hvanderwegen

@bezo What CPU are you rendering with? Is that the original render result? I ask, because there are major anti-aliasing issues with that render.

Actually, they can. But missing is the third variable: money, and the fourth one: skill. From my own experience, GPU rendering is brilliant. It revived my interest in 3d rendering once more (I solely focused on modeling before until a few years ago), and being able to render scenes like these (see below) in a minute or so at great quality and high resolution as well have almost real-time feedback in the viewport while working on lighting and materials... It's a breath of fresh air. It's so incredibly liberating! Another option is to learn Unreal or Eevee (Blender) which rely on game-type rendering techniques, and still produce excellent results, with real-time viewport feedback. Unreal is free. As is Eevee (gotta work in Blender then, though). But yeah, it does require a reasonable GPU, which are hard to come by at reasonable prices lately. I was very lucky to come across a 3080TI five weeks ago to replace my 1080GTX (which also rendered quite fast, btw). Then again, the one thing we cannot get back is time. And it's loads of fun now to render scenes in minutes or even seconds that took hours before. Adjusting the scene lighting feels more like being in a photo studio: things update in seconds in the viewport while playing around. Have you considered using Unreal for your animation shorts? https://www.schoolofmotion.com/blog/how-to-export-from-cinema-4d-into-unreal-engine You also get access to many free materials, objects, and textures for your personal projects if you use Unreal. https://www.unrealengine.com/en-US/bridge Out-of-the-box 1500x2000 full quality GI render produced in K-Cycles, 1:15 minute, 3080TI 12GB, Ryzen 3900X, 64GB

E-Cycles improvements are slowly making their way into standard Cycles as well. Yes, the light linking was part of the old internal render engine, but for some reason it never made it in Cycles. Light linking seems to be on the CyclesX roadmap now, however,

Finally, light linking for Cycles users. Combined with the rather awesome light groups implementation in E-Cycles, and the faster renderer, well... What's not to like? (Except for that it's not free 😉 A must for anyone working in Architectural/environmental rendering and Blender. It saves so much time. https://ecycles.gumroad.com/l/E-Cycles/launch https://blendermarket.com/products/e-cycles https://www.youtube.com/watch?v=5sBHOvHTq_g And the new Animation denoising.

That seems to be the case: there should be one vertex at the center, but in the converted mesh there are 64 individual verts! No wonder that the Lathe introduced phong edge breaks, because the angles for those 8 edges, even though they lie as good as on the same plane, might cause angles that the Lathe code interprets as >30/60 degree ones...

I exported as an OBJ file and opened it in Blender for inspection. As I expected, the normals are split at the edges: When I cleared those sharp split edge normals, there was still an issue with the normals. I selected all the top tris, and averaged the normals, which resolves the issue: If you open the Phong tab in the Lathe object, turn off the Angle Limit or adjust the Phong Angle to remove that cross. Still, it looks like a bug, because C4D should not be introducing edge breaks on flat lathed surface. Also, the back has a normal issue as well. I averaged those face normals in Blender, and they are no longer an issue in C4D. Note that if you increase the Phong Angle over 90 degrees, that issue also goes away in C4D.

Wow, where did that come from? I never claimed that, nor do you have any notion of who I was at that age. I began creating stop-motion animations at 11/12 years old when I got an inexpensive super 8 camera for Christmas. I converted our attic to a moon surface with DIY studio lighting setups to re-enact the moon landing and inspired by reruns of Space 1999 scenes on telly. I went to the local tiny library to find the odd book on cinematography and lighting. I blew up plastic models of space ships! I was doing wireframe renders of simple 3d objects on an Amstrad at the age of 13/14 by manually coding them in basic - self taught. In 1987 at 15/16 I created my first 3d renders and spaceship animations on the Amiga and completed my first commercial freelance work. Same for friends of mine (we were all part of the European hacker scene in the 80s and 90s). One of my friends was coding visual effects and games in pure assembly language on the C64 at the age of 12 and went on to work on AAA PlayStation titles later. Others in my circle of friends were accomplishing stuff that you would not expect at their age: one was an accomplished C64 and Amiga musician, and his career got him to work for Electronic Arts and became one of the more well-known game music composers/musicians. Another had an innate feel for business strategy, which landed him deals at 16-17 years old to convert commercial games like Lemmings for the C64. And so on. I teach young adults in design, and you wouldn't believe some of the quality of the work that they produced years before entering the program. Those people are not prodigies, but spent time and effort to learn to become good artists even at a young age. Don't tell me or others what they could have achieved at that age. We "first-gens" in computing and digital art were accomplishing things at a young age that were pretty amazing seeing the technical limitations. Besides, the available tutorials and pools of online knowledge are VAST compared to the 80s where we all had to figure stuff out by ourselves and the Internet didn't exist. Nor the available free quality assets and high-level user-friendly software. Just because someone is young of age doesn't mean they can't accomplish wonderful things if they set their minds to it. You don't have to be a prodigy to do so. Hard work, practice, persistence and dedication are the key elements. By stating someone is a 'prodigy' at that age just because they produced impressive art or accomplished a great feat, you basically ignore and downplay all their efforts and time that they put into learning that stuff and their persistence in finishing such a project. Perhaps I read too much in your comment, but I find it abrasive and rather condescending. And down-playing the hard work of many young artists and their future older selves. "Dude"

