They have different dimensions. The one is bigger and the other one is a little pice cut out of it.
I have gathered that size doesn't matter, as when resized the displayed resolution does not change.
The deciding factor seems to be the total resolution or number of voxels. The more voxels I've got, the more blurred details look in the viewport.
I played around with the values for 3d textures but they didnt change anything.
I guess it's not a big deal in general but a little unfortunate in this particular setup.
Found 12 posts.
Search results Show results as topic list.
Technical Discussion » Volume Rasterize is scale dependend??
- O_Speiser
- 12 posts
- Offline
Technical Discussion » Volume Rasterize is scale dependend??
- O_Speiser
- 12 posts
- Offline
Okay, further testing has been done.
Actually, only the viewport looks different (even after the pyro solver) but when rendered, they look the same. So the data is indeed the same. That kinda answers all former questions I guess...
But now my (hopyfully)final question for today:
Can I tell the houdini viewport to render both volumes more similar to each other?
Actually, only the viewport looks different (even after the pyro solver) but when rendered, they look the same. So the data is indeed the same. That kinda answers all former questions I guess...
But now my (hopyfully)final question for today:
Can I tell the houdini viewport to render both volumes more similar to each other?
Technical Discussion » Volume Rasterize is scale dependend??
- O_Speiser
- 12 posts
- Offline
Too make my question more clear, I prepared and example file and attached it.
There's a volume rasterized out of a long line of noisy particles and then a volume created just out of the center piece.
As a volume in viewport they look different. But as seen throgh a volumeslice they look the same. Why is this so?
Again, thanks a lot.
Oliver
There's a volume rasterized out of a long line of noisy particles and then a volume created just out of the center piece.
As a volume in viewport they look different. But as seen throgh a volumeslice they look the same. Why is this so?
Again, thanks a lot.
Oliver
Technical Discussion » Volume Rasterize is scale dependend??
- O_Speiser
- 12 posts
- Offline
Hello,
I am struggling with a pyro steup. I got a river delta like shape of a flip sim that represents lava. I want to use the flip particles as a source for a pyro sim.
For faster art direction, I deleted most points except a small area. When I volume rasterize and simulate this small area I am getting the desired results.
BUT: If I use all the source particles, the volume rasterize and the sim looks different very different from the little patch as an isolated volume. I have compared the density values and the grid size of the volume rasterized particles using a volumeslice and found that they look identic there. But seen as a regular volume in viewport they look vastly different.
Then I thought, "well maybe it's just the viewport that renders different" and played with the display parameters. Then I realized though, that even the sim behaves vastly different. It is a sparse solver in a dop network by the way. But if the volumeslice shows identic data, then it's maybe some interpretation by the viewport and the pyro network?
What am I doing wrong? And how do other people approach the issue of a pyrosim that covers a rather wide area but doesnt need much detail. I could split the sim in little patches and I guess the result would be fine but it really bugs me that I don't understand what's going on. I felt like I finally understand volume stuff reasonably well.
I'm on houdini 19.
I greatly appreciate your help!
I am struggling with a pyro steup. I got a river delta like shape of a flip sim that represents lava. I want to use the flip particles as a source for a pyro sim.
For faster art direction, I deleted most points except a small area. When I volume rasterize and simulate this small area I am getting the desired results.
BUT: If I use all the source particles, the volume rasterize and the sim looks different very different from the little patch as an isolated volume. I have compared the density values and the grid size of the volume rasterized particles using a volumeslice and found that they look identic there. But seen as a regular volume in viewport they look vastly different.
Then I thought, "well maybe it's just the viewport that renders different" and played with the display parameters. Then I realized though, that even the sim behaves vastly different. It is a sparse solver in a dop network by the way. But if the volumeslice shows identic data, then it's maybe some interpretation by the viewport and the pyro network?
What am I doing wrong? And how do other people approach the issue of a pyrosim that covers a rather wide area but doesnt need much detail. I could split the sim in little patches and I guess the result would be fine but it really bugs me that I don't understand what's going on. I felt like I finally understand volume stuff reasonably well.
I'm on houdini 19.
I greatly appreciate your help!
