3D-Pangel

This is really impressive work. A couple of thoughts: 1) Some basic road design principles and/or design rules should be considered. For example, there has to be a minimum radius of a round-about for cars exiting a highway at a certain speed or a minimum length of off-ramp that allows a car to stop safely when coming off a highway. I tried to find a highway design rule book but while the internet has a lot of links, I could not find a simple list of design principles as a reference. But I did find this site from the US Highway Department on "Geometric Design". It just has lots of links to other documents. So, while nothing straight forward, it may be a good place to start. 2) Banking. You need to allow for the road to have a slight bank when making a turn --- especially if it is to be a highway with cars traveling at high speeds. 3) Libraries. While the freedom to create "anything" is great, consider pulling in and connecting pre-defined libraries of intersections, exit ramps, entrance ramps, merges, one-way, 2-way, 4-lane, 3 lane, etc pre-built assets. Allow the user to build their own libraries with the tools you have already developed. Here is where you could incorporate the necessary "banking" of a road or the highway design rules (if you can find them) as mentioned in 1 and 2. 4) You need to allow for the integration into the plugin "Real Traffic" which also starts with a spline. These two plugins together would be an outstanding combination. Real Traffic needs to have things defined to it as stops, turns, merges, etc. so it would be cool that as you "build" the road from you pre-defined assets (see Item 3), those assets also have the traffic rules built into them to get Real Traffic to simulate accurate traffic patterns. Dave

Relative to the actor and writer's strike, then yes for US movies it will be a lean 2024 summer season for VFX in the big blockbuster movies simply because live action production had to shut down. But with that said, it will NOT be that bad for feature length CG animations because those movies are planned well in advance with voice acting talents locked down ahead of time. Not sure what the ratio is of reshoots (or re-rather rewrites) for a CG animation is relative to a live action movie but I would have to imagine it is a lot less. Therefore, I have a higher expectation for character animation in a 100% CG movie to be less impacted than for VFX animation in a live action film. So, anything from Marvel (given that their process depends on re-shoots as they actually budget for it more than most other companies) will be impacted. ....and of course, anything with a short development timeline (like advertising and TV) will be less impacted that anything with a longer development timeline (like movies, etc). Plus, there is streaming which can tap into global content not impacted by the strikes in the US. The only big summer 2024 summer blockbuster I would bet on would be MI: Dead Reckoning Part II simply because live action finished on that prior to the strike. Dave As an aside and not intended to derail this thread: Is lack of Marvel content during the summer of 2024 necessarily a bad thing? I only mention that given how poorly MCU's recent movies have been doing along with the rest of the movies coming from Disney: $1B loss in 2023 for all Disney films across all of Lucasfilm, Marvel and Pixar with bigger losses in Disney+ AND nothing to base a big Christmas 2023 toy campaign on. These missteps do not just impact movie revenue. The bigger golden goose getting killed by making movies no one wants to see is the toy market driven by these IP's. In fact, the initial strategy of Disney buying Lucasfilm and Marvel was because Disney (while strong in the toy market for girls with all their Disney Princesses) had no strong presence in the toy market for boys. So, they give a Star Wars reboot in 2015 with very little Luke and lots of Rey and kill Han off right away. Sorry, I just don't see a big draw for little boys to play with Rey action characters. What were they thinking? Also, word is that when the Mandalorian Season 2 ended with Grogu going off with Luke Skywalker, Hasbro was pumped!!! Now there was a toy merchandising opportunity: Luke training Grogu!!!! Pure gold. Nope, killed that quickly in the "Book of Bobba Fett". Hasbro was pissed as they pay a premium to Disney for the license to sell those toys and they have nothing to sell. On the Marvel side, they killed off Iron Man, turned Captain America into an old man, made She-Hulk stronger and more in control than The Hulk in "She Hulk", and made Thor look like a buffoon in "Thor Love and Thunder" in comparison to Lady Thor. So not much useable IP left in the toy market for boys from Marvel. I think Disney forgot their market and the simple fact that young boys are more likely to want to live vicariously through strong male oriented hero action characters. If that was not case, then Mulan would be selling like hot cakes into that market. Again, what were they thinking? But getting back to movies.... Anyone pumped up for the "Marvels"? Yeah....my point exactly. This movie is projected to tank badly (based on pre-ticket sales and theater bookings) well below the worst performing MCU movie which was The Hulk in 2008 ($264M worldwide gross). This probably explains why Kevin Fiege is already "soft" announcing a re-boot of the MCU ahead of "The Marvel's" release in November. These execs are realizing that they killed the golden goose in Phase 4 (also known as the "M-She-U"). I guess Disney stock price hitting a 10-year low can be a bit of a wake-up call. Well, that is what happens when you lose sight of your market. Not sure that is possible with SW or any other Lucasfilm property given current management. So how do you make a million-dollar franchise? Easy. Buy a billion-dollar franchise and give it to Kathleen Kennedy.

