JustinLeduc

Thanks a ton once again @noseman!

Well darn it, @noseman. You are too kind. It's funny, because I've been tinkering for at least 2 weeks trying to find a solution to this problem, and based on your recent Youtube tutorials (voxel-related ones specifically, which I've watched dozens of times), I recall a handful of times thinking "I feel like noseman would get a kick out of solving this problem". For all I know, you might have solved it in 2 seconds, so I feel a little validated in knowing that it also took you some considerable R&D on your part to find a solution, especially one that's plugin-free. Thanks again a million times. Very much looking forward to seeing your approach. Have a wonderful week-end!

Hey @noseman, "anti-tutorial". Ah! I can't thank you enough for the video you've made on Twitter. I'm very grateful for the well-worded insight you provided me. That said, I must reiterate that Power Cloner has indeed been a successful solution to my problem, minus the CPU performance of it. Speaking of, the performance is not that bad, but regardless, the sole reason I made this post was to find a faster alternative. I've just created and rendered a little "making of" to demonstrate how Power Cloner works for me. You can see how the "non lit" (obscure) part of my character is unaffected, which is the desired outcome. Video: https://streamable.com/xbig40 Please tell me what you make of it. Much appreciated!

@HappyPolygonAlright, thanks a ton for at least confirming my suspicions about needing a raycast-type solution. Power Cloner does the job pretty well for what it is. Ricochet crossed my mind a while back, but I forget why I never gave it a shot. Power Cloner is a bit old and hasn't been updated in a while, so Ricochet might be more CPU-efficient. I'll definitely try it out. Thanks a ton once again!

@HappyPolygon @Havealot @Hrvoje I can't thank you enough for wanting to assist me with my problem. I really appreciate it, truly. While I feel like you guys get the gist of my issue, I feel like I perhaps didn't do such a great job of illustrating my problem. While I'm not allowed to share the actual project file I'm working on, I made another project file (see attachment) that is a bit closer to what the real NDA-restricted project file looks like, but much simplified of course. Take this teapot, for example. It was built from an imported point cloud of a teapot that was then inserted in a Object Cloner. As stated in my first post, from the perspective of a given Camera, I want to dynamically select (for example) the top row of clones based on their Normal direction, as such: Same thing from, let's say, the right: What I'm dealing with in my NDA-restricted project file is an animated point cloud, so there's no way I would ever want to manually select each clone for each frame of the animation. Solution (but undesired): So far, the only way I've managed to make this work is by using Power Cloner, which allows me to "raycast" cubes onto my Cloned object (teapot in this instance) from any angle I want, then use the "collisions" between Power Cloner's raycasted cubes and Cloned object (teapot) as a "Point Object Field" Layer in the Shader Effector, which allows me to precisely color any side of the Cloned Object. This is all good in theory, but unfortunately, in practice, it is slow and clunky, which is why I've been seeking for a better solution for a little while now. Thank you so very much once again. While I have very little XPresso knowledge, I'll give this a try, @Havealot. Thank you! CLONER_PARAMETRIC_SELECTION_V002.c4d

Hey guys! Happy new year! 🙂 I have what I believe to be an interesting problem I've been trying to solve for a few days now. I'm starting to think it's simply not possible, but stubborn as I am, I keep hoping it's not. Goal: I have an array of clones and I want to dynamically select some of those clones based on: 1) their falloff direction 2) the position of the Camera. In other words, I want the selection of those clones to "stop" as soon as 1 clone is found/selected in its given direction. For example, take this simple array of clones (no active Camera): Now from the perspective of a given Camera, I want to dynamically select (for example) the top row of clones, regardless of their XYZ position in the World, as such: The Mograph Selection tool was used to better demonstrate the desired effect. Or, the right clones: My question is: Is this even possible, and if yes, how? As you can see, this isn't a matter of a simple Linear Field. I want a Field (or any other approach that works) to only select the first clone "found" in a given direction, regardless of its XYZ position. Any insight or tip would be immensely appreciated! Thanks a ton guys! 😁 P.S. I've attached the project file in case it's useful to anyone. CLONER_PARAMETRIC_SELECTION_V001.c4d

Hey @HappyPolygon, Big thanks for suggesting another path to explore! I've been staring at your message for the last 10 minutes and it feels like, at this point in time at least, I need to ponder a lot more on this potential solution, i.e. letting it simmer for a while. It definitely has its caveats, but your idea of converting the animation into a stop-motion "format" is definitely worth reflecting on. I think the biggest caveat to this proposed solution (and please correct me if I'm wrong) is the fact that I plan on having many lights emitting from different angles in my final scene. Said lights could also be blocked/intersected by other voxel meshes from all angles. I'm sorry if the videos found in my original post were a bit misleading in that regard, as they only showcase 1 front-facing light. Still, I'll experiment with a few things and get back to you on that. Thank you very much once again!

Hey @keppn, Huge thanks for the possible solution. I've just spent a chunk of time experimenting with SSS, both in Octane and with C4D's default materials, and I was unsuccessful in both cases. As far as Octane goes, I simply couldn't get the desired result (I will be experimenting a lot more on that front), and as for the default materials, SSS was taking so long to render (even with "Fast evaluation" checked off) that I might as well as be using the inefficient method showcased in VIDEO 003 of my original post. Additionally, I couldn't get a result that came close to what you suggested. That said, as I am very inexperienced with C4D's SSS, I'll have to dig deeper into ways of speeding it up, but I'm not holding my breath. I still think that Octane's SSS is the way to go if I'm ever able to put your theory into practice. Either way, I'll be experimenting with the SSS. Thanks again for the help!

