Jump to content

keppn

Premium Member
  • Posts

    390
  • Joined

  • Last visited

  • Days Won

    19

Everything posted by keppn

  1. Well, to be honest, I didn't have a point at all when starting the thread. Just an interesting tech preview πŸ™‚ But it evolved quite naturally to the discussion of workflows and simulation solutions, so, why not continue? πŸ™‚ It's always interesting to read different approaches and perspectives. Dave, for you the interoperability of simulation systems is a high priority, and also continous control without caching or baking steps. Totally valid point, I mean, you know best what kind of work you want to do and how you want to achieve it πŸ™‚ For me, I welcome every development that allows for a more playful exploration of possibilities. Coming from traditional arts, I value immediate feedback and friendly UIs that hide the deep tech from me. It's still seldom, but realtime stuff sometimes even bestows upon me the "happy accident". So, it's nice that there are different solutions with different approaches available for different kind of artists. I didn't mean to bash X-Particles, it's still a great tool and I use it quite often for Particle work. What I honestly don't know at all: Can a plugin escape the performance frame set by the DCC? So, could GPU support really make a difference for X-Particles, if it needs to constantly talk to "old" parts of Cinema 4D? I surely do hope so! On the other hand, a fresh start with a very modern standalone app seems to grant a massive head start in terms of raw performance. (With the obvious disadvantages like caching, etc.) In the end, everyone needs to evaluate for himself and make his own decisions... and I enjoy talking about that kind of stuff a lot πŸ™‚
  2. Ah, interesting view, and also a nice move back on topic - liquids πŸ™‚ Yes, multiple physics interactions in one system with good performance really is the simulation endgame. I wonder which package/platform/plugin will arrive there first? X-Particles is already there, functionality wise. They really need a huge performance gain, though, and I'm not sure, if they can break free from plugin-restrictions. Also, wasn't GPU support promised for late 2020? I just hope they get there some day... Houdini is also there, and it also has the performance. I'm really not willing to learn that thing, though πŸ˜… JangaFX... hm, Embergen is peerless for the reasons mentioned above, but I doubt they will go for multi-physics-systems. It seems, Embergen and Liquigen will be specialized realtime-tools for specific jobs, in their respective silos. I have no idea about Blender's competitiveness in this area. Seeing Blender's overall development, it surely ain't too bad πŸ˜„ Nextlimit... oh boy, Realflow for C4D was such a bad buy. It's nowhere near the capabilities of the standalone, only geared towards really small product shots etc. Also, they do feature GPU support inside C4D and... it doesn't do that much. Maaaaybe Maxon will totally suprise us someday with a fully integrated solution based on their highly performant scene-nodes? Who knows...
  3. Lot of words, but here's a simple example: In this short clip I draggeed down the slider for the amount of smoke emitted. It's so satisfying when you get the setting "just right" with immediate feedback: https://imgur.com/kF3HAYb
  4. Hi Dave, I understand your point completely, and in fact, it was my point of view not so long ago: "Why should I integrate another external tool with all the hassle of importing/exporting and keeping track of separate simulation files? Even more though, when I already shelled out a considerable amount of cash for an integrated solution (I own X-Particles, too)." Well, for me it's all about the quality of work you do in step 3. Here's an example: In Embergen, I very quickly get a desired result and then spend most of the time really finetuning my simulation. Not much technical knowledge is needed to keep everything fluid and fun. Typically, at the end of a session I think: "Yeah, whoa, that's perfect" (btw, I don't care about the shading in Embergen at all, since I have to redo that in Redshift anyway.) In X-Particles, it always begins kind of fluid, but very soon the simulation becomes so slow, that I can only dial in a dozen of artistic changes. Before it becomes just too frustrating to wait.. dial in... wait... Typically, at the end of a session I think: "Yeah, whatever, that needs to do it now". (Btw, Insydium has fantastic video-tutorials, but Bob always dials in the right values *BEFORE* seeing the effect, because he's just an expert. If you don't reach that kind of expertise, it's too easy to really fall into performance-pits left and right.) For me, it has been quite an expensice way to come to the conclusion: Simulation-Plugins just don't cut it in C4D. I moved on to expand my pipeline with external tools like Marvelous Designer and Embergen. It's true, you need to spend a little more time for setup and preparation (and of course, there's additional organizational overhead with files and versions). But for me, it's really worth the (managable) extra hassle since the quality is just better. And I have more fun πŸ™‚ btw, Embergen has a free trial period. Please, just try it out, and you will see πŸ™‚ (And, just in case: I'm just an excited user, this is not an ad in any way πŸ˜…)