Regardless this is what I heard: Are you sure that isn't you on the medical gurney? 😊 Dave

Nemetschek is the parent company of Maxon. Nemetschek's main businesses are in Architecture, Engineering and Construction software programs which makes up about 87% of the annual revenue ($800 M Euro). But Nemetschek does reference a "media" group in their financial statements. When you look at how Nemetschek lists their brands, you find that Maxon is the ONLY brand in that group: So, it is really easy to follow how Maxon is doing in Nemetschek's annual financial statements because all you have to do is search on "media". Now remember that Maxon is more than just C4D. So you while you can figure out how well Maxon is doing in Nemetschek, you can't determine how well C4D is doing in Maxon. But a rising tide lifts all boats, and Maxon's tide is rising. So I am not worried about the future of C4D. Solid program with a smart architecture in a strong company. You can learn more here Dave P.S. One thing you do hear in Nemetschek annual report is the move to "cloud" based computing. They are not saying that specifically about any of their brands other than the opportunity it represents. Something to keep your eye on going forward.

Honestly folks....Maxon is doing EXTREMELY WELL! This is not based on speculation but rather an analysis of Nemetschek's 2022 annual report (released in March 23, 2023). I have posted these numbers before, but the updated 2022 numbers for their "Media" segment (of which Maxon is the only member) are pretty impressive: This may explain the brand new 2024 BMW 8 Series that Srek is driving around in!!!! The man is just rolling in cash!!!😊 Plus when you read the various sections of the financial report, the outlook for their "Media" segment is nothing but roses and raindrops in comparison to their core business (AEC/O or Architecture, Engineering and Construction). A few choice excerpts are below:

I do have a semi-related question to the educators on this thread: What is the direction for DCC education given the insurgence of such AI based tools as Stable Diffusion, Dalle, Picsi AI, Insight Face AI, the list goes on. On top of all this is Unreal Engine with its free standard license that has modules for doing pretty much everything from fluids to facial animation via their Metahuman model. So, what are students looking to do digitally? Make digital assets (models, shaders, etc) or tell stories? I only offer those two choices because the entire DCC pipeline is extremely fungible right now with the advent of AI (just look at Wonder Studio). In short, are the days of the "3D generalist" becoming short lived simply because new tools are coming along that have automated vast portions of that pipeline? So what "career" guidance would you give to a freshman student wanting to learn 3D? Given the pace of technological development and the growth of AI, do students face the risk that what they have learned will be negatively impacted by changes to the entire DCC pipeline which could be made before they graduate? Thorny issues to be sure and (IMHO) far more perplexing than a low cost of entry to a "traditional" DCC application. Honestly.....my advice would be to learn "3D Theory" first. That is focus on the core concepts behind digital content creation independent of any application. Go deep on knowing the basics (e.g. topology, color space, studio lighting, rendering concepts, camera work) right down to the algorithms behind some of the most common 3D techniques. They will need this base to successfully navigate the constantly shifting CG landscape. Then I would steer the student into learning how to create the assets (e.g., modeling, retopology, and texturing) simply because whatever direction the technology moves in you will need a 3D asset. So far, they have yet to create a version of ChatGPT that creates models (though they are trying, and the results are laughably bad). Just my 2 cents. Dave

Your animations are getting better, and you are definitely developing a unique style. I loved the handheld camera work and the black and white photography. Given the gothic horror themes, I think black and white animation works best but there were at times it made everything appear a bit too flat. Subtle shading, soft shadows and strategic uses of color could really add a lot to the story. Or approach your next work with the mindset that it is being filmed in color but it is just a very black and white world that you are shooting in. Weird concept, but the best example I can think of is from an old animation called "Eroica" made back in 1998 on 3DS Max: Note the blue to the fog, the red in the characters eyes. A slight blue key light from the moon. But for most of everything else (until the end scenes), it is very black and white. Dave P.S. I have to ask....those names on the tomb stone. Are they friends and family? If still alive, did they want to be a name on a tomb stone? I would catch hell from my family if I did that.

