07-07-2021, 03:46 PM
(This post was last modified: 07-07-2021, 03:49 PM by josemendez.)
Hi!
Filo, without a doubt. The cost of simulation in Obi grows with rope length (as longer ropes tend to have more particles), in Filo it only grows with the amount of dynamic bodies involved in the cable system (in a typical lift, only 2: the pulley, and the lift platform.)
I see no reason to use anything more complex than Filo for a dark souls-style lift platform, that can only go up and down. The cable is never going to interact with anything, no need for self-collisions, no need to have frictional contacts, etc. You could even go as far as assuming the lift is not affected by the mass of whatever it is on it (a reasonable assumption, unless you want fatter characters to slow down the lift or something similar) and use no simulation whatsoever: just have a static cable mesh and manipulate its uv coordinates to match the platform height.
Filo already does this to a limited extent. Cables can dynamically split/merge when in contact with a CableBody, however this only works for very limited shapes (2D convex hulls, and discs). In the future we might support more complex shapes.
I assume you're referring to splitting/merging cable connections, this works fine in Filo but only against a very limited amount of shapes. Filo is an implementation of this technique, take a look at the research paper/video and it might give you insight on how it works:
https://matthias-research.github.io/page...Joints.pdf
https://www.youtube.com/watch?v=lGdBSGaiuAM
(02-07-2021, 05:27 PM)usemahrope88 Wrote: my question is, should obirope or filo be used for static lifts which simulate rope or chains pulling them? but would be used say, hundreds of times in a scenes life time
Filo, without a doubt. The cost of simulation in Obi grows with rope length (as longer ropes tend to have more particles), in Filo it only grows with the amount of dynamic bodies involved in the cable system (in a typical lift, only 2: the pulley, and the lift platform.)
(02-07-2021, 05:27 PM)usemahrope88 Wrote: also, would it be best to use filo on "cheap" lifts, which are semi-static but have the platform as physics tied to this filo rope?
I see no reason to use anything more complex than Filo for a dark souls-style lift platform, that can only go up and down. The cable is never going to interact with anything, no need for self-collisions, no need to have frictional contacts, etc. You could even go as far as assuming the lift is not affected by the mass of whatever it is on it (a reasonable assumption, unless you want fatter characters to slow down the lift or something similar) and use no simulation whatsoever: just have a static cable mesh and manipulate its uv coordinates to match the platform height.
(02-07-2021, 05:27 PM)usemahrope88 Wrote: also, have you considered making a filo type of asset which supports dynamically adding a cheap ray traced fast rope?
Filo already does this to a limited extent. Cables can dynamically split/merge when in contact with a CableBody, however this only works for very limited shapes (2D convex hulls, and discs). In the future we might support more complex shapes.
(02-07-2021, 05:27 PM)usemahrope88 Wrote: anyway this question is because filo is on sale right now, but i didnt realize i could make real use of it until just now, is it possible the attached bodies in filo can still be severed from the simulation if it leaves more than two connections, or i just change source code to do it? I read you cant add objects in runtime but i would want to be able to severe connections, im guessing the connections are stored as an array?
I assume you're referring to splitting/merging cable connections, this works fine in Filo but only against a very limited amount of shapes. Filo is an implementation of this technique, take a look at the research paper/video and it might give you insight on how it works:
https://matthias-research.github.io/page...Joints.pdf
https://www.youtube.com/watch?v=lGdBSGaiuAM