3D-Pangel

The Dock doors start to take shape.....finally getting something that I am happy with. At this rate, I should be done with this project by next Christmas...but I am having a blast and looking forward to every challenge and utter frustration along the way. Dave

Thank you! From you that is high praise! One thing for sure....if I don't get what I want, I keep trying. Given that this is a hobby, I can only eek out about 4 hours a week by the time the rest of work and life is taken care of. So those little dots represent about 2 weeks of time (or 8 hours of experimentation). Today is a rather unique day....I killed back hauling dirt yesterday (dead lifted a 150 to 200 lb sack over my shoulder.....I thought I could do it...I did...but I paid a price). So I hooked up a heating pad to my office chair and have been working on C4D all day. I am now working on the space dock doors. Three attempts today so far. Just not happy with it. I am using a design by Robert Wilde for inspiration. The inner door break on his design has some pretty neat geometry but I am struggling with the topological flow. To many cuts and then he indents and bevels the inner door "C" glow. If not done right, you get a lot of crappy triangles and phong errors. Here is what I have so far: But this is what I am going for:

My texture journey continues.....I was never happy with the black/white mask I had created for the tiny ship windows that I was getting from JSP. Not that JSP is a bad tool, it isn't. It's actually a must have in my opinion. But while there is some control of the random patterns it creates by creating vertical or horizontal sections, it places those sections randomly across the image. I was looking for something unique in that I wanted to lights to reflect floor levels in the space station. They needed to line up and be restricted to individual rows of pixels in the final image. Not easy to do with a random texture generator. So what I did was to create a plane that was roughly 184 x 324 polygon is size. I then textured it completely black and colored the 4 polygons in each corner completely red. I placed a camera above the exact center of the plane and had it facing it straight down. I set the render size to the same ratio as the number of polygon and did a few quick renders while repositioning the camera to insure that the red corners were exactly at the corners of the rendered image. Next was to go in and do a few loop selections to select the floor levels I was looking for and then remove polygons from each of those selections to get the randomness that I was looking for. I will admit that this part was tedious. When done, I painted the finished selections with a flat white material to generate the black/white mask of windows I was looking for. Once rendered, I then painted the red alignment squares in the corners black. The finished image is below: And this is how it makes the windows appear in the final render. It definitely sells the concept of floors within the space station.

Agreed. Scale is the hardest challenge and as I was not happy with the lights either, I took another whack at it. This time, I actually have the lights as a different texture that just uses the alpha and luminance channels. So when the lights are not part of the hull texture, then I can make independent scale adjustments to both. The problems I was facing is that as you decrease the scale on the lights, you can see a repeat to the pattern and it just looks like a virtual sea of random dots without an order to it. I really want to see the lights ordered along levels rather than a random pattern. JSP helped with that but at a scale value of 10 still ends up looking like a random sea of dots even when you lower the "Drawing Chance" settings to 1.9% (which is pretty low). I wish it had the capability for minimum as well as maximum pattern length as too many single dot lights add to the randomness when your expectation is for order as this is a man-made structure with floors, etc. So what I did was create two patterns using JSP -- one that was just vertical and one that was just horizontal lights with a minimum brightness setting of 0. This still created the random sea of dots but when you bring each image into an image editing program and decrease the color depth to 2 colors (0 or 255), it drops all dots at 128 or less to 0 and thus thins out the pattern a bit. As I am using this as a mask, I really don't care if the color depth is just black or white. I then selectively started to clone the vertical and horizontal patterns I liked to a new image until I started to get the pseudo man-made pattern I was looking for. Once completed, I copied it and renamed the original to "Lights" and the copy to "Less lights". I then opened up "Less Lights" in an image editing program and thinned it out some more with a black paint brush so that there were areas of no lights. Finally, as the last step in an insane bid to have as much control as possible, I made individual loop selections for each level of the doom (15 in all). On each one, I dropped either the original lights texture or the "less lights" texture and adjusted the offsets individually to insure that nothing looked repetitive. Now that I had that level of control, I could increase the number of tiles on each level until I got the sense of scale I was looking for. Finally, I added a bit of glow to the light textures as well. Interestingly enough, the goal never really showed up until I started to map textures to individual selections....not sure why. Here is a close up showing the glow: To answer your other question about modeling scale, I am matching a scale used by Alex 3D as he does have some outstanding Star Trek models which are available for download. He is also working on his own version of the Space Station for STIII. It looks like he stopped working on it for some time his WIP page is dated 2013. Once I get this to a point that starts to look pretty interesting, I will be reaching out to him. Dave