I only have the "Center Axis To" command in my layout so it hard to messy click "Center Object To". Now this is a rather random issue and tends to occur more with models converted from FBX files that were made up a bunch of separate objects that were then converted to one "object" in C4D. The Millenium Falcon model I worked on was a good example where I was kit bashing pieces of it together. So that piece I wanted needed to be split and disconnected from the original object but in doing that they picked up the Axis Center location of the original parent model. So for placement, scaling, etc. I needed to use the "Center Axis To" command but then ran into the issue of the object moving to the axis rather than vice versa as expected. I wonder if this could an FBX conversion issue as when you try to duplicate this issue with C4D primitives in various combinations of hierarchies (objects under objects, object under separate nulls), it all works as expected. When I find something, I will post it. Dave

Very informative. One other issue that I keep coming across is using the Axis Center command and rather than having the axis move to the center of the object, the object's center moves to the existing location of the axis. This happens regardless of the axis center command settings (or so I think) but only on rare occasions. Any ideas why? Thank you in advance!! Dave

I think Redshift 3.15.19 will improve CPU performance based on this comment in the MaxonOne release notes: Now, this only applies to CPU with 6 or more cores (12 threads) but I would imagine there are very few 4 core processors on the market these days. You can find out what version of Redshift you are running in the Preferences section. Dave

For me, a simple way of distinguishing between Xpresso and scene nodes/capsules when referring to their strengths and the best way to apply them to your workflow is as follows: Scene nodes/capsules gives you the tools to build the assets. Xpresso gives you the tools you need to help manipulate them. Now this is NOT 100% accurate and there are of course exceptions because there are always more ways to do the same thing in C4D. But viewing them this way helps me breakdown how to approach them when problem solving. Just my humble opinion. Dave

Excellent point and I never looked at it that way. I would imagine optimizing such a vastly complicated set of code was a hard slog and filled with frustration because there are so many connection points to other major code functions that the developers were faced with unintended consequences each time they changed something. All of this also points to the architecture of the program and the methodical way Maxon connected features to that new core over each release while "still" implementing new features....further increasing the programs complexity and interconnections. I get a headache just thinking about it. To use a construction analogy, they have essentially been rebuilding the foundation while still adding new floors on an ever-rising tower. That is no easy trick. Now...does this mean a rapid increase in new features and/or the much-needed revamp of aging portions of the program like TP and BP? I don't want to say that because it will lead to over-speculation and increased expectations. What I will say is that Maxon program management has obviously been pretty methodical over the years. It had to be given what they have accomplished. Most likely, that methodical approach will continue. Also.....capsules are amazing. I could not see the benefit of Neutron in its initial release and the back and forth on how to implement its capabilities was a bit confusing. But they nailed it with Capsules. Each is a non-destructive mini-plugin that provides more flexibility than any single plugin could ever provide. I could not have created some of the things posted on this forum without them and they certainly have come "into-their-own" in the past year. So, while they seem like such small things in the release notes, we do forget what each brings to the table in terms of flexibility and creative capability. Dave

Interesting comments in this thread. It appears that the thirst for new features outweighs the thirst for performance improvements. Now, no disagreement from me that both are great but let's not understate the effort required to re-engineer and optimize the underlying foundation of a program to improve its performance while still trying to add new stuff to it. Kind of like mounting new tires while still maintaining your lead in the car race. It can be done (we've all seen pit crews) but requires a lot of planning or else it will result in disaster. So let's give some credit to the Maxon developers for attempting to do both and pulling it off. I am actually quite impressed with the performance improvements, but I look at it a different way. Rather than ask "what new features did I NOT get because they were focusing on performance improvements?" I would ask "how much of a hardware upgrade would I need to make to my PC to get the same level performance improvements using R2023". If those hardware upgrades cost you more than the annual subscription price than consider yourself ahead of the game because you not only were saved the cost of that hardware upgrade with this release, but you did get some new tools as well. Just a thought, Dave

Kind of surprised to see R2024 not announced on the main Core4D page. Did I miss it? .....well...with the expansion to Houdini and Blender threads on the site and the loss of "C4D" in the sites title, I guess Maxon releases are not big news anymore. Even though I understand why...it is kind of sad actually. Dave

Many thanks to HappyPolygon for these links. Who would have thought that a Sketchfab model would could be so impressive. They offered it in an FBX format as well and so I downloaded it and went to work. While not modeled in quads with any polygon flow to consider what-so-ever, the modeling was solid with excellent UV's. There was very little corrupted geometry for me to correct. I think it was game rip, so there were some details that were just not adequate as textures and needed to be replaced. The Engine intakes and radar dish needed to be replaced. Likewise, there was no landing gear and the corruption of the geometry around the landing bays was pretty bad so that needed to be replaced as well. Landing gear rigging was added. Engine glow was added. Interior cockpit lights were added. The whole model was converted to Redshift. And now the finished model: Very happy to get that monkey off my back. Now I can start on the DS tunnel. Dave

