Hi all,

as far as I have understood, at a weld the welded vertex will receive the position of its weld goal (usually the corresponding vertex at the parent body part).

Are all poser versions doing this, or are poser versions known that use another formular for the welded vertices?

Thanks and best regards,

col

Attached Link: Weld tolerance in Poser figures

I did some tests on welding - I'm not sure if they'll be helpful? These were with Poser 6, by the way.

Hi Bob, thanks I think les has mentioned this one to me. Have you been moving the vertices in the .OBJ to see when Poser would stop recognizing them as a weld? Or have you modified the figure within Poser afterwards (with a magnet, morph or whatever and managed to destroy the weld afterwards?)

I moved the vertices in the original geometry file. You raised an interesting question though - does morphing or magnetising affect the weld differently? That would depend on the order in which the calcualtions are done. As far as I can recall from experience without devising specific tests, once vertices are welded they stay welded.

I did have a strange experience with joints splitting apart under some circumstances even though the vertices were perfectly aligned, and I never got to the bottom of that except that it seems to have been a bug that was introduced in P6. It may well have been fixed in later Poser versions, I never repeated the test.

http://market.renderosity.com/mod/forumpro/showthread.php?thread_id=2301883

http://poserpros.daz3d.com/forums/viewtopic.php?t=46304

I think you may have to already be a member to read the PoserPros forums: they're frozen now, but still a useful source of information.

Yes, these are all interesting threads. Appearently those welds have always been an interesting place :).

What I am looking for is, I think, a sort "official" piece of text that tells how the (active) welds (lets suppose Poser recognized them well) determine the vertices position. if its part of the rules that the parent dominates.

I think its this way but I would love a confirmation.

for my square test figure it seems to be that way. i am not making much progress with my support request at poser support :/. but it seemed to be just the same in daz studio. are there other programs that can read .cr2 that I could try?

Attached Link: http://cgscripts.colorcurvature.com/square.gif

animated gif :)

what i found most odd is that making the parent actor invisible breaks the weld

I'm not sure you can draw meaningful conclusions from a test figure with so few polygons - but I think you're probably right. I forgot to point this out earlier, but if you look at the first two pictures in my test thread, linked above, it seems to me that the lower square (which is the parent) retains its edge vertices in the same position, whereas the child square becomes more distorted as the weld distance is increased. This would bear out your theory that the parent has priority, as it were.

Quote - what i found most odd is that making the parent actor invisible breaks the weld

I was surprised but not astonished. I already had an idea there are some areas where making geometry invisible removes its influence; and others where it doesn't. If I think about it I may be able to come up with a list, but it's late here. :)