I am having the same problem. Bullet behaving weird with glue. This post seems to tackle the problem, by editing the bullet solver plugin: http://www.sidefx.com/index.php?option=com_forum&Itemid=172&page=viewtopic&t=25486&sid=3a56ab5c42932135a65e8bd03e457154 [sidefx.com]
I hope this will be fixed in a future build!
Found 14 posts.
Search results Show results as topic list.
Technical Discussion » Bullet and RBD Glue question
- csmnt
- 14 posts
- Offline
Technical Discussion » TwoState Constraint Relationship node not working correctly?
- csmnt
- 14 posts
- Offline
Hello!
In my scene I want to break a pin constraint when a certain force property value gets crossed. Unfortunately, the pin constraint doesn`t provide controls for that. So I took a look at the spring constraint, which does just that by using the “Two State Constraint Relationship” node - switching from a “spring relationship” to “no relationship” as soon as a force threshold gets passed.
So I rebuilt that using a “hard constraint relationship” node instead of the “spring relationship” node. And here is my problem:
When using the hard constraint node the force property is always set to zero, so it can never “break”. Is this a bug or am I missing the obvious? It definitly should work as this is from the docs of the “constraint relationship” node:
The quoted letting go never happens, only the “from spring to hard constraint” part works.
I have attached a simple example scene using the “constraint relationship” node to switch from hard to spring constraint and back with only the spring to hard part working.
Thanks!
In my scene I want to break a pin constraint when a certain force property value gets crossed. Unfortunately, the pin constraint doesn`t provide controls for that. So I took a look at the spring constraint, which does just that by using the “Two State Constraint Relationship” node - switching from a “spring relationship” to “no relationship” as soon as a force threshold gets passed.
So I rebuilt that using a “hard constraint relationship” node instead of the “spring relationship” node. And here is my problem:
When using the hard constraint node the force property is always set to zero, so it can never “break”. Is this a bug or am I missing the obvious? It definitly should work as this is from the docs of the “constraint relationship” node:
Two State Constraint
Acts as a hard constraint until certain forces are met, then acts like a spring constraint.
The constraint switches between a hard and a spring constraint based on the motion of the constrained object:
*
When the constraint is hard, if the “pull” of the object goes beyond the Two State Max Force value, the hard constraint will “let go” and turn into a spring constraint.
*
When the constraint is a spring, if the object gets within the Two State Min Distance of the goal, the spring will turn into a hard constraint.
Two State Max Force
Sets the force threshold for changing a two state constraint from the hard constraint state to the spring constraint state.
The quoted letting go never happens, only the “from spring to hard constraint” part works.
I have attached a simple example scene using the “constraint relationship” node to switch from hard to spring constraint and back with only the spring to hard part working.
Thanks!
Technical Discussion » displacing sphere type particles with rest position
- csmnt
- 14 posts
- Offline
Hi,
For my scene I need lots of individually noise displaced sphere particles. Instancing is not really an option as writing out the ifd takes ages. So I am using the sphere particle render type in pops, which works great as I can still displace them. The problem is though, that I dont want them to move through the noise but have it stick to them(and animate the noise later).
I tried setting up rest position and it also seems to work, bit it kills the noise and displaces the spheres uniformly, just adding to the pscale. My guess is this happens because sphere type particles only have just one primitive and one normal - it's just strange that applying displacement without rest position works.
I have added a simple example scene to illustrate the problem.
Is there a way to use rest position with particle sphere type?
Thanks!
For my scene I need lots of individually noise displaced sphere particles. Instancing is not really an option as writing out the ifd takes ages. So I am using the sphere particle render type in pops, which works great as I can still displace them. The problem is though, that I dont want them to move through the noise but have it stick to them(and animate the noise later).
I tried setting up rest position and it also seems to work, bit it kills the noise and displaces the spheres uniformly, just adding to the pscale. My guess is this happens because sphere type particles only have just one primitive and one normal - it's just strange that applying displacement without rest position works.
