2018 Booleans - Lots of problems.

[revel68]

Is anybody out there using the new booleans?

They seem highly problematic across all kinds of stuff i'm trying to create today.   I have extensive max experience so I know how to try different things to try to get it to work. Everything i'm doing is just brutal and broken.  I even sometimes get Invalid UV errors (that are unfixable even with all the little uvmap/edit poly tricks) on some things I create. This error comes up during rendering.

I would like to submit bugs but it just seems like the entire thing is bugged. 

Right now I'm trying to take 2 very basic legit models and Union them and it refused to.  All of the operations are not acting properly but I don't know how to make it any simpler of a case.  2 Oil tanks edited and collapsed and reset transform and they won't union properly. And i'm only doing this because when i do both of them as Subtract in layers on my base object one of them refuses to actually subtract.  So i'm trying to merge and then subtract.  

This seems unworkable. Am I the only one?

[hagen.deloss]

@Anonymous 

Thanks for posting about your issues with the boolean tool!

I would really like to help you figure out a way to get better boolean results, do you have sample file that you are having issues with that folks in this thread can test with? 

I used to work in game asset creation, so most of my experience with the boolean tool is with meshes that are a lot less dense then the examples above. What about the 2016 boolean tool works better then the 2019 version?

In regards to the Material ID issue you mentioned, are you experiencing this with both the boolean and Proboolean tools?  Are you using a boolean on a mesh that is "water tight"? (meaning the mesh has no border edges). I will test this out to see if I can log it in our internal database. 

If you have more specific issue to report, I would like to test them and maybe connect you to someone on the Max modeling team so you can pass along more pointed feedback directly 😄

Looking forward to speaking with you soon.

[Anonymous]

I tried posting sample files and a screencap yesterday but do not see the post.  I will try again and attach below this but I went back and confirmed that a boolean operation that did not work in 2018 did in fact work in 2016.  The main issue is that upon applying the boolean modifier (before even adding an operand) it changes the material id's on the object.  Previously in 2016 nothing happened to the object before the actual boolean operation and upon execution it would have a popup with 5 different material attach options; one of which was *do not modify mat id's or material* which would leave the assigned id's as is.

The attachment has a max 2016 version and a max 2018 version of a before and after of the boolean subtraction and a screencap of the dialogue prompt from 2016.  Proboolean in the 2018 version made the entire object disappear and deleted most of the object in 2016.  However the rule of thumb in the past was that if one didn't work the other typically would.  Now neither do and altering mat id's makes it mostly useless for my purposes.

[Anonymous]

Enclosed sample.

[hagen.deloss]

Thanks for the example file @Anonymous 

I will log this as an issue in our internal database and let you know what the development team says regarding it, as I am not sure this is by design, or a missing UI prompt.

I appreciate you posting this on the forums!

[hagen.deloss]

@Anonymous 

I'm curious, do you find this to be an issue if you first apply an Unwrap UV modifier to the object before applying the boolean? The example you provided seemed to have very large uvs by default.

[Anonymous]

I typically collapse the object before performing any boolean operation.  In the above example there likely was a simple UVW box mapping modifier used although all textures were removed in the file supplied and a simple multisub was used with different diffuse colors which represented the mat id's. 

The scale used is typical for architectural mapping where we use large scales to map large objects.  Using a small mapping scale would lead to visible tiling on large surfaces so we tend to use larger textures to reduce the any visible tiling in the texture.  This file was a piece of sidewalk that had to be cut for the base of a revolving door. 

I generally will not spend the time to use an unwrap modifier as they are time consuming and only needed in certain situations.  That being said how a map is scaled is not relevant to the assignment of material id's and the behavior I have documented.

[hagen.deloss]

Hi @Anonymous 

I am currently testing this to see how the boolean operation in 3ds Max 2020 handles a multi-sub object materials ID's, and i want to make sure I'm correct in saying that it behaves the way you would want. Here is a gif demonstrating this.

I have applied a multi-material with 2 material ID's to an object. I then apply a boolean operation to the mesh, and select the 2 other objects, the operation does not seem to over ride, or change the material ID's...is this the behavior you would expect? I realize there is no tooltip that pops up, asking about inheriting those materials...are you looking for the booleaned meshes to not inherit the material at all?

I want to make sure I fully understand the issue you are having so i can properly pass along the information to the dev team. (I have already made a ticket for the tooltip not appearing, I haven't heard anything back about whether that is an intended design choice though).

Talk to you soon 😄

[Anonymous]

That is correct.  By default or by a popup there should be a choice to not modify materials or material id's.  The screencapped popup I provided had 5 choices with different options but that was the one I generally used.  However we are not on 2020 yet and won't be for awhile.  This is an issue occurring in 2018 and perhaps 2019; though once again we are still on 2018 so a remedy for 2018 is what I am looking for.

[hagen.deloss]

@Anonymous 

I just wanted to keep you updated in this issue. Talking with the development team, it looks like this is possibly by design. Within Max 2019 and subsequent releases, there have been some additional work done on boolean operands including improved boolean scripting, UV functionality, and general performance fixes. Typically the development teams focus is on the current version of Max and one version back.

If you would like to suggest bringing back this material ID pop-up, you may want to post in the 3ds Max Ideas page, I'm sure other folks feel the same and would vote this being an issue, this is honestly the best way to make the dev team aware of this issue!

[Anonymous]

The lack of a popup might be by design but the lack of either a checkbox or default behavior that retains material id's is a screw up on the *design* teams part due to their lack of understanding on how booleans are used.  I am not surprised it would not be addressed nor honestly did I think anything would be done but after going through this lengthy exchange to end with *its by design* I won't be doing your development/qa teams job for them anymore.  Its simply a waste of my time on a group of people that do not care about making the product work but rather hitting deadlines for the next upcoming release. 

[smoore]

I've been working with booleans since the 1990s. I have yet to find a piece of software that does good job with booleans. In 3dsmax you got 'Boolean' and 'ProBoolean' - if you are lucky you can get one of them to work. 'ProBoolean' tends to crash while 'Boolean' doesn't crash, but gives wonky results.

I came here being fed up with booleans as usual and looking for another tool that might be better but it doesn't look like there is one.

So - a few tips:

Don't use booleans if you don't have to.

Booleans work fairly well with simple geometry.

Sometimes booleans fail when planes are co-planer - move one object by a very small amount and try again.

The proboolean option to remove edges right at the bottom - change that and it might fix things.

Try re-ordering your operands.

STL check on your operands.

Pray it works.