edwardIn my experience, the console is not reappearing.
If that console window is accidentally closed, then it will open again by itself when there's the next printf.
Found 129 posts.
Search results Show results as topic list.
Technical Discussion » How to bring up the console window on Windows?
- AnsonSavage
- 129 posts
- Offline
Technical Discussion » How to see why a TOP node failed
- AnsonSavage
- 129 posts
- Offline
Hi! I'm pretty new to TOPs and I've been following this super awesome tutorial [www.sidefx.com] series on it. When I got to the chunk on the ImageMagick node, every time I run it, I see that the work items failed but I don't know why.
As far as I can tell, ImageMagick has been properly installed.
Is there any way I can get the error message to see why this is failing?
Thanks!
Anson
As far as I can tell, ImageMagick has been properly installed.
Is there any way I can get the error message to see why this is failing?
Thanks!
Anson
Edited by AnsonSavage - 2023年3月30日 18:25:40
Technical Discussion » Move points to their closest unique neighbors
- AnsonSavage
- 129 posts
- Offline
I have a question along the lines of this [www.sidefx.com] one, but I have a feeling there's a more efficient and elegant way to handle what I'm trying to accomplish.
Basically, I have two geometries of points, each with an equal number of points.
Set 1:
Set 2:
I would like to smoothly transition each point from its location in the first set to its location in the second set. If I do this using traditional methods (like using a blendshape or lerping between the P attributes) there is a lot of weird motion because the point numbers in these sets are very different. I'd like each point in set 1 to move to the closest point in set 2, but I don't want any of the points from set 1 to end up on the same point in 2.
I've tried sorting the points across the diagonal vector, but simply due to the difference in point density, the point numbering is still different enough that there is too much weird motion when I blend their positions. I've also managed to make it so each point from set 1 moves towards the closest point it can find in set 2, which avoids the crazy motion, but this makes it so many points overlap (due to having the same closest point) once the motion is completed. Is there a way to achieve this effect?
Thanks!
Anson
Basically, I have two geometries of points, each with an equal number of points.
Set 1:
Set 2:
I would like to smoothly transition each point from its location in the first set to its location in the second set. If I do this using traditional methods (like using a blendshape or lerping between the P attributes) there is a lot of weird motion because the point numbers in these sets are very different. I'd like each point in set 1 to move to the closest point in set 2, but I don't want any of the points from set 1 to end up on the same point in 2.
I've tried sorting the points across the diagonal vector, but simply due to the difference in point density, the point numbering is still different enough that there is too much weird motion when I blend their positions. I've also managed to make it so each point from set 1 moves towards the closest point it can find in set 2, which avoids the crazy motion, but this makes it so many points overlap (due to having the same closest point) once the motion is completed. Is there a way to achieve this effect?
Thanks!
Anson
Technical Discussion » How to access SOP attributes in DOP context
- AnsonSavage
- 129 posts
- Offline
Hi!
I have an RBD simulation where I'm setting some of the points as active and others as inactive in the SOP context. When I use the SOP rbdbulletsolver node, this works great and it's respecting these attributes.
However, I'm realizing that I might want to try some more complicated things (like set the gravity or mass per each piece or limit the number of broken constraints based on how many constraints have already been broken). I figured that to do these things, I'd probably need to build my own DOP Network.
However, I'm finding that this new DOP network isn't respecting the active attribute, and it's also not respecting the rest length which is specific to each primitive in the constraints.
How can I make it so that my DOP network reads and respects these attributes from the SOP context?
Thanks!
Here's my HIP file for reference:
I have an RBD simulation where I'm setting some of the points as active and others as inactive in the SOP context. When I use the SOP rbdbulletsolver node, this works great and it's respecting these attributes.
However, I'm realizing that I might want to try some more complicated things (like set the gravity or mass per each piece or limit the number of broken constraints based on how many constraints have already been broken). I figured that to do these things, I'd probably need to build my own DOP Network.
However, I'm finding that this new DOP network isn't respecting the active attribute, and it's also not respecting the rest length which is specific to each primitive in the constraints.
How can I make it so that my DOP network reads and respects these attributes from the SOP context?
Thanks!
Here's my HIP file for reference:
Image Not Found
Edited by AnsonSavage - 2023年3月2日 18:13:53
Technical Discussion » Reference Houdini SOPs from other Files in Solaris
- AnsonSavage
- 129 posts
- Offline
Hello!
I am really new to the world of Solaris and USD, so maybe I'm asking the wrong questions, but I'll just explain the challenge I'm trying to solve real fast.
I'm on a team where we're pumping out a bunch of iterations of an animation in rapid succession. One of my team members is the main animator and I'm one of the main technical people that goes in to fix certain issues. This situation has happened a number of times:
This is inefficient and error prone.
I am sure there are more efficient ways to get similar effects such that he can continue working without pausing, and then the fixes I make can automatically be pulled into his file. I understand that using USD references can solve this issue in many cases. But he needs to have the top level controls to parameters of entire SOP networks for the animation we're doing, so if I understand correctly we couldn't save out the files as USDs with parameters (e.g., number of box subdivisions), but maybe that's an incorrect assumption.
What are some good ways to approach this issue?
Thanks!
Anson
I am really new to the world of Solaris and USD, so maybe I'm asking the wrong questions, but I'll just explain the challenge I'm trying to solve real fast.
I'm on a team where we're pumping out a bunch of iterations of an animation in rapid succession. One of my team members is the main animator and I'm one of the main technical people that goes in to fix certain issues. This situation has happened a number of times:
- My teammate is working on file_version_9.
- He asks me to make a fix while he's still working on the animation.
- I go in and save the file as file_version_9_with_fixes.
- By the time I've created the fixes, he might either be still working on file_version_9 or file_version_10.
- Either way, he has to pause working for a second so that we can open his latest file, find all the changes I made for the fixes, and then copy them into the most recent file.
This is inefficient and error prone.
I am sure there are more efficient ways to get similar effects such that he can continue working without pausing, and then the fixes I make can automatically be pulled into his file. I understand that using USD references can solve this issue in many cases. But he needs to have the top level controls to parameters of entire SOP networks for the animation we're doing, so if I understand correctly we couldn't save out the files as USDs with parameters (e.g., number of box subdivisions), but maybe that's an incorrect assumption.
What are some good ways to approach this issue?
Thanks!
Anson
Technical Discussion » How to determine material based on light path
- AnsonSavage
- 129 posts
- Offline
Technical Discussion » How to determine material based on light path
- AnsonSavage
- 129 posts
- Offline
Okay, I can confirm that using a raybounce node does work with a Mantra Vex material imported into Karma. I can also confirm it doesn't work with Karma XPU. I wasn't sure how to use a renderstate node, but that does sound interesting.
Thanks so much for the help!
Thanks so much for the help!
Technical Discussion » How to use light path conditions.
- AnsonSavage
- 129 posts
- Offline
Blender has a light path node that's really useful for setting up conditional shaders (e.g., I want a material to look like glass if my camera sees it directly, but I want it to appear transparent to any other bounce rays). Does Karma have any way to set up a conditional shader system like this? If not, does Mantra?
Thanks!
Thanks!
Technical Discussion » Thicken Node
- AnsonSavage
- 129 posts
- Offline
-
- Quick Links