Found 117 posts.
Search results Show results as topic list.
Work in Progress » ocean tool
- RudiNieuwenhuis
- 190 posts
- Offline
Technical Discussion » Python dopselect
- RudiNieuwenhuis
- 190 posts
- Offline
Hi everyone.
I'm trying to make a shelf tool to previsualise the breaking of a dop-object that has been set to be breakable.
I started out with the sop-part, which was rather easy. Just create a scatter, a voronoifracturepoints, a voronoifracture and a explodedview and automaticelly wire them in the right way and set a couple of parms on them to get a preview.
But that's just part 1.
Now of course i could add another shelf tool where i ask the use to select the sourceNode (in sops) and the targetNode (the fractureparms node in dops) and make all the relevant parameters relative references from source to target.
But this must be possible to do in just one single step.
So what i need now is to be able to select the RBD-object similar to anything in sops with toolutils.sceneViewer().selectedObjects()
When i have this RBD-object, i should be able to retrieve the node that is in the SopPath parameter and the fractureparms node is directly beneath it so there should be a way to get a hold of that one as well.
Come to think of it. I can't be 100% sure the next node is the fractureparms node so if i might have to go down the branch and select the first node of the “Voronoi Fracture Configure Object” type.
How would i check for this type?
As you can see, i could use some help. I promise to post the code here once i've got i all figured out though. :-)
I'm trying to make a shelf tool to previsualise the breaking of a dop-object that has been set to be breakable.
I started out with the sop-part, which was rather easy. Just create a scatter, a voronoifracturepoints, a voronoifracture and a explodedview and automaticelly wire them in the right way and set a couple of parms on them to get a preview.
But that's just part 1.
Now of course i could add another shelf tool where i ask the use to select the sourceNode (in sops) and the targetNode (the fractureparms node in dops) and make all the relevant parameters relative references from source to target.
But this must be possible to do in just one single step.
So what i need now is to be able to select the RBD-object similar to anything in sops with toolutils.sceneViewer().selectedObjects()
When i have this RBD-object, i should be able to retrieve the node that is in the SopPath parameter and the fractureparms node is directly beneath it so there should be a way to get a hold of that one as well.
Come to think of it. I can't be 100% sure the next node is the fractureparms node so if i might have to go down the branch and select the first node of the “Voronoi Fracture Configure Object” type.
How would i check for this type?
As you can see, i could use some help. I promise to post the code here once i've got i all figured out though. :-)
Houdini Lounge » suggest good tutorial
- RudiNieuwenhuis
- 190 posts
- Offline
It's actually quite easy, you won't need a full tutorial for that.
Take a look at the file i attached.
The box_object1 contains both the lowres and the highres geometry.
In the AutoDopNetwork the RBD-object node has its SOP Path set to an expression by default that points to the input for the dopimport node in your geo. If you change this expression and just point to the lowres version in your geo, it should work just fine.
Hope this helps.
Take a look at the file i attached.
The box_object1 contains both the lowres and the highres geometry.
In the AutoDopNetwork the RBD-object node has its SOP Path set to an expression by default that points to the input for the dopimport node in your geo. If you change this expression and just point to the lowres version in your geo, it should work just fine.
Hope this helps.
Technical Discussion » Modulate Cut Plane Offset in Voronoifracture Node?
- RudiNieuwenhuis
- 190 posts
- Offline
It would be horrible inefficient i guess but you could have a go at it like this:
First “allow editing of contents” for the voronoifracture node.
There is a NULL called INPUT_CELL_POINTS that defines the amount of chunks you'll get. Now from here all the way up to the “switch_create_groups” you would have to copy this and put this in a for each loop.
Then inside the foreach delete all the chunks except the one that corresponds to your for-value.
Now you have the “cut_each_piece” node (which is the actual voronoi split) inside the for each loop. In here you'd have to calculate the distance between your null and the point that corresponds to your for-value and use that to drive the Offset in the Voronoi Split.
It's a bit of a hassle but i guess it must be possible.
First “allow editing of contents” for the voronoifracture node.
There is a NULL called INPUT_CELL_POINTS that defines the amount of chunks you'll get. Now from here all the way up to the “switch_create_groups” you would have to copy this and put this in a for each loop.
Then inside the foreach delete all the chunks except the one that corresponds to your for-value.
Now you have the “cut_each_piece” node (which is the actual voronoi split) inside the for each loop. In here you'd have to calculate the distance between your null and the point that corresponds to your for-value and use that to drive the Offset in the Voronoi Split.
It's a bit of a hassle but i guess it must be possible.
Technical Discussion » keyframes in channel editor gone
- RudiNieuwenhuis
- 190 posts
- Offline
I found the solution, deleting all the preference-files in libraries/documents/houdini11.0/ did the trick.
Technical Discussion » keyframes in channel editor gone
- RudiNieuwenhuis
- 190 posts
- Offline
I've got a weird little problem that started to occur without any apparent reason: All the keyframes in my channel editor are invisible. They are there, i can see them in the timeline but not in the channel editor.
