Hi Ari, so one issue that I didn't notice before is that each of the tiles and the board also are cut out from a way way larger sphere surface.

One side effect of that is the "quick calculated bounding box" of each little tile piece is really the same size as the board, and that means if you try to union them all at once the sorting mechanism that is based on size won't really be able to gather together the large and small pieces one by one which is the best way for it to be processed.

So probably running ShrinkTrimmedSrf (http://moi3d.com/2.0/docs/moi_command_reference10.htm#shrinktrimmedsrf) on these things to get their underlying surfaces to just be the active trimmed areas, or unioning things more one at a time instead of in a big batch would probably help getting the tiles and the board combined to start with. Also it probably wouldn't hurt if there was just a little bit more space between the corners of each time, they are pretty close to running into each other and if they do get close enough to where the boolean tries to glue those 2 edges together that will basically make non-manifold object that has one edge that is trying to have 4 different surfaces joined to it.

- Michael

Hi Ari, so also another complicating factor is some of the leg structure is poking up just by a very tiny amount up past the top surface of the board, making nearly a kind of "single point" tip contact from some bottom pieces and the top surface. Here I've made a high resolution mesh and you can see some little tiny leg spikes sticking up through the top:



So anyway that's another complicating factor, having a pointy object rest its point right on the inside of the top surface of another solid will make for a difficult to handle boolean.


By doing ShrinkTrimmedSrf it was then possible to do one big group union for getting the board and the tiles togehter, now to get the legs on I'll probably first try to make a median surface going down the middle of the board (by offsetting the bottom up), and cutting the legs with that first to cut away these areas that are just barely poking through the top.

- Michael

Hi Ari, I've attached here a finished up solid version of the chessboard, I think it's got all the things combined ok now.

The main steps I did was that I first focused on getting the board and the tiles combined, when I examined their surfaces I saw that they originally came from a big sphere surface that was many many times larger than the current active trimmed area - that will make some various complications with such a discrepancy in size since some things like intersection tolerance will be based on the overall size of the underlying surfaces.

So running ShrinkTrimmedSrf on those pieces solved that problem - after doing that it was then possible to just select the board and all the times all at once and do a boolean union.

Then like I wrote above I also noticed that some of the legs were poking little pointy parts up either right on or just slightly above the top surface, you don't want to have that kind of thing with a sharp point barely touching some other piece. So to solve that I did an offset of the bottom surface of the board halfway up to construct a surface going halfway through the board. I then selected the legs and did boolean difference with that median surface as the cutting object. That took a couple of minutes to calculate and then it sliced off the tops of those leg pieces that were sticking up too much, then I deleted those. Then after that a boolean union with the board worked ok.

- Michael

Hello Michael,

thanks a million :) both for the model and guidance - must learn how to use the ShrinkTrimmedSrf.
BTW Should ScaleArray work ok in MAC as well (so for no luck with that)?

br Ari

Hi Ari, you're welcome! re: ScaleArray - I think that should be working on the Mac too. But in order for it to work you need to include a point object (Draw curve > More > Point) along with the arrayed objects, it then uses that point as the center of the scale. Is it possible that your array items did not include a point object like that?

- Michael

Hello Michael,

I have the point object ok - there is an error message:

ReferenceError: Can't find variable: ScaleArray
line 1
1: >> ScaleArray

I have created a 'commands'-folder and added the ScaleArray.js and .htm there - is there something else I should do?

br Ari

Hi Ari - you shouldn't have to create a commands folder, there should already be one there with all the regular commands (like Line, Extrude, Sweep, etc...) all in there already, you need to put those ScaleArray files into the same spot as all the regular commands.

So I think you probably don't have it in the right folder - when MoI tries to execute a command and it doesn't find any files of that name in the commands folder, it tries to execute the text as inline script code, that's what generates that particular error message that you're getting.

To get to the right commands folder on the Mac you need to right-click on the app and pick "Show package contents", then inside there go to /drive_c/moi/commands - that's where you'll need to copy the files into. Once you do that then you should be going ok.

- Michael

Also make a shortcut. http://moi3d.com/forum/index.php?webtag=MOI&msg=275.109

Hello Michael,

and so it was :) Thanks again - now I'll stop bothering you.

br Ari

Hi Ari, well definitely the plug-in installation mechanism is pretty crude right now, particularly on OSX where you have to do that right-click and go down in there a ways.

In the future I want to totally overhaul that whole mechanism and have something like some built in UI to browse a repository of plug-ins. There is quite a bit of work involved in that though, and since you only have to do the clumsy copy step just one time and not on a daily basis or anything it just hasn't risen to a high priority yet.

- Michael