Mstn CE Parametric Solids Roadmap?

Looking at the number of improvements that have been rolled out as part of Updates 1 to 6; it seems like the pace of improvements / enhancements is slowing down?

Working with a large engineer who has been using Revit because they found Revit's parametrics to be more capable, even though ProjWise and OpenPlant were used.

 

Parents
  • Hi All, we do take 3D modeling very seriously and we have been working on improving and expanding our tools and capabilities. We have introduced some enhancements in our MicroStation CONNECT Edition Update 5 (you can check our latest post in the MicroStation blog) and we are doing our best to bring in some more in the future releases.

    Stay tuned!

  • CE was first released September 2015, almost two years ago. It would be good to see progress soon.

  • Looking at the Mstn What's New section on docs.bentley.com, there has been a total of seven new or enhanced features, and two tweaks.

    U4: Array Feature, 3D Angle Constraint, Ability to Rotate Parametric Cells on Placement, Trim Solid (added to Ribbon)

    U5: Ability to Apply Radius or Diameter Constraint, Ability to Protrude a Profile Through Desired Distance, up to Solid or Selected Face, Ability to Use the Same Element as an Input in Multiple Parametric Features,

    Ability to Use the Same Element as an Input in Multiple Parametric Features, Ability to reverse the direction of the path

    U6: 3D Angle Constraint (for reals, now? Introduced in U4)

    This would equate to about three and half features per subscription cycle. Hopefully, the dev team are concerntrating on stability and bug fixes. Hopefully, the PS team has gotten to know Mstn expected behaviour better as well.

    I realise that there is a lot to be done to get even seemingly 'simple' stuff to work, but it would be good to understand what is being pushed. At the moment, my impression is there is little use of the Parametric Solids and constraints as most users are still waiting for the tools to be come more usable / stable; and the benefits more apparent.

    Goes without saying, PS is only one part of '3d' in Mstn. The ancient ACS system tools are way overdue for a revamp; and even trusty Accudraw could do with a push. And, the display system (including DV's) as well is not really working synergistically with PS. Yup, need to think holistically to deliver that psychoactive killer UX that will wake the punters sleeping in the pews up.

    Even leaving aside more ambitious goals like Parametric Content Manager (Family Editor ala Bentley?) RFA-I / IFC parametrics integration, iLogic-style rules-based solver or GC integration, there is still a lot of work needed to really unlock productivity gains for the user... even if it does appear tantalisingly close.

  • Unknown said:
    And, the display system (including DV's) as well is not really working synergistically with PS

    Could you please share more details on this?

    Thanks

    Mahendra

       

  • Hi Mahendra,

    1. Display System:

    a. Parametric Solids are usually based on 2d geometry that are part of a constraints solved set. There needs to be a means of controling the visibility of these elements in more granular way. Assigning them to the Construction Class is too clumsy. The user should have the option to switch the support geometry on/off by element or Cell.

    b. Parametric Solids will need a lot of use of Sprites/Glyphs. Things like manipulators or the widgets for manipulating the ACS are going to be a lot more important. There should be a combined Solids / Display API and templates. See how clumsily Progenio, Gold level Partner, uses ACS widgets and edge/face selection. There should a readily available means to display Solids information buried in the Parasolids kernel like vertex, edge etc index numbers.

    c. Sketch planes: when the user needs to manipulate the elements on the sketch plane, it would be good to be able to grey out everything outside or any element not a member of the sketch plane. This is a common thing on MCAD and other apps.

    d. Cells: Similar to above. It would be good to be able to grey out what is outside the Cell. This functionality should also recognise the nesting depth of the element. See the way Sketchup's Component editing works.

    c. ACS widget on the screen is pretty basic. It should be updated to function a bit more like the ViewCube or Catia's Compass. The user should be able to click on a particular plane of the ACS and get the view to rotate to that view. Actually, a lot of the functions are already there in Dialog ViewRotation, which just needs to be updated and integrated into CE.

