3D-Pangel

Volume builder would help you get the shape by manipulating basic primitives but my understanding is that it limits you with the application of texture decals. It may be a good way to start but you would still need to go through a number of steps to convert that volume to a polygonal mesh with adequate topology (quads, edge flow, etc) that would allow for accurate texturing and rendering. In that conversion you may loose the starting shape a bit which would then require hand modeling to get your quads back where they belong. But for complex shapes like this, it may still be a time saver (and less frustrating) than other methods unless your have the amazing skills of Cerbera or Vector. I have always viewed the volume mesher as a great tool for creating STL files that will be 3D printed but not for creating general purpose models that need to textured or rendered to real world accuracy. Either way, this will be a fun project to watch so please keep us posted.

Not sure a shader could handle the complexities of object interactions in a way that would look like fluids. It would work for static objects like clouds though.

The emitter was a rolling object discussed in an earlier post to get that rolling cloud look I was after. That object was all quads and thoroughly mesh checked because those weird artifacts seemed to be matching the emitter's polygons. It ONLY appeared once the temperature was activated along with density. Notice that after the ship emerges it all goes away as I key framed temperature off at that point (thinking being that the ship was the source of the heat). While you may have liked the early attempts shown in a previous post, as the emitter rolled that geometric pattern and noise to the fire only increased much more significantly very late in the animation just prior to the ship's emergence in those attempts. Smoothing and playing with the dissipation rates toned it down a bit but it never went away. Plus, for every change you would make, you would have to cache up to 500 frames just to see if anything improved. So it was slow going. Plus, remember that scene was huge. In real world scale that ship was a quarter mile wide (and made to look even bigger by scaling the buildings down) and the voxel size was 100 cm. Getting it to finish caching was a triumph in itself. So it was meant to push both my machine and C4D. The final cache size was 352 Gb! Some frames cached in at over 800 Mb. I honestly never thought C4D or my machine would be able to handle it....and so quickly. 352Gb were cached in 134 minutes. Render times were 70 seconds per frame. My understanding and ability to manipulate fluids were the weak points more so than the hardware or the software. Next time, I will play with less massive subjects. Leave that domain up to Houdini experts. Dave

After NUMEROUS iterations, I am still not 100% happy with the final animation. There is still a great deal I do not like about it but I think I have run out of either of the following: 1) The motivation to keep working on it, 2) The time to keep working on it, or 3) the skill to make it better. As for me, probably all 3 (maybe mostly the third). But here is the final animation: https://www.dropbox.com/s/pk8hn1f6m2ciub4/Cloud.mp4?dl=0 (mp4 version) https://www.dropbox.com/s/jwl61zuye2e4ltl/Cloud.wmv?dl=0 (wmv version) Now I made this to push what both pyro and my new workstation can accomplish. On that account, I was not disappointed. Both Pyro and my machine performed beyond my expectations. Dave

So I have registered and the course is massive. Interestingly enough, after registration it does not show up in my Enrollments page at Gorilla U (click on the "Go to Dashboard" in the upper left hand corner to see that page). There is the X-Particles training which I purchased from GSG before they went subscription but not this RS course. This leads me to believe that once you have registered for the course, you really don't have it forever. You really only have it until the offer expires on 3/1/23. I watched Nick's video on Twitter using the link above and there is no clarification. He just says it is free "for a limited time" which implies to me until 3/1/23. Nevertheless, still a great offer but treat it like it will go away in 19 days. Dave

Just released:

If my key purpose is video editing, then maybe this is the machine to get. But for 3D, well...this link says it all Cinebench R23 Benchmark Scores [2023] (nanoreview.net) M1 Ultra is in the 23rd position. M2 Max is in the 74th position. M2 Pro is 79th M1 Max is 123rd M1 Pro is 133rd So....if you believe that Cinebench is an unbiased means of measurement, then for 3D, there are better choices (22 of them to be exact). Dave

Really well done. Every time I see chainmail, I think of that poor guy at Weta working on LOTR who had to make all that chain mail for every character by cutting plastic plumbing tubes (the ones reinforced by a spiral loop of wire) into individual rings and then linking them into that cloth pattern and gluing the ends together. Kind of makes the work you put into learning Corona Pattern all the more worthwhile doesn't it!!!😀 Again, great work! Just love it. Dave