Technical Discussion » Sucking Smoke towards a point
- O_Speiser
- 12 posts
- Offline
Thanks a lot Tomas, it's working now!
As usual, the next problem showed up. Now the smoke dissipates once it's pulled to the area of negative divergence. Do you or anyone else know why this would happen?
I tested with and without the divergence and it's a clear difference how long the smoke stays around. I also tested with super low dissipation values.
I am using the sparse pyrosolver by the way.
Maybe there's some cutoff happening?
I want the smoke ultimately to build something like a ball of smoke that is still affected by turbulence.
Once again, thanks!
edit: again added some text
As usual, the next problem showed up. Now the smoke dissipates once it's pulled to the area of negative divergence. Do you or anyone else know why this would happen?
I tested with and without the divergence and it's a clear difference how long the smoke stays around. I also tested with super low dissipation values.
I am using the sparse pyrosolver by the way.
Maybe there's some cutoff happening?
I want the smoke ultimately to build something like a ball of smoke that is still affected by turbulence.
Once again, thanks!
edit: again added some text
Edited by O_Speiser - Sept. 9, 2022 09:12:37
Technical Discussion » Sucking Smoke towards a point
- O_Speiser
- 12 posts
- Offline
Thank you for the quick reply.
Indeed my negative divergence field encompasses the whole scene. Once I'm back at the desk tomorrow I will try tweaking the size.
Do I understand right that negative divergence reverses the effect of the pressure projection?
Thanks a lot!
Indeed my negative divergence field encompasses the whole scene. Once I'm back at the desk tomorrow I will try tweaking the size.
Do I understand right that negative divergence reverses the effect of the pressure projection?
Thanks a lot!
Technical Discussion » Sucking Smoke towards a point
- O_Speiser
- 12 posts
- Offline
Hello,
I am trying to figure out how to pull smoke towards a point.
The setup involves a fire consuming a source, and then that smoke is supposed to get pulled towards a point to form something like a sphere.
I tried sourcing velocities and divergence. Obviously velocities won't work because the pressurefield just cancels them out to avoid compression. But I want the compression! mh.
Negative divergence just makes the smoke and flames disappear.
Any idea how I can maybe scale down the effect of the pressure field, or some other ideas how to solve this isse?
All the best and thanks,
Oliver
edit: added some text
I am trying to figure out how to pull smoke towards a point.
The setup involves a fire consuming a source, and then that smoke is supposed to get pulled towards a point to form something like a sphere.
I tried sourcing velocities and divergence. Obviously velocities won't work because the pressurefield just cancels them out to avoid compression. But I want the compression! mh.
Negative divergence just makes the smoke and flames disappear.
Any idea how I can maybe scale down the effect of the pressure field, or some other ideas how to solve this isse?
All the best and thanks,
Oliver
edit: added some text
Edited by O_Speiser - Sept. 9, 2022 09:11:49
Technical Discussion » SVD Decomp and fluid sheets
- O_Speiser
- 12 posts
- Offline
Hello,
Short version: How to actually use the svddecomp vex function once I successfully calculated a weighted covariance?
Long version:
I am working on a liquid project and am trying to implement a formula from a mathematical paper (for the first time ever...). The poblem function deals with SVD decomposition and is found in point 4.1 of this paper: https://ryichando.graphics/sheetflip/download/sheetflip.pdf
Anyways, I learned a lot already, but I am hitting a wall looking at the svddecomp entry from the houdini manual:
"void svddecomp(matrix3 input_M, matrix3 &output_U, vector &output_S, matrix3 &output_V)"
First I don't understand the "void" aspect. How do I actually apply this function? Second, I am not sure what I need to feed into it and what the "&" sign means. I guess "&" means something like optional input. Is that correct?
Do I need to calculate the output_U S and V first? I'm confused.
I think I understand the general usage of the covariance and SVD decomposition,even though I cannot grasp the math 100%.
My goal is to detect the thin sheets and "edges" of my fluid sim, to then apply forces relative according to that mask.
I tried to do ths with volume workflows using density and curvature but my results were never really good.
Thanks a lot!
