Point Cloud Display Problem

Hi, WIP displays point clouds incorrectly. What is happening is the foreground portions of the point cloud is more transparent than the background portions. So when you rotate the view, the rearward points are dominate instead of the front. Here is an example:

V5 in Wireframe mode

WIP in Wireframe mode

Please help

Hi Art4me - thanks, I’ll check this.
@Art4me - So far, I do not see this - I have a similar scan (must be the other party in the same crackup) and it displays correctly here. Do you see this with other point clouds?

-Pascal

Well it seems that the latest update which I installed right after this post solved the issue for now. Very strange!

Alright, but now, after leaving the viewport in one place for about 5min without doing anything, when I try to rotate the view either by using the R mouse button or the rotate view icon, the viewport just shutters, does not rotate. Acts as if it is locked. And it is in read-only mode.

Hi Art4me - what video card do you have, and are the drivers the latest available?

-Pascal

Quadro K2000, driver 382.05

Hi art4me - Thanks - I have the same driver, I’ll see if I can reproduce any of this - so far it’s been working as expected.

-Pascal

Ok, thanks. And FWIW, My system is a Dell Precision T1650, Windows 7, has current updates except this last Tuesday’s.

Pascal,
I get the point cloud display issue to repeat now. Meaning I get the rear points are now dominate instead of the front. I loaded a new project with a couple of point clouds. I went to Curve-Point Cloud-Remove Points. I then used the Lasso Select tool, selected some points of interest, hit enter, voila! The rear points become dominate. I have done this 3 times now, same results.
Chuck

Hi Chuck - I see that now - thanks for the blow by blow. Here the display sorts itself out once the command is complete - do you see that as well, or is the display stuck in reverse? How about if you move the view?
https://mcneel.myjetbrains.com/youtrack/issue/RH-39886

-Pascal

Well, when rotate the view before I hit return, the rear points are dominate but after I hit return, the display goes back to normal.

Now this is different than the 1st posting. I wasn’t doing anything other than rotating the view when the dominate changed or when the display just stuck.

RH-39886 is fixed in the latest WIP

Well thank you! But I found another issue that may or may not be part of this. As I described earlier with the flopping of the display, try this. Start a scene with a point cloud. Create a clipping plane in one of the viewports like the right viewport as an example and have it clip part of the point cloud. In my case, I have it pointing in the -x direction. While looking at the point cloud (PC) in the perspective view, look at the clipped point through the clipping plane. Then go to Curve-Point Cloud-Remove Points. Select the PC, with the normal marque selection tool, you can visually see the marque capturing the points of interest. Now try that with the Lasso Points, looking though the Clipping plane, you can’t see squat what it’s selecting. However, if you rotate the view while the Lasso Points command is still active, you see what it is trying to capture. It’s very odd.

Thanks, I’ve logged RH-40059.

I have another anomaly with point clouds that seem to be related to the reversal of dominance foreground/background point display. I loaded a E57 point cloud into WIP that has several PC’s that make up the whole. When you select one, it highlights briefly then displays the selection color and the remaining PC’s become transparent as you rotate the view and then we are back to the background points being dominant once more.

Would it be possible to share the model? I can’t reproduce this.

When I get back to the office on Wednesday, I’ll send you a link.

Here is a link. I will make the file available for about 2-3 hours.

Hi Art4Me - thanks, I see this - if a pc is selected the front to back order gets messy. It looks like the portions of the selected pc that are behind get pushed to the front - here it does not seem to depend upon there being multiple point clouds, I can see it with one of them.
https://mcneel.myjetbrains.com/youtrack/issue/RH-40233
-Pascal

Hi Pascal,

Well at least I now know it’s not my computer system that’s messing it up.

Chuck