Hmm...for half the cost of an annual C4D subscription, is there anything that could not be replicated just as well using C4D's volume breaking and fields? What am I missing here given the cost at $305/year. Honestly, I was always a fan of Pull Down It, but this just did not impress me given the cost. Plus anyone notice this little defect in the shattering algorithm: The part marked by the red arrow is fully separated from the rest of the sculpture but moves together with the part marked in green. So a glitch in how the parts are shattered and identified as separate objects. As they are fine little details this may even be a challenge for C4D, but it is Pull Down Its demo reel!!!! Not good. Dave

The only reason I can think of when two material managers might be nice is if you should start assigning materials to different layers and then realize that you want to use ONLY one of the assigned materials to that layer on object that has its materials on a different layer. For large projects that has sometimes created a few extra steps to correct provided you were not careful is assigning your layers to begin with. Now if anyone has a pre-existing workaround to that than please share. I have only come across this problem once so I never deep dived into it. Thanks, Dave

I was referring the MCU Phases of their release schedules. MCU Phases 1 to 3 carried us from the first Iron Man to the Infinity War films. Phase 4 is everything that came after Infinity War until now. So as Phase 4 was a bit of a miss, Phase 5 is being re-thought. Honestly, what made Phase 1 to 3 great was that they were all telling one complete story from introducing each super hero, to forming the Avengers to then fighting Thanos with a common "mcguffin" or plot device of the infinity stones. People would wait until the end-credits to get hints of what was coming next because all the movies were part of the same overall story. There was a certain continuity to them all. What drove Marvel movies to blockbuster status was when people realized all these movies were related.....so you can't miss any of them. You have to watch them all. Honestly, Phase 4 has none of that. I have no idea how all these movies relate to each other and/or what is driving the overall narrative the way the infinity stones drove the overall narrative of Phases 1 to 3. The scenes at the end of the Phase 4 movies just leave me more confused rather than tie things together. Peace out Marvel. I will catch you on video. Also, when you add to that what Disney did to Star Wars, you have to begin to wonder. Yeah, I liked Andor, but honestly, that still does not wash the bad taste of the sequel trilogies out of my mouth. Obi-Wan was a bit of mess. Willow is a bigger mess (I could not even get past the first episode). And let's not even talk about She-Hulk. So peace out Disney+ (I did not renew my subscription -- no point to it now that they want you to pay and still watch ads for their crappy programming). Dave

Marvel is only studio where they budget re-shoots at a very high level even before they start shooting. Their reasoning is that their stories are a bit "high concept" and given the volume of movies they produce each year, they are pushing the pre-production teams to their limits. In short, they spend less time on pre-production planning than they should for the size and scope of movie that they are producing and prefer to just "fix it in post" should the edit not come together the way they expected. This plan works when each movie is a blockbuster as they were in Phases 1 to 3. So, if each movie is generating hundreds of millions in profit, then would you want to only produce one a year and take the time to plan it right or pay $20 to $30 million more in post-production to fix the edit so that you can get two blockbusters out the door each year. Unfortunately, Phase 4 was (IMHO) a dumpster fire for Disney Marvel. "Eternals" was too long and confusing. "Multiverse of Madness" was just a confusing mess built on a weak premise that even Sam Raimi couldn't fix (really...she gets scared by a bee and can now cross multi-verses? Are bees that scary? Is that the best they can do?). And "Love and Thunder" prompted a "What were they thinking?" type of response because it violated every concept of the "hero's journey" essential to good character development. It also made Thor look more like the village idiot than they have done in previous movies (which I think prompted Chris Harmsworth's pledge never to work with Taika Waititi again). "Spider Man No Way Home" was the only movie that actually lived up to and exceeded my expectations for good story and character development that we saw in Phases 1 to 3. Unfortunately, the deal between Sony and Disney over the Spiderman rights ended with that movie. Who knows what happens next. So that VFX reel proves to me that they are not even trying to do good pre-production anymore. We really don't know what the VFX will look like through the windows so let's not even try to match the lighting on set.... we'll build the whole room in 3D and fix it later (imagine if they had done some pre-planning and shot that set in a Stagecraft Volume). Where does the actor stand? Ahh...put him here, and if we don't like it, we will fix it later. Did we make a prop book? Whoops --- didn't have time. Here, hold this piece of cardboard -- we will fix it later. It may not have been as sloppy as that, but you get the idea. That "we will fix it later because we have the money" mentality keeps a lot of VFX houses employed by Marvel. Marvel's movie revenue for Phase 4 is not where Disney wanted it to be and did prompt this rumor on Phase 5. So that thinking may change. Dave

