oblong Posted February 15, 2012 Share Posted February 15, 2012 I have an example which shows rbd objects connecting constraints to each other, using apply relationships. When i change the rbdobject nodes (which have 4 objects set in each) to rbd fractured objects, which , when I look in the Dops details - appear to be almost exact (apart from objid difference) why does the apply relationship node no longer set up any constraints? As I say, the two object types appear identical in the Dop details list. There must be something fundamentally hidden from view which prevents this constraint loop.... Im simply trying to find a way to create a constraint for every object in the scene to connect to any other by using line segments as the information to apply relationships. This would be easy in SOPs. In Dops, it appears impossible ... I probably dont have the knowledge yet, but I must admit that there is simply no clue on how to resolve. apply_rel_constraints01.hipnc Quote Link to comment Share on other sites More sharing options...
rafaelfs Posted February 15, 2012 Share Posted February 15, 2012 I have an example which shows rbd objects connecting constraints to each other, using apply relationships. When i change the rbdobject nodes (which have 4 objects set in each) to rbd fractured objects, which , when I look in the Dops details - appear to be almost exact (apart from objid difference) why does the apply relationship node no longer set up any constraints? As I say, the two object types appear identical in the Dop details list. There must be something fundamentally hidden from view which prevents this constraint loop.... Im simply trying to find a way to create a constraint for every object in the scene to connect to any other by using line segments as the information to apply relationships. This would be easy in SOPs. In Dops, it appears impossible ... I probably dont have the knowledge yet, but I must admit that there is simply no clue on how to resolve. If you disable the "fracture by name" parameter in the RBD fracture objects it kinda works, with the exception that it places all the constrained points in the origin. The fractured obj doesn't calculate the individual positions of the objects until the simulation starts and that screws up the constraints in this setup you got. I'm looking into a way to properly place the anchorpoints. Cheers Quote Link to comment Share on other sites More sharing options...
oblong Posted February 15, 2012 Author Share Posted February 15, 2012 (edited) . Edited February 15, 2012 by oblong Quote Link to comment Share on other sites More sharing options...
oblong Posted February 15, 2012 Author Share Posted February 15, 2012 I made a set of points to represent the centroids of the objects and plugged those values in. Still doesnt fix the issue. I seem to have made a chain of constraints, which explode over a few frames. : attached a set up if you care to take a look. If I could get a handle on what variables are for counting through which objects, and how the stamp can be used to access these loop counts... but knowing Dops there are probably a fair number of hoops that i have to jump through before i can ever get this to work. Sending a set of line segments to Dops, to act as a rigging guide for auto constraints may now be a far off dream... is there a sesi programmer in the house!? apply_rel_constraints02.hipnc Quote Link to comment Share on other sites More sharing options...
anim Posted February 15, 2012 Share Posted February 15, 2012 if you want constraint between each box-ball of fractured objects then this is one possible solution first create pivot information according to your centroids, doing this on fracture object directly ensures correct ids then just reference pivot data as initial world space position of incoming objects. look at the file Quote Link to comment Share on other sites More sharing options...
anim Posted February 15, 2012 Share Posted February 15, 2012 second try apply_rel_constraints02_fix.hipnc Quote Link to comment Share on other sites More sharing options...
rafaelfs Posted February 15, 2012 Share Posted February 15, 2012 Yeah, I got to the same point and I start to believe that this would just not work the way we are trying. Even in conversations with a friend we had trouble figuring stuff out, so quite probably there's another way around this specific problem. It does make sense to me that what works on the left side of the switch certainly won't work for the right side. It'd be useful to figure out for something that I'm setting up myself too, but I guess for now I'll just keep it in the back of my mind. Cheers Quote Link to comment Share on other sites More sharing options...
oblong Posted February 15, 2012 Author Share Posted February 15, 2012 (edited) Yes - I have an intuition that there is yet another way involving use of Stamp with the apply relationship to do this, but its just not jumping out at me. Frustrating business , because I now have to resort to much more mundane non procedural methods involving opdigits(everything) etc ... and multiple seperate groups of objects for constraining individually. Pity there is no more info. The pivot idea was interesting to locate things, however chs("./object") didnt loop through anything and is always using the first location 0. Edited February 15, 2012 by oblong Quote Link to comment Share on other sites More sharing options...
Macha Posted February 15, 2012 Share Posted February 15, 2012 (edited) Something is odd here. If we switch to rbd fracture then the object id's are not what you expect. For example 1 and 6 is missing. Make sure they have correct objectid and I suspect it will work. Edited February 15, 2012 by Macha Quote Link to comment Share on other sites More sharing options...
anim Posted February 15, 2012 Share Posted February 15, 2012 isn't my file working for you? strange, it's deffinitely working here H11.1.22 Quote Link to comment Share on other sites More sharing options...
Macha Posted February 15, 2012 Share Posted February 15, 2012 Oh sorry Anim, I was referring to the original file. Quote Link to comment Share on other sites More sharing options...
oblong Posted February 16, 2012 Author Share Posted February 16, 2012 With object IDs that is what you get with Dops, and there is no changing those, also, notice the affectorids change to 0 2 3 4 6 etc instead of 0123456 etc Suspect something inside Fractured object that messes with this. Anim - check the hip you sent, it was trashed when i opened it with wrong rbdobject inputs. i will play with the contents of rbdFractured and see if the objids can be changed around ... Quote Link to comment Share on other sites More sharing options...
anim Posted February 16, 2012 Share Posted February 16, 2012 ... Anim - check the hip you sent, it was trashed when i opened it with wrong rbdobject inputs. ... I tried it with both versions I currently have 11.1.22 and 11.0.733, it's working fine sorry then, It must have been something changed since then, I will check when I get to newer build and post a new file Quote Link to comment Share on other sites More sharing options...
oblong Posted February 16, 2012 Author Share Posted February 16, 2012 (edited) . Edited February 16, 2012 by oblong Quote Link to comment Share on other sites More sharing options...
anim Posted February 16, 2012 Share Posted February 16, 2012 yes, they are connected, the only thing I can post right now is flipbook, I can modify the file as soon as I install new build to test for differences apply_rel_constraints02_fix.mov Quote Link to comment Share on other sites More sharing options...
anim Posted February 16, 2012 Share Posted February 16, 2012 (edited) I checked the file in newest build H11.1.206, it is still working, at least as in flipbook if that's the setup you are after are you sure you are using h11? Edited February 16, 2012 by anim Quote Link to comment Share on other sites More sharing options...
oblong Posted February 16, 2012 Author Share Posted February 16, 2012 (edited) I just re-checked ,yes: Totally broke in h12. Which is weird... I will get back to you when Ive compared the two and see if something can be learned from this exercise. Edited February 16, 2012 by oblong Quote Link to comment Share on other sites More sharing options...
anim Posted February 16, 2012 Share Posted February 16, 2012 I thought so, however it's easily fixable in H12, you can look at my post in alpha forums (about bug in RBD Fracture) there is the problem and solution written, or I can just send you H12 file Quote Link to comment Share on other sites More sharing options...
oblong Posted February 16, 2012 Author Share Posted February 16, 2012 (edited) Ok yes - see the fix now for this outstanding bug. Thanks for the information. Edited February 16, 2012 by oblong Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.