NOTE: Scan is for reference only, quality of this particular scan is not point.
I’ve noticed something peculiar the past week. Initially after fusing, the mesh recommendation will display some reasonable quality level and grid size. After meshing, the scale (or whatever) of the quality slider changes.
Then it gets stranger. Refreshing the quality slider to new recommendation meshes at a similar polygon count as the previous recommendation although the numerical value of the slider is much lower for the same grid size.
The end result is that the maximum polygon count achievable tends to be around 5.9 quality regardless of the scan. The example is autoturntable, but laser scans seem to show a similar issue although I’m lazy to do an equivalent number of tests with the longer processing times of the laser scans.
EDIT: The cap seems to be based on grid size, not quality setting. All my tests were objects of similar size. In this example, grid size .05mm is the peak, followed by a few levels of declining polygon count and then finally a cap.
You can’t get better quality scan than the Grid that represents the points at the fusion distance.
If you fused at 0.05 mm and you set Grid at 0.04 mm you don’t get more just induced artificial points that was not existent before at all .
I would also not meshing over and over the same point cloud as it can downgrade the results for some reason.
I think it may be a bug I am observing since RS 4.2
The Fusing and Meshing settings will depends on the total of the volume of the object scanned prior to fusing.
So recommend to clean out the scan in Raw mode before fusing to increase the higher settings.
Understood, but the observation was about weird quality setting behavior. There shouldn’t be a point on the slider scale where 1 notch up gives you 1/2 to 1/3 the number of polygons. I’m not sure I can trust the grid size that it’s telling me.
Now that’s weird. I’ll see if I can reproduce that.
I need to test it out , so far it always worked for me perfect as long I did not meshes it more than one after fusing.
The GRID is based on the fusing ( point distance), so you have to get the exact values you fused it with to get it correct. Sliding up and down is no the way. There will be only one perfect value and nothing else reflecting the exact point cloud point distance.