I have added a simple example scene to illustrate the problem.
Is there a way to use rest position with particle sphere type?
Thanks!
Technical Discussion » no viewport update using relative op: expression with cops
- csmnt
- 14 posts
- Offline
SYmek
op syntax works with absolute paths only, use opfullpath() for conversion:
op`opfullpath(“../../my/relative/path”)`
great, thanks!
..I think it would probably be good thing to have some info about this syntax in the houdini documentation. I didn't find anything there, though it seems to be used all the time.
Technical Discussion » no viewport update using relative op: expression with cops
- csmnt
- 14 posts
- Offline
Hi,
I am building a viewport reticle in cops to display frame number, camera info ect. for flipbooks. This is done by referencing a cop network on a constant shader which is applied to a grid that is parented to the camera. It all workes fine as long as the path to the cops OUT node is absolute. When using a relative path instead, the setup doesn't update anymore. This is a problem as I want to create a HDA from this setup.
I attached a simple example file including a simple cop2net with a font cop displaying $F. the absolute path updates the relative doesn't.
Thanks!
I am building a viewport reticle in cops to display frame number, camera info ect. for flipbooks. This is done by referencing a cop network on a constant shader which is applied to a grid that is parented to the camera. It all workes fine as long as the path to the cops OUT node is absolute. When using a relative path instead, the setup doesn't update anymore. This is a problem as I want to create a HDA from this setup.
I attached a simple example file including a simple cop2net with a font cop displaying $F. the absolute path updates the relative doesn't.
Thanks!
Technical Discussion » system("date /T") on windows not working
- csmnt
- 14 posts
- Offline
axebeak
I believe on Windows system() uses c-shell, not the native command processor, and this causes problems with built-in commands such as ‘date’. You can try the following expression (it works for me):
`system(“cmd /C date /T”)`
here too, thanks a lot!
Technical Discussion » system("date /T") on windows not working
- csmnt
- 14 posts
- Offline
Technical Discussion » system("date /T") on windows not working
- csmnt
- 14 posts
- Offline
Hi,
I have a font operator which I would like to display the current date. When I use `system (“date /T”)` I get no result. `system (“echo 123”)` is working though and so is `system (“ipconfig”)`, ect.
Is this a bug with houdini not understanding the “date” command or am I missing something?
Thanks!
I have a font operator which I would like to display the current date. When I use `system (“date /T”)` I get no result. `system (“echo 123”)` is working though and so is `system (“ipconfig”)`, ect.
Is this a bug with houdini not understanding the “date” command or am I missing something?
Thanks!
Technical Discussion » transfering color from particle to nearest particles
- csmnt
- 14 posts
- Offline
off topic: but by playing around with the two pop net solution something interresting happened. by just transfering the colors from one to another particle system this voronoi patterns emerged, wich makes total sense but I still think its cool
anyway…I still didn't manage to get the color transfer working within one pop net.
anyway…I still didn't manage to get the color transfer working within one pop net.
Technical Discussion » transfering color from particle to nearest particles
- csmnt
- 14 posts
- Offline
so here is a version using two pop nets and the attribute transfer pop.
it works, but I would really like to have it working in one single pop net.
it works, but I would really like to have it working in one single pop net.
Technical Discussion » transfering color from particle to nearest particles
- csmnt
- 14 posts
- Offline
this should be fairly simple and I know how to do this with geometry but from paticles to particles it wont work:
I am sourcing particles and puting every fifth in a group called “everyFifth”. those particles get colored randomly. now I want them to transfer their individual color to the nearest other particles placed in a rest group.
when using the proximity node I get a lot of groups ending with the ID of the particles in everyFifth (proximity5, proximity10,…) that contain the nearest paricles to each everyFifth particle. but now I would need something like a for-loop for those, using the number in the string of the groupname wich I am not sure how to do - also I think there must be a simpler solution.
the poppointid expression should work for getting the color of the individual particles in the everyFifth group but then I still don't know how to loop through every group.
