CompuServe Thread

#BOOLEANS!

8 messages in this thread
#11053From: Paul UnterweiserMar 23, 1992 6:46 AM
For anyone that is on the fence deciding whether to buy/upgrade to 3D Studio R2, I found a reason that made it worth it's weight in gold! I needed to model a fairly complicated cellular phone, and played around with the Boolean functions, GREAT!!! By being able to add or subtract volumes from my model my modeling power just quadrupled, thanks Gary and ADESK! I did, run into an unusual quirk. I'm subtracting a series of cylinders from an extruded object (holes in the plastic for the speaker) and out of 20 odd cylinders, all but 4 worked…. They are identical (duplicated) cylinders and have identical placings (4 rows of 5 each). Very mysterious… I tried new cylinders in the same location, but no luck… any ideas? Paul Unterweiser, Unterweiser Studio-Animation
#11061From: Yost GroupMar 23, 1992 10:13 AM
Paul, when you tried to subtract 20 cylinders from one object, you were actually changing the "speaker" object TWENTY TIMES. After each subtraction, the object became more complex, with much more faces around the area you had previously subtracted. If you re-read the documentation about boolean modeling, you'll see that the density and configuration of the face structure around the area to be boolean'd is very important to pay attention to. By the 10th cylinder you'd subtracted, you made your speaker unbelievably complex. In this case (making the holes in a speaker (or speaker grill?), it sure sounds like you could've used the lofter tools to loft nested polygons to created the holes. That would be approximately 1000 times more efficient. It is certainly not the type of operation we designed booleans to optimally execute. – G
#11073From: Paul UnterweiserMar 23, 1992 7:09 PM
Gary, lofter wouldn't do it, and I wanted something more 'real' looking than bump/opacity maps. The surface I'm punching holes in is very curved and complex… The funny thing is, it wasn't the *last* 4 holes that didn't work! It has something to do with the location, not the complexity of the object… Paul Unterweiser, Unterweiser Studio-Animation
#11078From: Yost GroupMar 23, 1992 7:46 PM
Paul, if you knew what was going on inside the boolean code, you'd be amazed. They're really intended for sculpting operations in which you're carving a few pieces out of an object… not for massive penetration of arrays of objects into other objects. That's what the lofter is for (and if you can't figure out how to do it in the lofter, you should probably use an opacity map). – G
#11089From: buck crowleyMar 23, 1992 9:37 PM
Gary this is like the cube you made for me with the holes on each face. Could you have done that in lofter? TNx Buck
#11097From: Yost GroupMar 23, 1992 10:55 PM
Nope. I certainly couldn't have done the operation on the cube that I did for you with the lofter. I must be misunderstanding the problem, because when I look at my phone, it looks like an opacity map with a little box behind it for some dark color would make pretty decent holes for the mike and speaker. I just don't see the need for sides to the holes unless you actually want to fly through them into the speaker or microphone. – G
#11232From: Mike MorrisonMar 26, 1992 11:46 AM
"1000 times more efficient"?… wow, that's a lot…:-, -MM
#11239From: Yost GroupMar 26, 1992 12:21 PM
Well, when you're talking NxN (or N-squared) algorithms, the faces can start adding up BIG TIME extremely fast. – G