Yes....same here. I would have responded earlier but we had a pretty severe snowstorm on Monday that brought down trees and knocked out power. Our power just came back on last night....just in time for another storm this evening. So thank you Corebot!!! Dave

Janan, I think this page explains it pretty well relative to bronze support. A couple of things to note from that page (screen shot below): Note that my contribution level (Silver Monthly) is highlighted in Blue. So that is how you get feedback of what program you are in. My understanding is that if you sign up for monthly then you are opting to go into automatic payments each month. In that program, I do get two emails each month: One telling me that a deduction is about to happen and then another telling me that there was a deduction, and it provides and invoice. The first looks like this: and the second like this: I would imagine that whether or not you are paying monthly or annually, you should be getting an email from "Core4D Community". It is all pretty well managed. Also note that regardless of whether it is monthly or annually, all programs come without ads. The yearly plans are the better deals for those who are on the fence about joining. I hope this helps, Dave P.S. I know that there have been some who complain about paying for Core4D and complain about the ads. I honored those complaints until I realized just how cheap Core4D sponsorship could be (I gave monthly before the start of the subscription program, so I never looked at this page). For 12 euros a year, all the ads go away (that comes to 1 euro a month.....you can't even get a candy bar for that price)! But as I don't know about everyone's financial situation, then there could very well be some for whom this is a hardship. In that case, my heart goes out to those who can't afford 12 euros a year or 2 euros a month and with all sincerity, I wish you a better future.

Yes....I saw that. Very interesting. But again, what is real or CG is still open to debate even in their minds as they do go back and forth. They even propose that it could be a combination of CG and live action. But notice also that however they do land in the argument, they will say that they could be wrong. Again, the big details being called out are around the micro behavior of water. Can you get CG water to seep into and pool up between the weave of the cloth or on the stitch of the leather? There was mention of a technical paper on new water simulation capability that came out of the Weta team a few years ago. So, is micro water physics the next great breakthrough in liquid simulation? I hope so. I have great faith in Weta to make anything happen. Dave

Very interesting in what needed to be overcome to do motion capture underwater. Those nasty bubbles that confuse the tracking cameras as they cannot distinguish between a bubble and a tracking dot on the actor's suit. But I don't think it was that bubbles were being read as balls more so than the bubble acting like lenses and distorting whatever was behind it. With that said, were RFID emitters ever considered? Apart from motion capture challenges, there were also some advances on simulating micro water behavior - that is how water seeps in between small features such as the weave in cloth as shown below in this image from the movie: But there is still some debate whether or not this scene is 100% CG or a combination between CG and live action. We will probably not know for sure until the movie has finished its theatrical run and the VFX artists are allowed to spill the beans. As for me, I saw the movie in IMAX 3D and it was just visually amazing. Definitely worth the $20 ticket price -- ouch! But only for a James Cameron movie would I pay that much. Story wise, you have to go into it realizing that it is setting up for the next 3 movies so there are some things left unanswered and not the self-contained and complete story that you got out of the first movie...but still good storytelling but not quite perfect. Dave

Thank you Corebot....I knew you would be on top of this and I appreciate the challenge that it represents. Dave P.S. So who is Corebot after all? Hmmm.....