Yes, ever so slightly intimidating that this was produced by a 14 year old teen in his free time. When I look back at my design work from that age - well, at least I was producing pixel art animations in Deluxe Paint and doing simple wireframe animations in Sculpt Animate 🙂 To think what we could have done with the tools that are now available to young artists and for free to boot! Very cool.

Open an Explorer window. Open a second explorer window. And a third one. Position these and scale them so that each has a unique position and size. Now close all three, and open one of the other two windows that were closed before the last one. Result: that window opens in the same position and size as the last closed window. I don't want that: I prefer to have each window to be opened where I left it last time, which is why I use a window manager tool to do this for me. But I read Windows 11 now behaves properly too, just like the Mac or Linux windows managers.

Is it true that Windows 11 FINALLY remembers the position and settings of each Explorer window and the position/size of app windows? That alone would be worth the upgrade.

A 14 year old 3d artist made this. In two months time.

@3D-Pangel That's actually an excellent notion - will keep that golden tip in mind when I get to design my own home - if ever the chance comes along. 😛

As a conceptual tool and communicate your ideas to an architect, sure: not a problem. But the actual drawings will have to be done and finished in a CAD tool and by someone understanding the structural limits, building components, and the building laws.

Well, it IS meant as an 80s inspired Halloween story 😉

The new Open Movie called Sprite Fright is out! Get all the production files on the Blender Cloud. https://studio.blender.org/films/sprite-fright/

If you are a new student of Cinema4D, you should have access to the latest student version (25) which also includes Redshift. 🙂 To answer your question: no, a simple scene like that should take a few seconds to resolve to a good looking render on a 3090. I have a 3080TI, and a similar scene renders in a few seconds in K-Cycles. The viewport renders it at near real-time, so your settings are probably wrong. I did not have time to test in Redshift - only have access to that at work. Your scene takes 30 seconds to render on my home system (Ryzen 3900X + 64GB)

Oh. My. So many nice tools in there. This would have saved me hours last week. @Igor Are these usually available by default in the Network node view? Or do these have to be activated? Because I did not have access to these before I installed them.

True, but for some reason the tools weren't there. It took a minute to copy the python commands, and execute them. Then a reboot of Houdini, and a new SideFXLabs desktop becomes available. I also tried finding the shelf, but clicked the wrong thing... Anyway, it's solved now. Thanks for the heads-up.

Ah, I got it! Installed via the python shell. (method 2 https://github.com/sideeffects/SideFXLabs) Some interesting tools in there. I wonder why my instructor hasn't mentioned these.

@Igor Where can I find those Labs nodes? I checked the available desktops, but it doesn't seem to be there.

PS there is also the Dissolve node, but when I tried it, it would not work. After some research, it seems the surrounding geometry must be blasted first for it to work properly. Or something. Like I said, it's a toolbox. Pros and cons.

That's the trouble (or advantage depending on your viewpoint) of Houdini: it's a toolbox, but often modeling is just simpler, more efficient, and more direct and user-friendly in other DCCs. Which is the reason many modelers just use other apps to create models, and then import those in Houdini for effects and sims. After dealing with Houdini modeling for a few weeks now, I tend to agree with this approach. Unless my goal is to create a fully parametric uber-object, I will stick with Blender for most modeling. That said, here is a SOP tool that will do what you want: https://orbolt.com/asset/Houdini_Zero_Tom::curvegenerator

From experience I never update the OS until: [A] the new OS has received a year's worth of patches and updates; [B] essential software that I need requires the newer OS. Or: [C] I have to (re-)install an OS for whichever reason and I must start from scratch. [A] and [B] go hand in hand generally.

In the current version (2.93) Nodes can be appended and linked like any other asset. If you have a nice geometry noodle in one file, it may be appended or linked in a different file and reused. In the upcoming v3 Geometry Nodes have seen quite a few changes and many new additions. It is possible to mark geo noodles as assets which are then exposed in the asset browser. These can then be reused in the geo node editor, or even dragged into the node editor as pre-built groups. So presets are possible, but not as flexible (yet) as C4D's capsules with exposed parameters and usable in the outliner or Houdini. It is necessary still to keep working in the geo nodes editor in Blender. But having said that: yes, it is possible to save presets in the upcoming v3 and it is already possible to append/link geo-noodles from other files saving the user from having to redo it all.

Any reason why you are not using Geometry Nodes Point Instancing? It works with lights too.