    d. The grid plane in Mstn follows the ACS. What would be good is if the extents of the grid plane can be controlled better. This will give the user much better visual cues and situational awareness. It would be good to have a clipped 'workplane' instead of an infinte plane for a lot of cases. It would be good to have a tri-workplane element. When working on a solid or component in 3d, it is almost a certainty that the three 'cardinal' planes with be used a lot. Eventually, I suspect/hope the Parametric Solids will need to include the means to save local coordinate systems with the Functional Component. Most of the constraints solving only works on a plane in 2d. It would be good to be able to convert this to workplanes quickly, when needed.

    e Synchronised Views. Modeling in 3d is a multi-window process. Mstn has long had a lot of powerful stuff that is based on the view windows like Active Depth, Display Depth/Front/Back Clip etc. It would be good to allow the user to synchronise mutiple windows so that when the view is rotated in one view, the other views can be synchronised. I think Descartes can already do this. It would be good to mainstream this.

    f. Staging area tools: I noticed that the new OpenBridge Modeler has what looks like a Cell Manager-like dialog that has 'full' access to geometry (ACS, dims and snapping etc). It would be good to update the Cell Manager or provide a Parametric Content Manager-type tool so that the user can 'customise' the parametric object before it is placed and mixed into the active model. See also how Catia's PowerCopy works. Working with parametric solids will mean working with parameters or geometric features  that will need to be linked to what is already in the model.

    g. ...?

  • ...

    g. Displaying like centre-lines, insertion points / local coordinate systems, grids, sketch planes associated with the Parametric Solid / Functional Component is becoming very important. Especially when some of that '2d' geometry is going to be used to place, orientate, connect and constrain objects accurately and smartly. See how Smartplant, SolidWorks etc provide 'Display Rules' for this.

    h. ..?

    2. Dynamic Views

    a. It would be good to enhance the DV by Element tool to recognise Cells. Parametric modeling will commonly involve modeling components / Cells in assemblies. Dynamic View's Clip Volume functionality would be used a lot to isolate and modify the Cell's contents.

    • User calls DV tool, selects Cell. DV tool highlights Cell's Boundary Box. Tool offers option for user to hover over the BB faces to select the Section Clip plane. Also offers an option to tunnel into the Cell to find any nested Cell's BB (Reset or Tab?)
    • User selects the Cell he wants to isolate, and is then offered the option to select another Cell or element to isolate. If the user selects another Section Clip plane, he is asked if this will override the previously picked SC plane.

    The DV Tool should also have the option to use any local coordinate systems or sketch planes stored in the Cell. As mentioned above, these coordinate systems  are pretty key to the way the Assembly:Component modeling that Parametric Solids / Functional Components will be used. The default BB's generated are usually based on the global coordinate system and is not as useful as a result.

    When the user sics the DV tool on the desired FC, it should harvest the local coordinate systems and sketch planes, and present these to the user on screen as temporary Markers. This will allow the user to select the sketch plane he would like to edit... right off the bat. The visibility of the SP should be separate to the Clip Volume / Section Clip plane. Please note the need to allow for projections of off-plane geometry to allow this information to drive/participate in the constraints solving. This is something that most of the MCAD apps have had to provide over the years after their users started to bump up the limitations of 2d-only sketch planes. Pretty sure AEC users will find the same problems in due course.

    Interesting to note that, Parametric Solids doesn't do 3d sketches, while new partner Siemens' / old parent Intergraph's Solid Edge has a 3d sketch tool.

    b. BIM workflows rely heavily on the use of LOD or View-dependent representations of the Parametric Solids. All the BIM authoring apps on the market some sort of tool to generate these proxies. A 3d door Cell will have a 2d proxy for plan drawings, for example. DV/CVE's provide an elegant means of automatically doing this, but needs a few enhancements to make this production-environment ready.