Okay....as MS Edge is based on a Chrome architecture, then essentially they are the same. So I thought I would try Firefox. Well, after logging in I no longer get white space but I do get a ton of ads: Plus they are constantly changing and sometimes video ads pop up either to the left or right of the forum text fields and below the banner (thus explaining why main section showing the forum messages is so narrow). Oh my! That is distracting. What was that line from "Ready Player One"? So again, it appears that Firefox is NOT the browser to use as a Silver Contributor. Just seeing what non-paying users put up with makes me happy with just seeing the white space...even though that is NOT ideal. Dave

There is also this weird page formatting too going on: Look at all that dead space below the banner and how narrow the actual forum entries are appearing. I thought maybe there was some browser setting I needed to change or something in my Core4D user profile, but I could not find any. Plus it only appears on this web-site and when I use other browsers and flush the cache, the problem does not go away. Dave

Glad I am not the only one (well...sorry that it is affecting others on the forum....but you no what I mean) If it is ad related, then am I still supposed to get ads as a silver contributor? This may explain why I am getting the ad window without the ad -- just a white box. Dave

So I am seeing this strange white square at the bottom of every page of the forum....and only on this forum but on other browsers as well (so both Edge and Chrome) It is like Core4D is pulling up a message on accepting cookies but that message is not loading....nor is there any embedded location where "close" or "accept all cookies" would normally be located. Cleared cached, cleared cookies, tried other browsers ...problem is still there. Anyone else seeing this? Dave

It looks like a sub-step issue. The simulation only calculates the collision by comparing where it is now to where it would be in the next frame should the collider object NOT be there. If that point is past the collider object's boundaries, then the simulation determines the collision behavior from its current position -- which may not be close to where the collider object is now if the simulation is moving fast enough. Thus the large gap. There are some settings to play with. Decrease the object thickness on the collider tag and increase the sub-step count on the pyro object. Increasing the sub-step count simply takes that distance between current position and the next position on the next frame and sub-divides it by that count. This will get the collision behavior closer to collider object and eliminate those gaps. Now....I am doing this from memory as I do not have C4D open at this time --- so hopefully those two settings are both called and located as I am remembering them. Dave

No animals were harmed in the making of this short and all actions were monitored and approved by the SPCA. I loved it. The lighting and atmospheric early morning fog was spot on! Though the wolf's growl was a bit off (sounding more like a hungry stomach than an actual wolf), I loved the overall sound design especially the footsteps in the woods. Every footstep was perfectly keyed to the sound effects. That took some work. Honestly, for anyone who desires to do more with CG than stills or motion graphics, Unreal is growing to be the application that provides the broadest set of capabilities to "story tellers". Its toolset is geared toward providing "on-set" tools. By "on-set" I mean its tools are designed to create finished elements rather than provide the capabilities for developing those elements (e.g., do you want flowing "water" or a fluid system for creating flowing water - there is a difference). If modeling is not your thing, then there is Quixel for environments, Kitbash3D for structures., and a host of other sites for characters and character motions. I resonate with Dardo's comment of loving the virtual on-set environment that Unreal provided. And it does all that at close to photo real quality and pretty fast as well. Unfortunately, after watching a few tutorials, the UI appeared daunting to me. I actually could wrap my head around Houdini's UI better than I could Unreal engines. Maybe it was the tutorial. So for those who have stepped more into Unreal than I did (admittedly a passing interest on my part with no real time investment), what were your experiences with the learning curve? How would it compare to Houdini? To Blender? To C4D? Dave

