Jump to content

MJV

Contributors Tier 3
  • Posts

    794
  • Joined

  • Last visited

  • Days Won

    16

Everything posted by MJV

  1. Here is my objective. Simulating this real world gravity simulation. I tried to recreate it in Houdini last year but couldn't get soft bodies directly working with rigid.
  2. This post is specifically about dynamics in Cinema 4D but I assume all software uses the same terminology, so I hope for Houdini and other software users' thoughts too. 👨‍🏫 I have been trying, rather feebly, to figure out just how much is possible with Cinema's new unified dynamics and it has me wondering about the use of "Density" and "Mass" in dynamics, what the values mean and when to use one over the other, and also, isn't "Weight" a product of an object's mass and gravity? If so, why not let users work in Weight, which everyone can easily understand and conceptualize, instead of mass and density which seem much less meaningful. I have a scale at home and it conveniently tells me my weight when I get on it instead of my mass and density. It just assumes I'm on Earth, instead of say, Mars. Why can't dynamic simulations do this too, given that all the required values are known? In Cinema, when a user sets an object's dynamic mass value to 10, what is that? How can such a number be conceptualized? Why do ridged bodies have a "Use Custom Density" vs Use Custom Mass" setting. How can a user use that option effectively, and how should they conceptualize it? Now matter how much I think about it I can't intuitively conceptualize how changing Mass and Density values should effect the simulation. If I knew and could choose objects' weight instead of mass or density, it would be as intuitive as rolling down hill. Surely someone must have thought of this before, so one may assume there is some logical reason, that isn't obvious, why we can't work in weight. But I can't figure out what that is. What say you all?
  3. Thanks. That's not even the one I was referring to (it's driving me crazy I can't find it now) but yes his is even more sophisticated than the one I saw before. Even Houdini's soft bodies do not interact with their ridged bodies, especially not like that, which I recall being very disappointed and somewhat astonished to discover, it being Houdini and all, so the fact that Maxon pulled this off is really remarkable to me.
  4. Btw, did anyone else see the video where someone made a dynamic cube float while tethered to a cloth parachute? That blew my mind and now I can't find it again.
  5. I don't rember but I must have read it in this thread or heard it in one of the videos in this thread. I don't have the imagination to have made it up.
  6. It was my understanding that that is the main thing, and in my brief testing so far I have only been able to detect major speed-ups where deformers are involved and so far nowhere else. If you know about other areas of speedup please let us know.
  7. Yeah only Maxon can change a default or a menu and call it an upgrade. The editor speedup (thanks to deformers now being multithreaded) on the other hand is something that is sure to please everyone, and is the only new feature till now that intices me to use any version later than r21.
  8. In any case I don't think there is any thing wrong with doing videos about Neutron (or whatever Maxon is calling it now) at any particular stage in development. The tree branch/lightning generator thing that guy built is amazing as is and it's the concept and clever approach behind it that is most interesting. If some nodes or specific connections change in the future it will still be easy to rebuild using the same concepts, just as it would probably be easy to rebuild even in other already well established node systems like Houdini and others.
  9. Seems like a huge assumption that it's still under development.
  10. That looks brilliant! Such a smart and elegant tree generator. I was not expecting that.
  11. For the record, I never use an FFD without also using a point morph object to go with it. It's the only way to animate it or store different states.
  12. The LOD object only affects what objects within it are visible (as if that is the only thing that impacts viewport performance). Any deformers acting upon it will still deform all the objects inside the LOD Object, so this is not going to help with your issue at all. A better solution to your problem would be to use an Instance Object and apply the deformers to the Instance Object. Then build a simple Xpresso setup that allows you to toggle between the instance object pointing to the high poly object or to the low poly object.
  13. MJV

    scamsale.jpg

    Great stuff, and entertaining!
  14. I thought there were no perpetuals after r21. Has that changed,? I am interested in your r21 perpetual if you sell it.
  15. The Split-Lohmann Multifocal Displays thing looks like some seriously clever optical engineering. Would help relieve eye fatigue as well as make it way more immersive by being able to optically focus naturally and see different things in focus. I haven't followed VR development much in large part because the focusing thing was always a deal breaker for me.
  16. If the align normals command doesn't fix it, it means you have bad geometry. Overlapping dubicate polys, points, edges, etc. You could try selecting all points, making sure none are hidden first, and optimizing them, and then see where you are with that.
  17. It could have happened in a different time but I don't think there is a suitable incentive structure currently existing for Maxon to deliver on the original promise of scene nodes, with all the time and development cost that would entail. Why would a subscription service go so far out on a limb like that when all that is needed is to maintain the subscription service and the system operability and performance while adding as few new features as needed to placate those demanding something new. Not saying Maxon does that, but as we've seen with Adobe, it can easily and not surprisingly unfold that way.
  18. Yeah that video is what people are so angry about. 😆 Karma XPU improvements are super important no doubt but they were promised upon release of R19.5, so nothing really new above what was already widely anticipated and expected. Feathers is a super niche feature that won't impact many users day to day. There has to be something substantially more than this or, fair or not, Houdini R20 will be considered a huge bust.
  19. Seriously? Side FX closed down the r20 thread on their own forum because they haven't released a preview, as they always have in the past, and people are panicking like crazy at the complete lack of buzz. https://www.sidefx.com/forum/topic/86202/?page=26
  20. Oh I see. It has an option to pin on top while floating whereas no other Cinema window has this option. Yes it does seem odd that only this one thing in all of Cinema has this option and nowhere and nothing else does, which seems silly and inconsistent to say the least. However, if you create a specialized tool AM for modeling settings you can leave it floating and it will always stay on top or you can alternatively dock it anywhere in your layout, as I do.
  21. This is already possible, no?
  22. If you Current State to Object the spline and clone that, it works, but of course it's not parametric anymore.
  23. I feel like that video is trying to convince me not to use Octane for Pyro.
  24. No idea. I didn't even know that they had been updated all the way to 2023 until recently. Someone should grab them all before the full server goes down.
×
×
  • Create New...

Copyright Core 4D © 2023 Powered by Invision Community