I'm running Microstation V8i corresponding to Power Geopak 08.11.09.878. I'm plotting to a PDF from a DGN that has many attachments--some displayed, some not. When I attempt to plot with all the attachments I want to display, the rasterized option is locked in and greyed out.
I have noticed this issue before, and because I don't know better, I had associated it with 3d models. I have two 3d attachments in this situation: attachments A and B. In this particular case, I need to detach only attachment A in order to correct this rasterized option lock. (It isn't enough to just turn off display on that attachment.) Attachment B can remain attached and displayed, and I can still disable Rasterized assuming attachment A is detached.
I've determined that it has nothing to do with presentation settings (i.e. Wireframe, Illustration, etc.). It isn't Plot as 3D (PDF).
Attachment A has a point cloud in it. Attachment B has linear, mesh, and I believe volume features.
Any advice on this? Where do I look for the correction?
It's the point cloud. Point clouds can only be printed in rasterized mode.
There was an issue in the past where point clouds in references whose display were turned off would still trigger the rasterized requirement, but that's been corrected in CONNECT Edition.
.
Answer Verified By: Derek Schmidt
Thanks. Knowing this now, could it make sense to have Microstation simply not print the point cloud if I don't want to print rasterized? This would be preferable to entirely detaching a point cloud model.
And as it stands, I can have a single point element (which results from any two-vertex line where all but one vertex is deleted) that isn't in a point cloud and still print non-rasterized. Is there an important distinction between these?
Like raster attachments, you can control the display of a point cloud on a per-view basis via the point cloud properties dialog. You can work in one view, seeing the point cloud, and print from another where the point cloud is not visible.
Vector points, e.g. single-vertex line elements, and point clouds are drastically different as far as implementation is concerned. But the basic problem is the amount of data generated at printer resolution. A hypothetical design file containing millions of single-vertex linestrings would not print successfully in non-rasterized mode either.
I do see that if I turn off display of all point clouds in view 1, I can print from view 1 without rasterized locked in. This helps considerably. Thanks. Though until I have a good set of principles lined up, I will probably find it at least a little bit painful to have to turn off point clouds on a per-view basis, since they're usually in a different file than the one I have open.
I can understand that a whole mass of single-vertex linestrings may not print satisfactorily, but I also feel like that doesn't really justify forcing an option that I may not want to be active. I'd rather determine that for myself. And further, perhaps it could be a view attributes and print attributes option whether or not to view or print point clouds.
Anything but the most trivial point cloud would likely crash the program were its generated data converted to vector elements and sent through the non-rasterized printing pipeline. So it's not an issue of printing satisfactorily, but printing at all. MicroStation allows the Rasterized checkbox to be toggled off in many situations where it's questionable the desired output will be achieved, but avoiding a near-certain hard crash is a different matter.
Investigating, I see that MicroStation does not take into account the point cloud level's "do not plot" flag when determining whether Rasterized should be enforced. It should, and that's essentially the same functionality as a non-view-dependent point cloud print attribute. That's something we'll fix in the next CONNECT Edition update.