Oliver
Short version: How to actually use the svddecomp vex function once I successfully calculated a weighted covariance?
Long version:
I am working on a liquid project and am trying to implement a formula from a mathematical paper (for the first time ever...). The poblem function deals with SVD decomposition and is found in point 4.1 of this paper: https://ryichando.graphics/sheetflip/download/sheetflip.pdf
Anyways, I learned a lot already, but I am hitting a wall looking at the svddecomp entry from the houdini manual:
"void svddecomp(matrix3 input_M, matrix3 &output_U, vector &output_S, matrix3 &output_V)"
First I don't understand the "void" aspect. How do I actually apply this function? Second, I am not sure what I need to feed into it and what the "&" sign means. I guess "&" means something like optional input. Is that correct?
Do I need to calculate the output_U S and V first? I'm confused.
I think I understand the general usage of the covariance and SVD decomposition,even though I cannot grasp the math 100%.
My goal is to detect the thin sheets and "edges" of my fluid sim, to then apply forces relative according to that mask.
I tried to do ths with volume workflows using density and curvature but my results were never really good.
Thanks a lot!
Oliver
Houdini Lounge » Geometry to Milk
- O_Speiser
- 12 posts
- Offline
Hello again,
I think I managed to more or less implement your workflow, but my results are much worse than what you showed here. I guess I am just lacking the experience.
I came to a point where I feel that a sculpting workflow maybe wouldnt be so bad in this particular case, but I have even less experience with sculpting.
Then I started playing again with FLIP and I think it's still the way to go, even though it takes a lot of patience to find appropriate values.
I am happy to collect more tips, so if anyone has more ideas or knows how to further improve those mentioned in this post, I would be very happy and thankful.
I think I managed to more or less implement your workflow, but my results are much worse than what you showed here. I guess I am just lacking the experience.
I came to a point where I feel that a sculpting workflow maybe wouldnt be so bad in this particular case, but I have even less experience with sculpting.
Then I started playing again with FLIP and I think it's still the way to go, even though it takes a lot of patience to find appropriate values.
I am happy to collect more tips, so if anyone has more ideas or knows how to further improve those mentioned in this post, I would be very happy and thankful.
Houdini Lounge » Geometry to Milk
- O_Speiser
- 12 posts
- Offline
Hello Konstantin,
Thanks for the answer. It looks very promising. Maybe as the starting point of a sim.
Could you elaborate a bit on the workflow?
Thanks!
Thanks for the answer. It looks very promising. Maybe as the starting point of a sim.
Could you elaborate a bit on the workflow?
Thanks!
Houdini Lounge » Geometry to Milk
- O_Speiser
- 12 posts
- Offline
Hello and thanks for the answer!
I checked out your video and I like the approach but I'm not sure if the look I can achieve with that method is gonna be organic/liquid enough for the client.
I feel like I don't get around simulations.
I checked out your video and I like the approach but I'm not sure if the look I can achieve with that method is gonna be organic/liquid enough for the client.
I feel like I don't get around simulations.
Houdini Lounge » Geometry to Milk
- O_Speiser
- 12 posts
- Offline
Hello,
I am trying to achieve a similar effect as seen in the attached jpg file. My arm gemoetry is animated.
Last time I did splahes, I prepared the source points with various noises and some modeling to achieve an interesting starting shape and velocity values. Here, the difficulty is to preserve most of the shape and only have certain areas work.
Maybe it doesnt need a sim at all. Extrusion, scattering of spheres to get tendril shapes and vdb smoothing. After that, bring in movement through either noise on the tendrils or even vellum. Ideas?
Thanks a lot,
Oliver
I am trying to achieve a similar effect as seen in the attached jpg file. My arm gemoetry is animated.
Last time I did splahes, I prepared the source points with various noises and some modeling to achieve an interesting starting shape and velocity values. Here, the difficulty is to preserve most of the shape and only have certain areas work.
Maybe it doesnt need a sim at all. Extrusion, scattering of spheres to get tendril shapes and vdb smoothing. After that, bring in movement through either noise on the tendrils or even vellum. Ideas?
Thanks a lot,
Oliver
-
- Quick Links