[David Hadden]

We appear to have something that is giving the correct color -> directions, however we’re still fine tuning this a bit, and trying to get a more elegant alpha channel solution ( currently we’re using object passes and only allowing the objects that are moving into the pass, however if we have objects that aren’t moving the whole time, then we’d have to keyframe the inclusion of those objects which I’d like to try and create a better solution.

We’re also getting a bit of trouble with some of the information, so it’s not perfect yet, but it’s getting closer.

A great big shoutout to the blenderartists.org community for helping make this happen btw.

[David Hadden]

We appear to have something that is giving the correct color -> directions, however we’re still fine tuning this a bit, and trying to get a more elegant alpha channel solution ( currently we’re using object passes and only allowing the objects that are moving into the pass, however if we have objects that aren’t moving the whole time, then we’d have to keyframe the inclusion of those objects which I’d like to try and create a better solution.

We’re also getting a bit of trouble with some of the information, so it’s not perfect yet, but it’s getting closer.

A great big shoutout to the blenderartists.org community for helping make this happen btw.

[David Hadden]

Hi Pierre,

We’re working on a node layout that I think will fix this, however we’re getting some issues with keeping non-moving areas of the scene from changing off of the middle value. I’ll let you know what I end up coming up with.

Currently we’re taking the X Frame from previous and next, and mixing it into the Red Channel, and then the Y movement from Previous and Next and mixing it into the Green channel, and then we’re outputting that image, and then putting the alpha channel of all the moving objects through to the alpha channel of the render layer so that we have the alpha channel information pushing through which I believe is the way that the RSMB plug-in is working it out. We’re also normalizing the Red and Green channels separately, however I believe that’s where we’re running into the issue in blender because it *appears* from my limited understanding of how the normalize function is working that it’s not normalizing it the way we want :(.

I’ll let you know what we come up with, in the meantime, you’re quite right, RSMB Pro is a very good tool for adding Motion Blur post render on it’s own.

[David Hadden]

Hi Pierre,

We’re working on a node layout that I think will fix this, however we’re getting some issues with keeping non-moving areas of the scene from changing off of the middle value. I’ll let you know what I end up coming up with.

Currently we’re taking the X Frame from previous and next, and mixing it into the Red Channel, and then the Y movement from Previous and Next and mixing it into the Green channel, and then we’re outputting that image, and then putting the alpha channel of all the moving objects through to the alpha channel of the render layer so that we have the alpha channel information pushing through which I believe is the way that the RSMB plug-in is working it out. We’re also normalizing the Red and Green channels separately, however I believe that’s where we’re running into the issue in blender because it *appears* from my limited understanding of how the normalize function is working that it’s not normalizing it the way we want :(.

I’ll let you know what we come up with, in the meantime, you’re quite right, RSMB Pro is a very good tool for adding Motion Blur post render on it’s own.

[David Hadden]

I have a the information for what channel is used for what information here…

The image (RGBA) is based on screen space pixel movement of vertices… R being the x direction from current frame to previous frame, G being y direction from current frame to next frame, B being x direction from current frame to next frame, A being y direction from current frame to next frame.

So it sounds like I’m simply out of luck trying to use RSMB to interpret this information due to the fact that it’s not even giving a “speed” information, it’s just giving movement to and from previous frame utilizing all 4 channels of an RGBA image.

Any ideas on how I might be able to use it would be very much appreciated. I’m not tremendously familiar with blender’s internal renderer as I much prefer the look of 3rd party tools, such as octane and the like.

[David Hadden]

I have a the information for what channel is used for what information here…

The image (RGBA) is based on screen space pixel movement of vertices… R being the x direction from current frame to previous frame, G being y direction from current frame to next frame, B being x direction from current frame to next frame, A being y direction from current frame to next frame.

So it sounds like I’m simply out of luck trying to use RSMB to interpret this information due to the fact that it’s not even giving a “speed” information, it’s just giving movement to and from previous frame utilizing all 4 channels of an RGBA image.

Any ideas on how I might be able to use it would be very much appreciated. I’m not tremendously familiar with blender’s internal renderer as I much prefer the look of 3rd party tools, such as octane and the like.

[David Hadden]

Hi Pierre, thanks for your help,

I found some information regarding the pass information, but it’s a bit old and I don’t know if it’s still accurate. Lot of people asking the question when I search, not a lot of answers 🙁

https://blenderartists.org/forum/showthread.php?133266-RGB-Speed-Vector-Pass-in-external-compositing-apps

[David Hadden]

Hi Pierre, thanks for your help,

I found some information regarding the pass information, but it’s a bit old and I don’t know if it’s still accurate. Lot of people asking the question when I search, not a lot of answers 🙁

https://blenderartists.org/forum/showthread.php?133266-RGB-Speed-Vector-Pass-in-external-compositing-apps