Jonas Dahl

So I am working with an orthographic camera because I am doing some animations that will be projection-mapped. So I thought it's the most straight forward to render it orthographic. The thing is, with orthographic view, I lose the perspectiveness of the Z axis of course. I have some light orbs that move around in front of and back of the building. I want them to follow a spline. But they will not appear bigger or smaller in distance, only disappear once they pass through on object. Is there some expression that I could tell objects to scale in size depending on their position in the Z axis? Like to define a point A for Scale A and a Point B in the position for another scale value and in between the object shifts? Then I would not have to animate the scale by hand. Thanks so much!

Just wanted to say thanks for the exact explanation! This helped a lot!

Now I am thinking I can maybe do this with a light and a texture, but I cannot manage to get it mapped correctly...

I am doing some 3D animations for projection mapping and since I don't have a physical model of the stage, I want to preview my animations inside a 3D model. So I want to project the image, I rendered, back onto the scene. So I set up a new material, set it as Camera Mapping, dropped my camera in, all is mapped correctly. But I don't manage to get the material to behave like I want. I'm not (yet) an expert on materials. So I want the white parts to be transparent and the black parts to be pitch black. So the wall gets illuminated by the projector material. I want to setup one material that I can drop on all the objects in the model that are being hit by the projection (there are trusses, LED screens, sage elements, so on) and not overlay the texture in every single material in the scene. The other thing is that I want the projector material to not be affected by any lights. And instead work as an illumination for the wall. So the white parts are more see-through and the black parts stay black. I work in Redshift. Is there a way to set this up in one material, that I can drag and drop onto all the objects so when I make changes, it's just on quick change in the projector material?

I have the exact same issue with my M1 Max Pro with Cinema 4D 2024. I'm just experimenting with a cube with a subdivision surface, simulations and field forces and every 10 min it tells mit it has run out of RAM and I need to force quit. I emptied out my hard drive already,no improvement. I mean, it's just a cube with 50 steps and a subdivision surface. The simulations (cloth) are running smooth in the viewport. So it's far from a heavy scene. I t seems like it does not empty out the RAM? Are there any settings I can modify? It's really annoying.

Hi there, so I am working on a large 3D model of a building that is supposed to have a long simulation (destruction or similar) happening to it. Since I am still in the mode of trying out things, I for now use a low-poly dummy model and when I am satisfied with the simulation, cache it on the real model. That's something I already learning from this forum. Now since I am in stage of trying out stuff, I don't want to cache the simulation over the whole minute (1500 frames) all the time. But also for the final animation, it might make sense, to not cache the simulation over 1500 frames but break it into parts, since there are supposed be some glitches to the model first, reverse back to normal, and then at the end one final simulation destroying it. What would be the best way to go about this? For now I made a null and a copy inside it of the model for each simulation. Here are my questions: - Can I store my cached simulations and switch between different caches? That's a feature I have in pyro, where I can select form the dropbox menu the different caches. But it does not seem to possible with cloth/soft bodies? - Can I keyframe / move positions of several caches in the Timeline after caching them? Ideally, I would cache different simulations always from frame 0 to wherever. And then move them around in the timeline and stack them together. Is this, what "Cache as alembic" is for? - or is it best like I am doing now of duplicating the model for each simulation, render the frames, and then put together in compositing? - How can achieve caching only a few frames instead of the whole timeline? For now I always do it by changing the number of frames in the output but it's a bit annoying. I also tried in the Project settings to limit the simulation until a certain frame but it always runs through the whole timeline. - If I disable an object with a simulation tag in the viewport and render, the simulation is still being calculated, right? Unless I disable the simulation tag. Is there a quicker way to disable several other object while I work on one?

I did but always skews some parts of the topography! Since this is for projection mapping on this exact building, I need to make sure, that none of the edges and points move positions. I have been experimenting with getting more subdivision details with the shader in the render output and trying linear sub-dividing. But in the end I think we will manually go about it. We have some budget left so if we are getting to the serious stage, I can decide if I want to do this or if we give it to someone experienced.

Thank you both for your throrough replies! I will use your advice and see how far I get. 🙂

Hi, I am doing some projection mapping on a building and therefore working with the 3D model/scan of this building. Since they work with it every year, the model was scanned years ago and it works really well. But the topography is very chaotic/skewed. The past years they just dealt with that but now I want to do some things such as simulations where a smooth topography is necessary. I am a bit at a loss though on how to fix this. Remesh did not work, since there some very weird stuff happening (picture attached). Now I guess that's normal for 3D scans out of the box that they have bad topography. The last time I did one I managed to clean it up but the topography was, at least, evenly detailed. Now, I am also not a pro at modeling. I would just like to get an idea on how to approach this and on how difficult this would be - if it was not for projection mapping, I would rebuild the model from scratch. But everything has to line up perfectly on the end. So I don't want to do that. Any tipps would be greatly appreciated.