I'm most likely just overlooking something very simple but I just can't find it.
I already tried to restart and reinstall but that didn't help.
I'm most likely just overlooking something very simple but I just can't find it.
I already tried to restart and reinstall but that didn't help.
Technical Discussion » Wheel Rotate based on distance
- RudiNieuwenhuis
- 190 posts
- Offline
I have found a solution to the “total distance” problem using a Python script in chops that is very well explained by Steve Twist on 3DBuzz's Technical Rigging V.
Technical Discussion » Wheel Rotate based on distance
- RudiNieuwenhuis
- 190 posts
- Offline
Hi all,
I have searched extensively to find a “car-rig”-like solution to make wheels rotate based on the distance they have traveled.
I came across: Lyonz thread: (http://forums.odforce.net/index.php?/topic/13258-direction-spin [forums.odforce.net]) but that didn't came up with a solution.
I found tyson ibele's elegant solution in 3Dmax over here: http://www.simplycg.net/viewtopic.php?f=32&t=4630 [simplycg.net]
So i started to try to get a simple version of that in Houdini.
For this i want to calculate the total distance traveled and here comes the first trouble. I calculated the distance traveled in the last frame.
I created a null that does follows a path, and in chops offset it to 1/$FPS. Then subtract the second from the first and you've got the distance traveled in the last frame.
So my idea was: get the rotation of the wheel and add (distance * (360/2*pi*wheelradius)
But this isn't working because in sops += doesn't work as explained here:
http://www.sidefx.com/index.php?option=com_forum&Itemid=172&page=viewtopic&t=20063 [sidefx.com]
Alright, so i thought i'dd get the totaldistance covered so i brought it into dops and use a sop-solver, similar to how Peter Quint explains here:
http://vimeo.com/6663811 [vimeo.com] but i'm getting into trouble again. I CAN get the distances calculated in a vopsop in the sopsolver in dopnetwork but i just can't get them properly added to the already existing distance.
Needless to say, I'm starting to feel a bit stupid and i could really use some help. Maybe someone could shed some light on this?
Thanks in advance.
Rudi
I have searched extensively to find a “car-rig”-like solution to make wheels rotate based on the distance they have traveled.
I came across: Lyonz thread: (http://forums.odforce.net/index.php?/topic/13258-direction-spin [forums.odforce.net]) but that didn't came up with a solution.
I found tyson ibele's elegant solution in 3Dmax over here: http://www.simplycg.net/viewtopic.php?f=32&t=4630 [simplycg.net]
So i started to try to get a simple version of that in Houdini.
For this i want to calculate the total distance traveled and here comes the first trouble. I calculated the distance traveled in the last frame.
I created a null that does follows a path, and in chops offset it to 1/$FPS. Then subtract the second from the first and you've got the distance traveled in the last frame.
So my idea was: get the rotation of the wheel and add (distance * (360/2*pi*wheelradius)
But this isn't working because in sops += doesn't work as explained here:
http://www.sidefx.com/index.php?option=com_forum&Itemid=172&page=viewtopic&t=20063 [sidefx.com]
Alright, so i thought i'dd get the totaldistance covered so i brought it into dops and use a sop-solver, similar to how Peter Quint explains here:
http://vimeo.com/6663811 [vimeo.com] but i'm getting into trouble again. I CAN get the distances calculated in a vopsop in the sopsolver in dopnetwork but i just can't get them properly added to the already existing distance.
Needless to say, I'm starting to feel a bit stupid and i could really use some help. Maybe someone could shed some light on this?
Thanks in advance.
Rudi
Technical Discussion » workflow hdri environment map
- RudiNieuwenhuis
- 190 posts
- Offline
sanostol
put the gamma 2.2 in the viewer (mplay or renderview)
the output of the renderer is linear and it is better feed the renderer with linear data and correct the outputimage after rendering
That would be a good solution if it weren't for all my other stuff to get wrong. If i just put a constant shader on an object right next to it, setting a gamma for the overall image to 2.2 messes up that constant shader.
Technical Discussion » workflow hdri environment map
- RudiNieuwenhuis
- 190 posts
- Offline
I'm trying to make a hdri environment map but i'm having some trouble with the gamma or exposure.
Here is what i'm doing:
* take 3 bracketed photos (-2EV, 0, +2EV)
* in photoshop: automate Merge to HDR Pro to get a 32bit .hdr file
* in HDRShop do the panoramic transformation to get a vertical cross.
* with isixpack convert this to a .rat file
* use this .rat file as an environment in an environment light with render mode set to raytracing background and light the scene seperately with another environment light with this .rat file but render mode set to direct light.
* put a mantra surface on a sphere and set the specular intensity on the Reflect Lights tab to 1 and the specular angle to 0.2 to get near perfect reflections.
The resulting image is way to dark, only if i bump up the gamma or the lightintensitiy of the environment light do i get reasonable results.
