a few years ago (2008) there was a labs preview technology named iMap for Inventor.
Maybe someone could remember. I think this was the great tool. But why this tool was never released? If this tool is extended to part feature browser to know which sketch is used by which feature, and wich projected geometry is from which edge, or which adaptive thing is from which part adaptive? e.g Reference 1 in Browser like now, could not help anyone.
please bring back the iMap feature for assembly, part and drawing!!
very good question.
maybe for many of us it would be better time spent rolling this out than all the shiny materials and colours and ribbon buttons that we have forced upon us.
This is so right.
Too much of our valuable design time goes to blindfolded data search.
CAD/PLM developers should understand that the designs that we create are not only 3d component assemblies. They are in fact large collections of interconnected data items. With - like it or not - very complex structures of dependency.
In order to let these dependencies become interesting again for designers, CAD software should focus on managing (i.e. visualising) this data structure.
The only way to limit cad complexity to a useful extent and have it work for you, is when you can SEE the complex data structure while you are making it.
Eventually people will see that CAD assemblies are no longer to be represented as mere tree structures, as they are networked structures.
Visualising not only 3d-models, but also data structure, would also help managers make better decisions CAD policies, even when they have no CAD experience.
I would argue that without something like Imap, soon we will all be stuck. The amount of money that is currently wasted, is just hallucinating.
Also I think that allowing the user to have access to the real data structure of his or her work necessitates an open source approach to CAD, one way or another. You cannot open data and keep it closed simultaneously. Business models will just have to cope with that.
For the record: I have been made to switch from Inventor to Creo, only to find that in such a so-called 'high end' application these data structure issues become even more painful. If anyone is familiar with the Reference Viewer in Creo 2.0, they will know what I mean. It seems easier to map the whole data history of Facebook, than to find one faulty link in your own assembly ;-)
I think if CAD vendors don't come up with something, maybe user groups will. Why not have both of them team up on this one?
Sorry to have picked up this thread so late, but I gathered it is jus too important an issue to drop.
Tom De Prins
Hi! Currently we are doing some investigation of this tool to see how it will work with Inventor. Aside from the existing functionality provided by the original iMap on the Lab, do you have any suggestion on new capability? Or you are pretty much happy with what iMap offered?
I never got to test the imap tool because it was not released for XP64. At the time, my idws did not fit in 3 GB of RAM so I could not use XP32.
I would like to see a tool that shows me the source of any projected geometry/copied workfeature/derived geometry and lets me edit the link.
For example, I have a sketch with projected edges. I want to see which features/sketches control those projected edges. Later on, I change the source feature. Now the projected edge/endpoints get sick (pink). I want an option to edit the projected geometry and attach it to a new edge/point. Inventor should then update the sketch using the new source edge.
If I had this option, I would not have to delete the sick edge, break any sketch endpoint constraints to the sick endpoints, move the lines/arcs away from the sick endpoints, delete the sick endpoints, re-project the new edge and then re-constrain any sketch geometry back to the projected geometry.
I would also like to see the ability to find all assembly constraints/joints attached to a particular face/feature of part A and change them to apply to a particular face/feature of part B. That way, I don't have to edit 27 individual constraints to replace part A with part B in an assembly.
The new tool should also show/edit any shared parameters.
Finally, I would like to see both parents that push info into my part/assembly and children that consume it. The latter option may require Vault to track all the dependencies.
Hello Johnson and others,
Nice to hear iMap is not gone. Regrettedly, I never got to testing it, though.
Anyway, here is what I think:
People are reluctant of using parametric performance because they can't figure out how to manage it later on. Still, all cad users really do is create links between data items. Both spatial and non-spatial. One should not be harder to do than the other. When you draw a circle in a sketch, you make a link between the 'circle' concept and the sketch containing the circle. When you use projected geometry, you also make a link between this circle and values from another model.
To me this means we have to integrate both sides of the story into one GUI. Visualizing our cad database as a web of connections. And make better informed edits to it. Cad designers may start appreciating data and data people may get interested in cad ;-) New tools require new ways of thinking.
I would like to track down every thinkable relation between items, like how/how many times a subassembly is used, the type of constraints which define it, constraints to it, in turn defining other assemblies, who is/was working on them, links to reports... basically: I'd want to use it for everything you would do in cad ànd plm. Like e.g. a Copy Design with on-the-fly subassembly sustitution.
Add Helicopter view: simultaneous view of multiple dependency threads. It is very hard to grasp the complexity (==> hard to simplify it!) when you have to track down dependencies for every single file.
- At all times you should have control over how detailed the graph should be. You also should be able to start editing things, while the rest of the information is still loading in the background.
Looking further into the future:
- Why should we stick to assemblies, parts and drawings as the sole carriers of parameters? Couldn't we see files as only a partial and momentary output of this networked dataset. Since only the timeline and the connections are really what matters. Not the file formats.
In reality we rarely change all of a file at once, just links, somewhere along a model tree (model web, that is!). Still, with every change, the system considers the whole file to be altered and iterate. This makes us send lots of mostly identical datasets across our networks, worring people about their outdated versions.
I was inspired by information mapping stuff you can find on visualcomplexity.com, Infovis, Cytoscape, D3, gephi, ... Also by the way Touchgraph Googlebrowser represents highly intertwined datasets.
Tom De Prins
Access a broad range of knowledge to help get the most out of your products and services.