    When a FC is modified or flexed, the proxy representation needs to be updated as well. If the mods are done in the dgn or .cel containing the FC, the updates could be stored there. But, this is unworkable if the FC can flex to generate an infinite number of variants. Ideally, the proxy representation should be stored with the FC in the host dgn model. The DV/CVE etc settings could/should be stored in the master .dgn or .cel.

    It would be good to look at what verticals like ABD have been providing over the years and mainstream that functionality at platform level. ABD has been wrestling with view-dependent Compound Cells, Single-line modeling / centre-lines for things like piping/ducts, walls etc, Drawing Rules for annotation, unification etc etc for years with mixed results. I can see a lot of other verticals new to the drawing production game like OpenPlant or OpenBridge benefiting.

Reply
  • ...

    g. Displaying like centre-lines, insertion points / local coordinate systems, grids, sketch planes associated with the Parametric Solid / Functional Component is becoming very important. Especially when some of that '2d' geometry is going to be used to place, orientate, connect and constrain objects accurately and smartly. See how Smartplant, SolidWorks etc provide 'Display Rules' for this.

    h. ..?

    2. Dynamic Views

    a. It would be good to enhance the DV by Element tool to recognise Cells. Parametric modeling will commonly involve modeling components / Cells in assemblies. Dynamic View's Clip Volume functionality would be used a lot to isolate and modify the Cell's contents.

    • User calls DV tool, selects Cell. DV tool highlights Cell's Boundary Box. Tool offers option for user to hover over the BB faces to select the Section Clip plane. Also offers an option to tunnel into the Cell to find any nested Cell's BB (Reset or Tab?)
    • User selects the Cell he wants to isolate, and is then offered the option to select another Cell or element to isolate. If the user selects another Section Clip plane, he is asked if this will override the previously picked SC plane.

    The DV Tool should also have the option to use any local coordinate systems or sketch planes stored in the Cell. As mentioned above, these coordinate systems  are pretty key to the way the Assembly:Component modeling that Parametric Solids / Functional Components will be used. The default BB's generated are usually based on the global coordinate system and is not as useful as a result.

    When the user sics the DV tool on the desired FC, it should harvest the local coordinate systems and sketch planes, and present these to the user on screen as temporary Markers. This will allow the user to select the sketch plane he would like to edit... right off the bat. The visibility of the SP should be separate to the Clip Volume / Section Clip plane. Please note the need to allow for projections of off-plane geometry to allow this information to drive/participate in the constraints solving. This is something that most of the MCAD apps have had to provide over the years after their users started to bump up the limitations of 2d-only sketch planes. Pretty sure AEC users will find the same problems in due course.

    Interesting to note that, Parametric Solids doesn't do 3d sketches, while new partner Siemens' / old parent Intergraph's Solid Edge has a 3d sketch tool.

    b. BIM workflows rely heavily on the use of LOD or View-dependent representations of the Parametric Solids. All the BIM authoring apps on the market some sort of tool to generate these proxies. A 3d door Cell will have a 2d proxy for plan drawings, for example. DV/CVE's provide an elegant means of automatically doing this, but needs a few enhancements to make this production-environment ready.

    When a FC is modified or flexed, the proxy representation needs to be updated as well. If the mods are done in the dgn or .cel containing the FC, the updates could be stored there. But, this is unworkable if the FC can flex to generate an infinite number of variants. Ideally, the proxy representation should be stored with the FC in the host dgn model. The DV/CVE etc settings could/should be stored in the master .dgn or .cel.

    It would be good to look at what verticals like ABD have been providing over the years and mainstream that functionality at platform level. ABD has been wrestling with view-dependent Compound Cells, Single-line modeling / centre-lines for things like piping/ducts, walls etc, Drawing Rules for annotation, unification etc etc for years with mixed results. I can see a lot of other verticals new to the drawing production game like OpenPlant or OpenBridge benefiting.

Children
No Data