Where should i change my workflow to just use a gamma of 1 and a light intensity of 1.
Here is what i'm doing:
* take 3 bracketed photos (-2EV, 0, +2EV)
* in photoshop: automate Merge to HDR Pro to get a 32bit .hdr file
* in HDRShop do the panoramic transformation to get a vertical cross.
* with isixpack convert this to a .rat file
* use this .rat file as an environment in an environment light with render mode set to raytracing background and light the scene seperately with another environment light with this .rat file but render mode set to direct light.
* put a mantra surface on a sphere and set the specular intensity on the Reflect Lights tab to 1 and the specular angle to 0.2 to get near perfect reflections.
The resulting image is way to dark, only if i bump up the gamma or the lightintensitiy of the environment light do i get reasonable results.
Where should i change my workflow to just use a gamma of 1 and a light intensity of 1.
Technical Discussion » Car Paint with image planes
- RudiNieuwenhuis
- 190 posts
- Offline
circusmonkey
Hi the code is fully available look at the tab titled code for the otl !
r
The code is available, but i don't understand that, i only know how to do this in the Vex builder.
Technical Discussion » Car Paint with image planes
- RudiNieuwenhuis
- 190 posts
- Offline
circusmonkey
Ive not seen this shader but what about drop a default shader , RMB on the vop shaders output and view the code . Look how the exports have been written. Then try and apply to the car paint shader.
Thanks Rob, i know how to export the different variables for use in image planes, that part is not a problem. The problem is the car paint otl is closed, and i can not enter in it. So i'dd have to make one myself and that's quite a daunting task. (for the moment)
I did find a metalic shader thread by JR Gauthier that's helpfull: http://www.sidefx.com/index.php?option=com_forum&Itemid=172&page=viewtopic&t=17846&highlight=metallic [sidefx.com]
His shader is not closed so I think i will take a better look at how he did his, that will be helpfull i think.
Technical Discussion » Car Paint with image planes
- RudiNieuwenhuis
- 190 posts
- Offline
Technical Discussion » Car Paint with image planes
- RudiNieuwenhuis
- 190 posts
- Offline
I'm searching for a metallic car paint shader. The otl that's in exchange is great but i can't get any image planes out of it. (like diffuse, paint, specular, reflections)
Best would be to write a shader myself but i'm a noob when it comes to shader writing. In the vex builder i can get around a bit.
If anyone has some tips, they are more then welcome.
Rudi
Best would be to write a shader myself but i'm a noob when it comes to shader writing. In the vex builder i can get around a bit.
If anyone has some tips, they are more then welcome.
Rudi
Houdini Indie and Apprentice » Apprentice challenge winners
- RudiNieuwenhuis
- 190 posts
- Offline
Does anybody know when and where the winners of the Apprentice Challenge are (will be) announced? I participated and i understood that one month after the deadline (deadline 1st may so 1st of june) the jury would decide and ask the winners to send their files to prove they used Houdini for the animation.
I haven't received any requests so i'm afraid i'm not one of the lucky ones but i'm quite curious to see who did win.
I haven't received any requests so i'm afraid i'm not one of the lucky ones but i'm quite curious to see who did win.
Houdini Indie and Apprentice » Cloth in H10
- RudiNieuwenhuis
- 190 posts
- Offline
Wow, thank you very much for the reply in such detail, that really helps me a lot to get my head around cloth in Houdini. I just downloaded the latest daily build and your edited hipnc file and it's simulating right now, so far so good.
Thanks again!
Rudi
Thanks again!
Rudi
Houdini Indie and Apprentice » Cloth in H10
- RudiNieuwenhuis
- 190 posts
- Offline
I've read some older post (on odforce) where people state that cloth in H9 was close to unusable so with the new H10 i thought i'd give it a try.
Allthough at very low res i get reasonable simulations with multiple rdb-objects, the moment i triangulate2d the grid i use for a cloth a little too much it slows down tremendously and errors out often with memory allocation errors even though I have 16 GB ram in this machine.
Has anyone found a good workflow for cloth or is it still not really usable?
Also the Aprentice HD version does not allow an AutoDopNetwork to be saved out with a ROP output driver so how can i get a initial state to work?
I also very often get the error on the clothsolver: “cannot retrieve data from previous solve”. When this happens, solving just stops and all objects remain static for the rest of the timeline.
Allthough at very low res i get reasonable simulations with multiple rdb-objects, the moment i triangulate2d the grid i use for a cloth a little too much it slows down tremendously and errors out often with memory allocation errors even though I have 16 GB ram in this machine.
Has anyone found a good workflow for cloth or is it still not really usable?
Also the Aprentice HD version does not allow an AutoDopNetwork to be saved out with a ROP output driver so how can i get a initial state to work?
I also very often get the error on the clothsolver: “cannot retrieve data from previous solve”. When this happens, solving just stops and all objects remain static for the rest of the timeline.
-
- Quick Links