Generic forum to discuss Octane Render, post ideas and suggest improvements.
Forum rules
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
Recently people seems to debate a lot about how easy/difficult would it be for Octane to implement Motion Blur.
I'm not a programmer, so I have no idea what are the technical challenges behind this, but considering that when you render out a turntable animation in Octane standalone (probably not many people use this feature), it DOES HAVE Motion Blur, so what's the problem? Once the Alambic animation is supported, couldn't Octane just use the same Motion Blur it already uses for turntable?
Sorry if this is a dumb question, I just want to understand why a lot of people are concerned about it...
Yes, it is MB, but only camera MB.
This means, if the camera is fix and the meshes are moving to get the same effect, that every object/vertex has the same speed.
To calculate this is relative simple, each vertex is added or subtracted by the value, on which the camera moves, rotates or is zoomed.
On a scene, where the meshes moves with different speeds, or vertices changed her position in different ways, it is a little bit different.
To calculate this, you must know the position which the vertices have before or after the current frame.
You see, on the camera MB, you have only some camera values to calculate the other vertext positions. This can be the camera position, rotation or a zoom. Let it be 5 3d-vectors.
With full MB you need the same amount of vectors, which each mesh has.
It is also possible to use different interpolation techniques to blend between the start and end point of the animation, like linear, cubic and so on.
With linear it is simple, you need only the vertex start and end point. So you have the same amount of data to calculate the MB, which the mesh vertices use.
With cubic or spline, you need positions between the start and the end point, so you can have 3 or 4 time the amount of data.
I think the problem isn´t to calculate the MB, it is the amount of the VRam you need, to store the data with which you can calculate it.
Maybe a simple MB is comming soon, not vertex based, but mesh based.
So you have only on transformation matrix for the entire mesh to calculate the MB.
With this it isn´t possible to generate MB on vertex based transformations like skeletons/bones or mathematical ones.
But you can scale, rotate and translate the mesh to get a MB.