Jay, I did checkout JSPlacement some time ago, but it was a little glitchy on my old Win 7 machine. I tried it again on my newer Win10 machine and it works well. I have been doing some texture tests on the top dome using maps for specular and luminance from JSPlacement. The base texture is an old deathstar texture that was originally much darker but I passed it through the Colorizer shader (I love that shader) to lighten it up a bit. Here are the results of my testing: Getting there! Thanks for your help! Dave

Wow.....I wonder how long it has been full as there were no alerts from the forum software....or any that I can remember. I had messages dating back to 2013 so it is very likely I did get some warnings but completely ignored them. I cleaned a few out so now it is 75% full. Thank you for letting me know! Dave

Finally found some time to get back to this WIP and make some progress. As C4D is a hobby, I do bounce around a lot with my projects -- trying, experimenting, learning, but never completing. I wanted one big project for this summer because all that jumping around was not giving a sense of satisfaction of accomplishing something. Here is some of the mesh details: The model is huge because it will be seen with Star Trek ship models as well. The works of Robert Wilde is the inspiration for this and I have over 66 reference images found from searching the web. I finally got the shape/dimensions/proportions correct because rarely are there any orthogonal views to work with for every element of this model. Now that the basic shape is done, next is to insure that I have the just the right texture with the appropriate specular highlights. Here is an example of what I hope to achieve (for me it is a lot harder than it looks): Any ideas on how to get this affect? Thanks, Dave

Not sure if anyone from GSG pays attention to the Cafe threads, but if they do then my recommendation is for them to consider what Helloluxx charges for tutorials. Based upon what I have read, the Hellolux tutorials are also well presented and very informative but are priced about a third of what GSG is charging. Any XP4 tutorials planned from Helloluxx? Dave

Okay.....for someone who is inexperienced, you picked one of the more challenging subjects to model, light and texture and that is automobiles....and you did a great job with it. If you had said that you were extremely experienced with 3D modeling, I would have no reason to doubt you based on this work. Also, you have drawn the attention and respect of both Vector and Cerbera in this thread, two of the best modelers I have ever seen -- so don't let that go unnoticed. So please feel good about what you have done because you have some serious skills. Relative to when you hit a wall or not....I have found that talent will always beat out native tools or external resources (plugins, textures, 3rd party renderers). What it really comes down to is time and your creativity. There are always clever work-arounds to most limitations of any software package, the question though is what makes sense for you, your business, and/or your client. For example, you could position 200 lights manually to get some pretty good ambient lighting effects and fast renders or just turn on GI --- each approach costs you something and the question is whether or not you are willing to pay those costs (time, money, mental health)....there is no right answer. It all comes down to whether or not you are getting the render you want or need (eg. for a client). In fact, the only pragmatic piece of advice I could give someone of your skill (which vastly exceeds my own....and I have been futzing with C4D for 10 years now as a hobbyist) is to study the real world with an extremely critical eye. Pick apart the shadows and the highlights found in nature. What is it about the subject be it a car, enironment or character, that makes it look real. And always remember that the best details in a WIP are those which you would ONLY notice if they weren't there. Dave

Greyscale Gorilla is offering a 30% discount on this course. It provides over 20 hours of training into XP with 12 hours on XP4 alone. Unfortunately, even with the 30% discount it is over $200. So has anyone taken this training? GSG site does not offer any sample chapters as you would find at Lynda.com so I have no idea on how effective is Jon Bosley as a teacher. I am pretty confident this guy knows his stuff, but is anyone familiar with how well he can teach? The other detraction is that it is available via streaming only. Sometimes that can be a problem. Plus, my use of the product is dependent on how long GSG sticks around. While I have no concerns about their future, $200 is still a lot money to bet on. Finally, I am most interested in XP4 new features and find the Ensydium tutorials pretty good. Unfortunately, XP4 fetarures are presented as an Appendix rather than project based lessons for XP3 found in this GSG tutorial. Not sure how that is different than just using the Ensydium tutorials. Hmmm....as I type this...I have come to the conclusion that it is not worth the money just based on my questions and concerns. Besides, cheaper alternatives are out there. I would still be interested if someone who took the course said the training literally blew his/her hair back and felt very strongly that there was nothing else better on the market at any price (which would probably be cheaper as this has to be the most expensive tutorial I have ever seen). Thanks

You know you have been using Cinema 4D for too long when you find yourself saying this: "Honey!!! You better get here quickly! The Rebend Deformer is finally kicking in!" .....sorry...I couldn't resist. Dave

Agree....excellent work and a real valuable resource. I hope you keep it maintained or get some compensation from somewhere to keep it maintained. I do have one issue though with the sample animations though....in particular Rebend Deformer: Really? Can we find another example shape? This is a family site after all! Dave