I actually asked a similar question some time ago and the general consensus was that depending on what you were trying to do, 12Gb should be fine and 24Gb should be more than enough for "most" situations. Now, I put "most" in quotes because if you are looking to generate fluid simulations, then VRAM gets quickly consumed. From experience, 8Gb is just too small these days for most things. My last workstation used an 8Gb card (albeit a 7-year-old gpu) and it could not render certain models from Kitbash3D as they tend to use 4K or 8K textures. If your limit is 12Gb, then you may find certain fluid simulations may be out of reach but that does NOT mean all simulations are out of reach. Just be careful with the voxel size and frame the shot accordingly (eg. higher voxel sizes will be okay as long as you don't want a close-up shot of an overly detailed simulation). Also, relative to the CPU, focus on clock speed and not core count. With GPU rendering, the CPU is nothing more than the traffic cop setting the speed limit on moving data to and from the GPU. The GPU does all the work. With that said, I got the cheapest Threadripper available on the configurator which worked out because the cheaper CPU's have the fewer cores but the fastest clock speeds. My choice was the AMD Ryzen Threadripper Pro 3945WX (12 core, 4GHz). Another consideration is not only DRAM speed but whether or not the CPU/motherboard can support multi-channel memory. Some tests at Puget Systems have not (by their own admission) thoroughly tested the benefits of multichannel memory in 3D because their 3D scene files were not that large. But for video editing, etc. they did find that 2, 4 and 8 channel memory configurations made drastic improvements in performance (eg. two channel is 20% faster than single channel, four channel is 20% faster than two channel, etc). You will do better to first determine how many memory channels your configuration can support (either 2, 4 or eight) and then dividing that number into how much memory you want and then configuring your PC with that as your max DIMM size. So if you want a total of 64Gb of memory in your system, and your configuration of CPU and motherboard can only support 2 channel memory, get two 32 Gb DIMMs. If your configuration can support 8 channel memory, then get eight 8Gb DIMMs. Think of multichannel memory like configurating a hard drive array in a RAID 1 (stripping for speed) configuration. I hope this helps. Dave

Igor, Happy to. The biggest issue is with creating the massive scale of the City Destroyer (or CD for short). That ship is supposed to be 15 miles wide so you can imagine how big the fluid simulation would need to be if I just decided to model everything to scale. So you have to cheat and the biggest cheat I could think of would be to have the ship pass "just above" the buildings when in reality it would be much higher in the movie. Therefore, you would need to insure that there is no light or reflections from the buildings onto the ship otherwise it would give away just how close they were and destroy that sense of scale I am looking for. So here is a side view of the scene set-up. The building are only 600 cm's tall. The ship is about 44740 cm wide and 6300 cm tall. So if you assume that the buildings are about 40 stories tall (around 400 feet) then at that scale the CD comes to around 6 miles wide. Not quite the 15 miles discussed in the movie but good enough to get me where I want to be. Also, not a lot of buildings were needed, and they are strategically placed to cut off a good portion of the wide-angle view with more being placed closer the camera and fewer behind with a pseudo main street down the middle to leave enough of a gap to see the fire effect. The next challenge was how to do a rolling fire cloud. Not sure why that took two iterations when in reality it is pretty straightforward. I thought I could just do it with a high rate of turbulence on the fluid simulation but that killed the scale again because to see that turbulence you would need to allow a long dissipation time. You don't want big whispy clouds because again that would kill the scale. The frame captures above that show the ship are from this first attempt. But here you have a static cloud that just kind of simmers rather than rolls. So, to get that rolling affect than it was a simple solution: just create a odd-shaped tube to be the emitter and have it roll along just ahead of the ship until the point you want the ship to emerge from the cloud: Remember that the odd shaped tube is just the emitter and it will be deleted after you have cached the simulation so it will never be seen in the render. When getting the integration of the fluid simulation with the ship I was a little concerned that the ship was made with nothing but booleans (sorry Jay 😉 ) but the connect object worked better than I expected. I just put a collider tag and set Bounce to 0 (that would kill scale) and Friction and Stick to very high values as I wanted the fluid to be dragged by the ship and stick to it because the ship should be the source of the fire and the ship only emerges from it because the ship is cooling --- but cooling unevenly so that is why parts of it are still emitting smoke. A few problems to still solve. The polygon count on the cloud is too high and needs to be reduced because while it is giving a great look, each VDB file per frame is over 1Gb. So I need to reduce the polygon count, shorten it up a bit (strategically place a few more buildings to make that possible) and remove velocity from the simulation cache. So much for really good motion blur but as the ship is supposed to be massive, it moves slow so you don't need motion blur for the ship but it would have been nice for the rolling cloud. That's about it. Hopefully I can get the cache size down to get this done. There were no VRAM errors but C4D just crapped out after 139 frames which I think may have something to do with the size of the VDB files. The ship is moving slow so it will take at least 1000 frames to pass over the buildings. Any faster would not match the pace shown in the movie. Dealing with large scale is a bitch in CG but that is what makes it memorable (IMHO) if you can pull it off! Now....what can you do in Houdini? Sorry. I had to ask and lay down that challenge!!!!!! Dave