I also tried to get this working using the attribute transfer pop, directing it to the popnet as the sop but as expected this resulted in an error. I dont want to use 2 different pop networks wich I think would work using the attribute transfer pop.
thanks.
I am sourcing particles and puting every fifth in a group called “everyFifth”. those particles get colored randomly. now I want them to transfer their individual color to the nearest other particles placed in a rest group.
when using the proximity node I get a lot of groups ending with the ID of the particles in everyFifth (proximity5, proximity10,…) that contain the nearest paricles to each everyFifth particle. but now I would need something like a for-loop for those, using the number in the string of the groupname wich I am not sure how to do - also I think there must be a simpler solution.
the poppointid expression should work for getting the color of the individual particles in the everyFifth group but then I still don't know how to loop through every group.
I also tried to get this working using the attribute transfer pop, directing it to the popnet as the sop but as expected this resulted in an error. I dont want to use 2 different pop networks wich I think would work using the attribute transfer pop.
thanks.
Edited by - 2011年10月29日 14:17:39
Technical Discussion » animate noise time
- csmnt
- 14 posts
- Offline
thank you! .. this is exactly what i was looking for. by using the lenght of the antialiased noise with 4d input as displacement along normal i now get noise that looks almost like the perlin noise on the turbulentNoise node - but in an animateable version. very cool!
the thing where i am still stuck at, is getting the look of the aligator or sparse convolution noise in an animatable version. on the curlnoise node those types don't have any effect when using a 4d input so i tried using the worley and voronoi noise as the help mentions that they are very similar to aligator and sparse convolution noise.
the look i get is very different though and there are not many settings for further adjustment avaliable. any ideas?
i have attached an example file with animated perlin style noise and the look i am after when using turbulentnoise set to aligator type.
the thing where i am still stuck at, is getting the look of the aligator or sparse convolution noise in an animatable version. on the curlnoise node those types don't have any effect when using a 4d input so i tried using the worley and voronoi noise as the help mentions that they are very similar to aligator and sparse convolution noise.
the look i get is very different though and there are not many settings for further adjustment avaliable. any ideas?
i have attached an example file with animated perlin style noise and the look i am after when using turbulentnoise set to aligator type.
Technical Discussion » animate noise time
- csmnt
- 14 posts
- Offline
thanks for the reply. ..well i was actually doing excact what you described, the problem is that i dont want to see a direction in the noise as what happens when animating the offset.
but the flownoise node seems to give the result i need. the only drawback when using flownoise ist that i can not choose between different noise types like spatial, aligator, perlin.
plugging the noise vector from the flownoise in the position vector of the turbulent noise seems to work though. but is that the correct workflow for getting animatable, say, alligator noise?
but the flownoise node seems to give the result i need. the only drawback when using flownoise ist that i can not choose between different noise types like spatial, aligator, perlin.
plugging the noise vector from the flownoise in the position vector of the turbulent noise seems to work though. but is that the correct workflow for getting animatable, say, alligator noise?
Technical Discussion » animate noise time
- csmnt
- 14 posts
- Offline
hi,
i am a maya user trying to make the switch over to houdini. while using maya i got used to having a time attribute on noise textures to animate variance in noise without seeing a directional motion as when just animating the offset. i tried plugging another noise in the offset of my main turbulent noise but without much success(i get a noisy back and forth motion).
i am using it on a sphere in a vop sop just displacing the geometry.
id really appriciate it, if anyone could explain to me how to animate noise in houdini.
thanks
i am a maya user trying to make the switch over to houdini. while using maya i got used to having a time attribute on noise textures to animate variance in noise without seeing a directional motion as when just animating the offset. i tried plugging another noise in the offset of my main turbulent noise but without much success(i get a noisy back and forth motion).
i am using it on a sphere in a vop sop just displacing the geometry.
id really appriciate it, if anyone could explain to me how to animate noise in houdini.
thanks
-
- Quick Links