Hey guys, A huge thanks for helping me out on this. I really, really appreciate it. @3D-Pangel @keppn The idea of having a low-res shadow map crossed my mind, but here's the thing: my voxel scenes will not just be made of flat walls, meaning that the Z-Depth of said "voxel walls" will vary. So with that in mind, wouldn't the idea of having a low-res shadow map (that still somehow matched the location of my voxels) simply not work, even if up-scaled without sampling in post? It's hard for me to put into words as to why, but I feel like you guys might know what I mean. Plus, I would need a certain amount of flexibility regarding the position/rotation of my camera, and I feel like this method would limit my options greatly. @HappyPolygon Yes! While the end result will resemble a pixelated emission/shadow map, as mentioned in my original post, I'm very committed to having the boundaries of those lights/shadows (that are projected onto my voxel meshes, like the walls) be strictly restricted to the walls' voxels/cubes, which is why I've resorted to Mograph and Fields so far as opposed to dabbling into any texture trickery (which I'm still opened to, but doesn't feel like it would yield the result I'm expecting, the variable Z-Depth of my voxel matrix being one of the reasons). A big thanks once again for the help, truly.

Hey guys! I’m dealing with an interesting challenge. Basically, I’d like to light my scene in a pixel art fashion (except in 3D, so I guess that would be in a "voxel" fashion). I initially wrote a wall of text, but I've decided instead to make a few renders to better illustrate what I'd like to achieve. First, this is my setup (viewport screenshot): https://i.imgur.com/nzJK5iP.png Now, a traditional way to light this scene would be to use a Spotlight, as such: VIDEO 001 : https://i.imgur.com/XYmDKFk.mp4 But this is not the result I want. I would like to find a way to light my scene/meshes in a way that resembles this: VIDEO 002 : https://i.imgur.com/Xza6tnQ.mp4 Problem: The method I used to create the render above is utterly inefficient and CPU-intensive. In short, I'm duplicating each voxel of the bicycle, extruding those duplicates on the Z-axis and detecting any intersection between those extrusions and the "wall". See here (in red): VIDEO 003 : https://i.imgur.com/gH8ZwSG.mp4 Objective: I would like to find an efficient way of having the boundaries of the lights/shadows projected onto my meshes (like the voxel walls) be restricted to the voxels/cubes they (those same meshes) are made of, i.e. in a pixel art fashion. I've attached the demo scene to this post. Please note that Octane was used in this project file, but I'm also opened to using default materials if needed. At this point in time, I’m wondering if this is even possible to do with Octane, or Cinema 4D for that matter (unless I decide to dive into Python or something). That said, given my novice status, there could be a ridiculously obvious solution that I haven't thought of. I feel like this could be more likely... Any advice/tip/insight would be hugely appreciated. Seriously. Even just keywords to Google and what not, because I'm running out of keywords to Google over here... Thanks a ton! BIKE_DEMO_001.c4d

I can't thank you enough, @jed! I must admit that I was secretly hoping you'd reply, given that I saw this message of yours yesterday and was bummed to find out that the project file was unavailable. I'm happy to report that your node system in the provided project file is fully understood, thanks to your reply + remark nodes. I learned a lot. If this helps anyone in the future; I've removed the "Visible in Editor/Renderer" inputs and changed them for "Scale X/Y/Z". After baking the XPresso animation ("Bake Object" in Timeline with all parameters selected) and putting all of the grouped objects into a non-welded Connect, I was able to bake and merge the Connect into a single Alembic file. Thanks a million once again!

Hey guys, I'm attempting to make some kind of "stop motion system". Goal : In short, picture a scene with 100+ objects of the same character in different poses (ex: a walk cycle) and I want to iterate through the list of objects and show only 1 object per frame. So, in other words, some kind of sequential visibility. Problem : I was able to accomplished this, but I think I did it wrong, because I don't understand why it works. If you peek at my scene file (huge thanks in advance), my "Compare" node has a "!=" operator. Is this the right way to do this? I thought it'd have to be a "==", but as expected, this gives me the opposite effect. Additional requirements : Additionally, I would very much like to achieve 2 things: - Have control over the framerate of the animation. For example, I'd have XPresso jump to the next object every X frame, as opposed to every 1 frame. I'm hoping that this isn't a huge addition. I'm currently attempting to achieve this. - Export the final sequence animation as an Alembic file (merged in 1 file). I tried to put what I currently have in a "Connect" object and export that as an Alembic, but I was unsuccessful in doing so. Nothing was animated. Do I need to bake the Xpresso? I own Cinema S22/R23, as well as the NitroBakeV2 plugin (perhaps relevant here). Thanks a million in advance for your help! I very much appreciate it. Justin CHARACTER_ANIMATION_TEST.zip

Yeah you're most probably right. Not keeping my hopes up. Really really appreciate your scene file though! My end goal is just to have an object set as an obstacle blocking the waves on a shore/beach shape. I just be able to get that done from your scene file if I watch the Liquid simulation tutorial again. All I have to do after is the foam and spray! Thanks again.

Ah. I stumbled upon this thread 20 minutes and I was also on Wayback machine. No luck so far. I'll post here if I find something!