Speckle intelligence - Issues and features ideas

Hi,
I recently Started testing Speckle Intelligence using both IFC files and direct uploads via the Speckle connector in Tekla Structures.

I faced few issues

  • When using IFC files, the interaction between the model viewer and other visuals appears to be broken. For example, hovering over an element in a pivot table correctly highlights it in the viewer, but when I click the element, the entire model becomes ghosted. I haven’t had this issue with models uploaded through the Speckle connector.

  • Performance with large models the dashboard page crashes when working with larger models. I’m not sure whether this is related to my laptop’s memory.

I also have few ideas for improvement:

It would be great to have the following features:

-Configurable units of measurement: Currently, volumes display in cubic centimeters and areas in square millimeters, which are not standard units for reporting.

-More control over tables and pivot tables: For example, the ability to rename columns, sort values, and search within tables.

-Ability to link external data (e.g., CSV, XLS) with the 3D model data.(This could help mitigate the fact that Tekla connector doesn’t move assembly level data)

Thank you the great work so far the dashboards look very promising, and I’m excited to see how it evolves.

3 Likes

I would also appreciate this addition for the same reason

Is there any chance you could point us to a minimal example where this is reproduced? it would save days of faffing on our end!

All other things noted. Linking external data sources is on our roadmap, as well as most of the others.

Units - we’re using whatever the host software gives us, but luckily we have the lib to convert them to others easily - there will be some ux things to nail but it’s all in the doable realm.

1 Like

Hi,
what do you need from my side to check this? would a link to the dashboard be enough? or do you need the IFC file?

the dashboard where you observe it happening

We have logged these features for the backlog @MoLaach