like doors, windows, ladders, curtain walls etc are luckily relatively few. They also suffer from being designed as 'super do all' objects that are problematic to use. The strategy here should probably be to match and superset Revit's system families in functionality using Mstn's Constraints and Parametric Solids libraries. Some manual tinkering will be inevitable.

    Visualisation will be a big deal but importing, orientating / scaling and managing material / texture maps associated with the objects -whilst tedious- should be relatively straight forward? Again need to match Revit and Sketchup, FBX materials functionality.

    Should the imodel exporter plugin that is available for Revit users be updated to include an export RFA function in the meantime?

    Mstn is a 3d modeler. Mstn is a Parametric Content (re)modeler. Mstn is also a comprehensive 2d CAD app. Mstn File open, Ref attach, Cell Manager, Bentley View and DgnPreviewer are viewing apps. You need all of them in tip-top condition to produce BIM deliverables, effectively.

Reply Children
  • Mstn File open, Ref attach, Cell Manager, Bentley View and DgnPreviewer are viewing apps: Working in the assembly:component BIM world highlights the need for a step change in functionality. Mstn has a lot of powerful tools to deal with both Models and Cells. But Mstn's interface is geared primarly to editing models.

    What if we looked at how Mstn can improve working with Cells or at the component not just the container or model level? A lot of BIM is about finding and selecting those Cells before we get to the part where we place them in the container models.

    Question1: Why should we have only one Cell Manager dialog per session? Why can't we have multiple Cell Managers that can look at multiple Cell libraries or folders, instead of forcing the user to flip back and forth in a linear and unproductive way? This 'modal' way of working and accessing content content content is a big bottleneck that I think is pretty easy to fix, no?

    Question 2: the DgnPreviewer apps already allow you to click on and measure geometric elements, extract element info etc. Pan, orbit and  zoom about. The Cell Manager should also have this functionality and more editing functions. Working with components will mean needing work on the components in their libraries and not just working on them after they are inserted and replacing them once you are done.

    Similarly, other entry points like the File Open and Ref Attachment dialog should be more capable and better leverage the DgnPreviewer functionality.

    Question 3: Should Cut/Copy and Paste be supported between the Cell Manager windows? Cell managers are slightly unique for the user as it allows the user to edit a file that he hasn't opened through the File Open dialog or double clicking via Windows Explorer. It would make a lot of tasks easier and allow the user to use the Cell manager as a staging area to assemble and tweak a Cell before inserting into the active model (or another Cell).

    Question 4: All of this too much for Mstn? Why not then use Window Explorer folders as 'pallettes'. This require the DgnPreviewer to be enhanced to provide write, docking capability etc.

    As mentioned elsewhere, it would be great to leverage Screen Menus to avoid loading the dgns and provide the means to extract the Cell directly from the Dgnpreview in Windows Explorer. Screen Menu stores the ElementID with the Hyperlink?