Screw what the audience sees, whichever camera was "live" is where they switched the background to match that camera's position. Also, I agree that while the size of the screen and its "projection" method (front, back or self...meaning LED) is not defined, to do this all-in post is a bit of a challenge on a weekly show unless it is REALLY more fake than we all believe. Fake audience reactions, fake judge reactions, all of it. Plus, at some point particularly towards the end of the season, the acts start to appear on the same stage. None of that can be determined ahead of time because no one knows who those acts will be. So is it THAT FAKE (predetermined winners and losers) just because we can't accept the size of the screen? We accept real time rendering but we can't accept a big screen? Really? Dave

Well...this video required a little inspection. Personally, I felt it was a little scripted...from the camera cuts (some appeared carefully framed) to the back and forth between the judges and the animation. Not saying that the whole thing lacked a "live on stage" believability, but something just did ring not as true. Now....how did they do it? Most likely Unreal Engine including the new hair rendering capability that came out early this year. Notice that the dog had a very binary movement to his mouth for the lip sync. Open or closed. That is all you got. While the body had deformation that was built into the rigging, the lips not so much. So while there are really only 8 basic mouth shapes used in animation, breaking down the "unscripted" dialogue into those shapes and morphing between them in real time may have been a "bridge too far" for real time animation. What the audience saw? While it was apparent that there was a screen on the stage that replicated the background of the real stage. The screen extended from the floor to beyond the ceiling. The stage design also helped because there were lots of vertical columns on the sides to hide the edges of screen and make the blends better regardless of camera angle. Now, if this was in fact live, then you would need multiple renderings going on at the same time to match the angles from each of cameras recording the performance. This was not a cheap performance. To cut down on the computational complexity, there was only one moving camera and that shot was for a very short time. It could have been added later for the TV broadcast. Note also the final shot of the dog walking off the stage. They cut it right before he broke the edge of the screen and cut back to the judges. All in all, great planning and quite the investment in time and equipment. I would imagine that this was a professional VFX house specializing in CG for stage productions who put this together. Even if they don't win, it will provide great PR. Dave

The Sketchfab model comes in FBX and while i have not dived into it too closely, there modeling seems pretty solid (unfortunately, very non-quad, no edge flow, etc) but other than complex poles and boundary edges as all the detail is just slapped on...it looks pretty good. The only downside is that the UV mapping is a little off and UV re-mapping is my weakest skill. The big issue for me with the original model was that it would render incorrectly in Redshift which I attributed to the incorrect normals and corrupted geometry. It renders fine in Physical Render. Let me see how the Sketchfab model renders in Redshift. I want to use Redshift given the polygon size of the environments I will be putting it in (the outer Death Star). Also, I want to start working on the Death Star tunnel to the reactor chamber which will also be pretty heavy on geometry and lights and definitely in need of Redshift. I should be done in 2035 at this rate. Dave

I have been running mesh checking but upon deeper investigation found that there are more boundary edges than expected and when you zoom in you find a duplicate cylinder just under the exterior cylinder by an amount equal to the thickness of the hull plating. It was as if the original modeler conducted a extrude on the faces but left the original faces intact (Lightwave must be okay with non-manifold surfaces). But not sure why that would cause a texture error when you reverse the normals on the outer layer. More work is required. Lots more. Ugh. If anyone knows where I can buy really good C4D model of the Millenium Falcon, let me know. Sometimes as hard as you try to restore something, the level of rot is overwhelming. Dave

Some people restore old cars, I restore old models from other formats. One model I have been working on is: This model was originally done by Andy Crook in Lightwave and was graciously donated to the site Sci-Fi 3D for free download. The conversion to C4D was not easy as about 20% of the geometry was corrupted/or missing and/or needed to be redone. Overall, I reduced the polygon count by almost 37,000 polygons. I then rigged the landing gear, boarding ramp and engine power. But one problem still remains that I have not been able to solve: Note that with reversed polygons, it renders well under flat lighting though when you get to shinning a light on it with shadows in Redshift you get back-face culling issues. Reversing the normals gives you this: I have no idea how to resolve this. Dave