Almost as good as Black Friday is NABS and the sales that come with it. Toolfarm has quite an extensive list found here A couple of C4D related items to note: Pixel Lab - 30% off Video CoPilot - 25% off (some of their model packs come in C4D format) Curious Animal - 15% off HelloLux - 20% off all training Laubwerk - 10% off e-on: 30% off (includes Vue-xStream which works in C4D) Some of these sales run to 4/18/2018 but most expire on 4/14/2018. If you know of any others, please post them here (models, plugins, textures....you know the drill) Thanks, Dave

I thought the model was great until I saw the larger image you posted. Now it is phenomenal! The skin and lips are just perfect and the hair is amazing. Even his coat (with the seams and buttons) is wonderful. Best of all...the eyes are alive! Very tough to do. Well deserved exposure Dan! Congratulations. Dave P.S. Please keep us posted where this leads you because I do believe it will lead to some pretty tremendous opportunities. I wouldn't be surprised if Peter Dinklage himself has seen it.

When you look at the edits, camera angles and overall sheer amount of character animation, it must have been like putting a jigsaw puzzle together because you kept it visually interesting. The close-ups and overhead shots were well timed and gave it some punch. I can understand why it would be mentally exhausting putting this together and staying focused for so long. So how did you approach it? Did you act it out on video first and do a rough edit to figure out the timings and direction? One small nit though....please add motion blur. The platypus was making some pretty broad dramatic gestures and the strobing was noticeable. Dave

So that is how you have been spending your time! ;-) Awesome devices. Actually, I think it is the basis for a new form of Animusic. 3D-Kiwi would love this. Dave

I remember hearing some time ago (and I mean years) stories about Blender being crash prone or that commands did not operate as intended. I will admit that it formed a bias that stuck because (in my mind) it was open-source software after all. In actuality, as I think about it, I would imagine that because it is open source, Blender probably has a more thorough peer review process of new code than other private development groups out there. I wonder if the comments about commands not working as intended has more to do with the confusing UI than the program's stability. So my apologies for jumping to old conclusions. Now Blender 101 sounds like a very interesting way to test the waters without too much of a time commitment. I will definitely continue to watch that space. Dave

Along with the long over-due update to C4D's UV capabilities and nodal materials, then Cerbera's wish list for modeling tools, Briankoko's wishes for VFX tools and Rectro's well thought-out and comprehensive list pretty much sum-up my hopes for not just R20 but probably for a few more releases after that. Many would think that R18 and R19 put a good dent in the gap between C4D and the competition, but after reading this thread you realize just how much farther they have to go! But if I could take this in a different direction here is my ultimate and obviously unrealistic set of wishes: 1) MAXON gives us everything we have mentioned in this thread prior to 2020. Revamping the core has been the bottleneck to any progress in a number of aging areas (BP in particular) and it has been the long held belief that MAXON will not invest in improving any code wrapped around the old core. So as a community, C4D users have patiently waited as the industry has moved forward. Remember that MAXON has assuaged our impatience with statements about how this great new architecture will set us up for big changes in the future. There have also been insinuations that improvements will come faster than anything we have seen in the past. These promises have kept the truly faithful camped on the MAXON doorstep. So if the new core is finally here in its entirety (having taken 3 to 4 versions to fully release), then if you don't want to break that faith, you better deliver -- especially if you are raising MSA prices. I don't care how many releases it takes - but I do care how long it takes. The expectation you have established around the new core is that catch-up the competition will happen and it will happen fast -- how you program manage new feature releases needs to strongly respect how patient we have been. That is why my expectations are not for R20, but for 2020 and I don't care if takes you 8 releases to get there...but I do care how long it takes because I fear our patience is coming to an end. 2) Either MAXON meets our expectations or Blender makes its program easier to use. Honestly, there are some really neat areas of development within the Blender community and at a fraction of the cost of C4D. The total cost of ownership (Blender and plugins) is still far less than C4D Prime but you get so much more. MAXON needs to respect Blender and realize that Blender sets the floor for what users want in terms of features. We all hate the UI and the instability, but amazing things are still being done with that program. 3) MAXON stops relying on plugin or 3rd party developers to round out their tool set. Focusing on built-in pipelines to other programs like Houdini, Allegorithmic, etc is nice, but it is not a winning strategy for staying competitive. In fact, it is almost and admission that they can't stay competitive on features. We are not going to give you nodal materials, but you can use Substance Design and import the .sbsar file. We are not going to give you fluids or OpenVDB support, so please use Houdini and import the file. Now, it is okay to not compete on features but you have to compete on something and the only thing left is to compete on price. Sorry, but in a highly technical market with a strong development pace, competing on stability and ease of use will only get you so far. People will always want what your competitors have and will take stability for granted when they jump ship to another program to get it (and its not like Maya, Modo or Houdini are bad options----they are great options). They may regret it in the end, but what does that matter to you when you have lost the sale. So my wish is that MAXON commits to changing what C4D can NOT do rather than what C4D and its plugins can do. If they don't make this commitment, then they should lower their price. In summary, my hope is that everyone at MAXON is extremely nervous about over-promising and under-delivering with R20. Since R16, they have set the expectations that the new core will bring great change and while they never said that R20 will be the "big" one, the general consensus of the community is that R20 is when the big changes will yield BIG NEW FEATURES. Right, wrong or indifferent, it is MAXON's time to deliver on those promises. Dave

