We value your input! Please participate in Archicad 28 Home Screen and Tooltips/Quick Tutorials survey
2023-11-01 07:39 PM
Our office has been working with point clouds in the last two years, and we keep having issues navigating the models when the layer the point cloud is in is turned on. We do a lot of retrofit projects, and the point clouds are a fantastic tool for getting a lot of existing information, but it has been very difficult to model these features because of the difficulty in navigating the model, whether in 3d or 2d.
So the question is: is there a better way to import or work with these point clouds so they don't bog down our model? Sometimes just turning on the point cloud layer makes the model inoperative for five or ten minutes.
I don't remember having these issues in Revit, and from looking at tutorials, it seems like we should be able to navigate the model a lot easier with these files.
The details:
- We scan our projects using a Matterport scanner, and then download the point cloud from their website as a '.xyz' file format.
- In AC, we then go through the process of turning the point cloud into a '.lcf' file, and then we upload the file to our BIMcloud server.
- The point cloud is then placed in our AC file as a BIMcloud library.
- All of our libraries are in the BIMcloud server, and all of our projects are teamwork files.
All of these projects are single family homes or apartments. The latest point cloud file that I'm having issues with is 389.5 MB
Everyone in our office is on AC24
My computer is running: MacOS 12.6.8; Apple M1 Max, 64 MB Memory
2023-11-02 03:59 AM
Have you been able to test the Apple Silicone version of AC26 to see if it is still a problem ?
2023-11-03 02:56 PM
I haven't tried that yet. We are waiting a little longer to get the next upgrade, but the problem is still there even in the other Macs in our office that have the Intel chip.
2023-11-03 03:07 PM
389MB point cloud file for a residential project seems quite dense. Have you tried reducing the points via subsampling?
Cloud Compare is open source and is fairly straightforward with fast way of doing point cloud subsampling (space/random methods).
There are several video tutorials on YT on how to achieve this.