Some names to consider: Clinton Jones (pwnisher on YouTube) - Founder of the 3D Challenges on YouTube and an amazing C4D Artist. Niko Pueringer and/or Sam Gorski: Founders of Corridor Digital. At one time they were both C4D users but have since moved on to Unreal Engine as well as many of the cutting-edge AI based apps (eg. Stable Diffusion, etc) in their quest to push the boundaries of making amazing digital streaming content on an indie budget. Their ability to innovate in this space and their views on AI's impact on CG art, the artists and the industry would be very interesting. Chris Schmidt (RocketLasso) - Great educator and master "nodesman" (not to be confused with Noseman). Just hearing his story from a GSG employee to a major independent C4D influencer would be fascinating. Andrew Price: Founder of Blender Guru and Poliigon Textures/Models. Honestly, anyone thinking about moving to Blender probably starts their journey at Blender Guru. His beginner's tutorial on making a donut with sprinkles is to Blender what Nigel Doyle's little blue jet was to C4D. What started as a hobby has now become two major elements of the Blender eco-system. Nigel Doyle - So how is life in the modo-verse? What has he been up to in the past couple of years (has it been a decade yet? --- Feels like it) and is 3D still a part that life. While he has been gone, I really don't think he will ever be forgotten --- that is a testament to the impact he made. Nick Seavert (JangaFX Founder) - Visionary, passionate developer of 3D software. Honestly, he is (IMHO) the "Steve Jobs" of the 3D software industry. Unfortunately, the addressable market for 3D software is a lot less than consumer electronics but I don't think that matters to Nick. Listening to him is always a treat as he tells it like it is. David O'Reilly and/or Mike Batchelor - CEO and CPO respectively of Insydium. Now that Nexus is fully rolling out what is next for Fused? What is next for X-Particles? Where do they see major threats or opportunities in the ever-changing VFX software industry? GET TO KNOW OUR MODERATORS - I would have to imagine juggling a day job to pay the bills and being a C4D moderator probably leaves very little time for things outside of C4D or their own personal projects. They've influenced many with their help and with their artwork but who or what influences them? Dave

While not evident in this example, I also find problems if there are manifold surfaces and unoptimized points (e.g., two points in close to the same space). So run the mesh checker and look for any errors, fix them and then try again. Dave

First off --- Many thanks to Happypolygon for the continual updates. This is becoming an amazing resource and one that I hope will keep going. Probably should be its own sub-Forum (IMHO). Also, I am rather impressed by Insydium's Fused - in particular the work done with Nexus. The foam shader examples look really good and the wave modifier looks both easy and powerful. The upscaling too across all simulations too is rather impressive. Now, I know that the Houdini faithful will have something to say but before you do, think about capability vs. learning curve. True, Houdini is more powerful and that is why no major VFX production studio is banking their next movie production on Insydium. But for the clod-kicking, knuckle-dragging hobbyist crowd like myself who have a day job, I find it very approachable and capable of delivering amazing results without requiring huge sacrifices in time or patience. Starting from ground zero (knowing nothing about either program) I am not sure I can get to the same level of quality with Houdini in the same amount of time and effort required by Fused. Just saying!!! Dave

Been thinking about subscribing to Octane. $267 USD/year is pretty reasonable and (imagine my surprise) about equal to Redshifts annual subscription cost of $264 USD/year. What makes Octane attractive is its ease of use, but then again my biggest barrier to mastering any node based program is being able to understand which node to connect next given all the options. That flexibility is what makes nodal programs so powerful, but it also complicates the workflow (IMHO) and you fall into "analysis paralysis". So the ability to Octane to be context sensitive at the node connection level is very inviting for experimentation and therefore learning. In comparison, my original workflow for Redshift was to first start with what I know best, the Physical Renderer, and then convert to Redshift. That gave me the bones of what I want to achieve as a starting point. But I have to say that conversion is very crappy. So much is left out and the reflectivity was way off so it was a complete waste of time. Nevertheless, I would still have to agree that it does take a lot of work get Redshift to look right. There really are too many settings. I have yet to play with Octane so I have no idea of just how much easier it is to get acceptable results. But my journey with Redshift has been a struggle so anything that is easier and just as fast is very attractive to me!!! A word about the free models, plugins, etc that come with Octane. First off, Embergen is a huge draw for going with an Octane subscription. But I fear that will be coming to an end. As Embergen (and the soon to be released Liquidgen) has really taken off in popularity, JangaFX no longer needs that relationship with Otoy and may be pulling out of it. So keep an eye on that. Plus Kitbash3D is no longer part the deal as well. I think they pulled out ever since Kitbash3D launched their own subscription service called Cargo. I originally was very anti software subscriptions but have changed my opinion because the companies behind them (like Otoy and Maxon) are continually improving the software. Not so much a fan of model subscriptions where I can't use that model anymore once the subscription ends and it is not like they are improving the models every year. Cargo has been launched for a while now and they finally included their first "Cargo exclusive" model called Police Station. Looks cool...but would I use it? Probably not. Dave