Congratulations. The trailer does not do it justice. I encourage others to view the rest of the short clips out there on Vimeo and DeCarlo animation to get a true sense of the heart of this story (I love the clips of the selfies with the alien) and how well it is directed (as evidenced in the first scene which does a good job of setting the tone). Was all the modeling, rigging and animation done by one person (Thomas DeCarlo). If so, that is quite an achievement. Keep us updated on how well you do at the festivals. Dave

Well, one of the more impressive elements of that animation are the sound effects and editing. Very professional and add huge amounts to the enjoyment of the film. But I don't think this is the right forum to discuss that (in fact, I hope you are posting this animation as well to the Final Cut Pro crowd....BTW: I just love the voice "Oh...this can't be good"). One thing I would probably want to learn more about is how to rig a model for rag doll physics as I am sure there are some constraints that need to be applied to insure that joints bend realistically. Was it a simple matter of just using hinges? Was it that simple? Also, while you go from key frame to physics simulation, can you actually combine the two? That is, have the arms key-framed to flay about and grab for things while the body is falling. Or the legs kick a bit after they hit the ground. Not sure if that is possible (eg is motion blending possible?). In short, just tell us everything! ;-) Dave

Just outstanding! I almost felt guilty enjoying it soo much (I mean, they are scenes of human based characters being squashed, chewed on and crushed after all. It reminds me of the first time I played with the Endorphin demo years ago....I had a sadistically good time whacking those little rag dolls around). But apart from the subject matter, the animation, music, editing and animations were just spot on impressive. I am saving that link as I anticipate many viewings in my future (oh my...what does that say about me?). I also watched Ragdolls 1 and 2. Loved those as well but Ragdolls 3 takes it to a whole new level. MAXON should put this on their home page with an article about how you made it. Just tremendous. Dave

A quick scene in Forestor --- so nice to use!!! As I was making it, I was wondering if 3D Quakers would be creating a new version of trees to use on their landscapes....You don't want to create a full tree as that could become too polygon intensive. Rather, what you want is some variant of multi-floral plant that looks like a full tree from a great distance and can then be painted onto their landscapes but is in actuality a miniature tree with a much lower polygon count than a standard tree. Just a thought. Dave

I love the work you are doing. For a number of reasons which I won't go into here, relative to pricing, one thing which I would like to direct your attention to is Poliigon Textures. You purchase credits either in blocks or on a monthly or annual subscription plan and can then use the credits for individual texture purchases. All textures and texture maps come in a variety of sizes and you are essentially using one credit for each texture/texture map. You want color, diffuse, normal then it is 3 credits total regardless of size. If you want the bump map as well, that is another credit (4 total). Now, I will admit that the web-page really makes this system sweet to use but it is rather powerful for the user because it is a GREAT way to search for textures and you are only paying for exactly what you need (meaning what you are going to immediately use). Personally, I would like to see more texture developers sell in this manner....but I know that the web-development is probably not trivial (and creating quality textures is hard enough). Dave

Don't hate me for pointing this out as the modeling work is nicely done and I don't want to detract from the effort you put into it....but I did find this rather humorous.

Wow....just wow. Charbel, I only have about 40 or 50 questions...I hope you don't mind ;-) But I will keep it down to a handful. Is that landscape procedurally generated ---- that is, the detail holds up the closer you get? Is the interface nodal based for creating erosion types. Will it use C4D sculpting tools? C4D seems to suffer with scenes that are object heavy more so than polygon heavy. How is the viewport performance with these terrains in R19. Any chance of posting a video just showing that object being manipulated in the viewport As these will be two different programs (Moonrocks and Forester), what level of inter-operability is being considered. For example, you build a landscape, you populate it with trees but then you decide to change the landscape. Will the tree positions automatically update to the new terrain especially if the mutti-cloner has the option for placing trees based on terrain slope (well, that is my assumption)? Infinite terrains.....in scope? I know you can't answer this question....but at least you know where my hopes are heading. I know you can't answer any of these (well....maybe question 5) but the results are right up there with World Machine. Even Vue users would have to export their terrains to World Machine to get half decent erosion results, so you are already years ahead of Vue in terms of terrain output quality. I see another Insydium in the making here with 3D Quakers. Better start hiring. ;-) Dave