  5. They seem to have a lot of work still ahead, but the first teaser of Liquigen is quite promising: Realtime liquids: In case you haven't yet tried Embergen, their realtime fire&smoke sim - do try! It has a great UI, thanks to it's realtime approach you can cram a ton of artistic decisions into a session, and interoperability via VDBs is quite okay. But above all, it's plain fun πŸ™‚ I'll never touch xParticles/Explosia again πŸ˜…
  6. Ah... so, if I understand correctly, a voxel in the background should have only 2 states: 1. Completely in the light OR 2. Complety in the dark (shadowed) ...Like a pixel with two luminance states. Then how about this: - Build the wall from voxels - Assign individual color materials - The base color (for example, diffuse) should be the shadow color - quite dark - Now give the material a strong Subsurface Scattering color. Now, when a voxel is in the shadow (not hit by light source), it should stay dark (the shadow color defined in diffuse) When a voxel is hit by light source, it should light up complety, due to SSS. With a medium SSS setting, you could even produce some kind of "pixel flickering" like dithering in pixel art. ... it's just a theory, though πŸ™‚
  7. Just an idea: - Start with approach 1 - Lower the output resolution to something like 160x100 - Carefully position the Camera, so that a voxel roughly equals an output-pixel (you could make an overlay for the camera with a grid of output-pixels) - Everything including shadows should now be pixelated - Upres in post as needed (without filtering for "clean" pixel look)
  8. No, I wasn't aware of C4D-Usermeetings around town... but I would actually be super-interested in joining, as soon as the situation allows. "Normal" friends tend to be bored soon enough by 3D talk πŸ˜…
  9. Gerresheim, far in the east πŸ™‚
  10. Hi Marc, greetings from the other side of DΓΌsseldorf! The world is a pie πŸ˜…
  11. Wow! Totally surprising, and a real power-move. Whenever I dip into zBrush, I get lost in that labyrinthic GUI and also step in a thousand pitfalls. But when actually sculpting, it's simply peerless. So organic, so responsive, so natural. I guess, the new acquisitions will live unaltered alongside each other for quite a while, but it's obvious now, that Maxon builds it's own CreativeCloud now. Not sure about the Forger buy now that they acquired the big one, but ok. Let's see πŸ™‚
  12. I'm not a Blender user, but from the outside, that's looking like a massively useful update! Have fun, guys! πŸ™‚
  13. Yeah, from the outside, Insydium fits so obviously, that there must be a deeper reason that they haven't been bought yet. Maybe it's money, maybe it doesn't fit with Maxon's core-rewrite/scene-node strategy, maybe their tech has a hard performance ceiling... who knows...
  14. I'm tempted, too. But when looking at tutorials, it seems just like another grave to bury fiddling-time πŸ˜„ Also, Explosia is of course far worse performance-wise, but it's so neatly integrated with the rest of x-particles. I just hope, Insydium gets their GPU-support up and running sometime... ( I think it was promised to the end of 2020, wasn't it?)
  15. Pixellab 30% off everything https://www.thepixellab.net (I love their stuff, btw -- really good quality)
  16. I found this helpful to understand Insydium's maintenance conditions:
  17. Insydium 25% on Maintenance 40% on Fused https://insydium.ltd/
  18. Topas Video Enhance AI 100$ off, additional 30$ off with code "RUMOR" https://www.topazlabs.com/video-enhance-ai
  19. Adobe 40% off of the first year of Creative Cloud https://www.adobe.com
  20. I have not much experience with Xpresso, so I can only add a very general idea to a different approach: Could you use an animated noise to offset the particle position? That could provide the flowing change between attraction and repulsion and it would be art-directable via the noise-settings. It should work somewhat similar to X-Particles Advection to a Explosia-sim.
  21. Haha, yeah, MusicMaker is always plain fun πŸ˜„
  22. Kitbash 3D 50% off, all kits https://kitbash3d.com/
  23. Hi guys, let's collect and share relevant "Black / Cyber - Friday / Monday / Week / Season"-Deals πŸ˜€
  24. Chris Schmidt's tutorials are a delight to watch. Most of the times, I'll never actually use the techniques shown; but he has broadened my awareness to unusual solutions sooo much. Also, it's just plain fun to watch, because he's fast and witty and quite thorough in his explanations.
  25. I couldn't go back to a version without the dynamic placement and scattering tools. For the stuff I do, those feature were like a Quad-Christmas πŸ₯°
Γ—
Γ—
  • Create New...