I recently had the same issue using particle emitters and dynamics. One solution is to create a cloner, add you sphere to the cloner. Then set the cloner to "object" mode then select the emitter as the object. You don't need any objects under the emitter and the simulation should work fine.
in my experience, one of the things that mess with the simulation is the distribution of polygons on your sphere, when I change it to Hexahedron instead of standard I get better results
The final render's lighting looks fucking insane bro. My only wish is that you showed us how you light and rendered everything. Please consider a tutorial on that
I usually have better simulation results when I work with 10x the scale of the real objects. All my settings like substeps, collision distance, particle size don't have to be so small and engine runs them faster and with less issues. I would suggest to anyone to try that regardless of the software.
Have you tried changing the scene scale in the simulation settings, from 10cm down to like 1cm, instead of scaling the objects to get a more stable simulation on small scale simulations? I havn’t yet tried it myself. Great vid as always! ❤️ Edit: ok, Im an idiot commenting before watching the whole video 🫣 never mind my suggestion hehe
You can have motion blur with exported alembics by forcing deformation and enabling motion vector in RS object tag. In 2023 I was also adding motion blur tag (regular c4d one) before baking alembic
Alembic isn't suitable for motion blur because every frame is exported as an entirely seperate mesh internally, the software has no way of knowing which vertex from the previous frame is linked to which vertex from the current frame. This would normally be solved with motion vectors (which can be imported as vertex maps in the redshift object tag). However, for some reason, C4D itself has no way of actually generating these motion vectors in the alembic exporter, making alembic pretty much useless for caching inside cinema4d. I wish Maxon would take the time to actually perfect these core features, instead of leaving them half-baked for years while rushing to add flashy new features to every release.
@@cyrilruegsegger9759 not the same thing. Motion vectors provide exact values for each pixel, allowing for accurate MB without any guesswork. Post process MB techniques all just compare current frame to previous frame and then try to figure out which pixel has moved where, which is far less accurate.
Hey Vincent thanks for the amazing tutorial! But how are you saving a startup or new scene, with the folder structure in the object manager already in place?
it's good to see you work through your issues, it makes it more educational.
I recently had the same issue using particle emitters and dynamics. One solution is to create a cloner, add you sphere to the cloner. Then set the cloner to "object" mode then select the emitter as the object. You don't need any objects under the emitter and the simulation should work fine.
nice
in my experience, one of the things that mess with the simulation is the distribution of polygons on your sphere, when I change it to Hexahedron instead of standard I get better results
the first thing that popped into my mind when i started this video
You need to do a masterclass on your lighting setups. They're so damn good.
Thank you for testing out the new features! And thanks for showing your workaround!
매우 매우 매우 좋은 영상이다! 감사합니다
Richtig gutes Video Vincent!
thank you a lot Vincent!!!
I think redshift fixed the cache thing now within the new updates
Thank you my teacher
Thanks man!!
thanks for sharing 👍
This is so inspiring, thanks for the hacks.
Thank you for sharing your knowledge Vincent :)
Hi! Nice tuts! I love all your arts and style. Also You can use in volume builder Intersect mode, and you can create only one cube, not four
yeah, haven't thought about that. Would have saved some time.
This so helpful, Thanks!
Thank you very much man
Excellent result. I would like to see the material settings
this comes on Patreon in a few days
The final render's lighting looks fucking insane bro. My only wish is that you showed us how you light and rendered everything. Please consider a tutorial on that
The second part of the tutorial is going to be on Patreon in a few days
@@vincentschwenk Will that show exactly how to light and texture everything to look identical to your final render? Cuz if so I will sign up
yes @@YouAlreadyKnowBabi
I usually have better simulation results when I work with 10x the scale of the real objects. All my settings like substeps, collision distance, particle size don't have to be so small and engine runs them faster and with less issues. I would suggest to anyone to try that regardless of the software.
yes, that's a good value. But I hate working in a non real world scale, as I have a photography background and then everything changes.
@@vincentschwenkI have the same preference, what I’ve done before was to export my sim as alembic and scaling that down for rendering.
gracias por aporte
Very nice
Have you tried changing the scene scale in the simulation settings, from 10cm down to like 1cm, instead of scaling the objects to get a more stable simulation on small scale simulations? I havn’t yet tried it myself.
Great vid as always! ❤️
Edit: ok, Im an idiot commenting before watching the whole video 🫣 never mind my suggestion hehe
very nice tutorial / the exciting part is on pateron and idont have acess to pateron in my country :/
Which country is that?
could you please share material setup and lighting? and Thanks for the video. Amazing Job!!
www.patreon.com/posts/51-sponge-90327474?Link&
thx man! awesome tut! usefull!
You can have motion blur with exported alembics by forcing deformation and enabling motion vector in RS object tag. In 2023 I was also adding motion blur tag (regular c4d one) before baking alembic
I have to check that out. I only know that with houdini alembics
Alembic isn't suitable for motion blur because every frame is exported as an entirely seperate mesh internally, the software has no way of knowing which vertex from the previous frame is linked to which vertex from the current frame. This would normally be solved with motion vectors (which can be imported as vertex maps in the redshift object tag). However, for some reason, C4D itself has no way of actually generating these motion vectors in the alembic exporter, making alembic pretty much useless for caching inside cinema4d.
I wish Maxon would take the time to actually perfect these core features, instead of leaving them half-baked for years while rushing to add flashy new features to every release.
didn't know that, but it makes sense. Also it is so stupid, as motion blur is so important.
Otherwise you can put a Forced Motion Blur effect directly in After Effects
@@cyrilruegsegger9759 not the same thing. Motion vectors provide exact values for each pixel, allowing for accurate MB without any guesswork. Post process MB techniques all just compare current frame to previous frame and then try to figure out which pixel has moved where, which is far less accurate.
Hey Vincent thanks for the amazing tutorial! But how are you saving a startup or new scene, with the folder structure in the object manager already in place?
Amazing!
I dont know why my c4d crash when i caching the simulation and i put specs lower than yours
Don't hold your breath over having these issues resolved. They look similar to the issues with the old collision dynamics and they were never fixed.
I thought I was the only who cancels render inbetween cause I saw some mistakes🤣
What a nightmare - C4D is a buggy mess and is ready to be replaced by a newcomer
Stop complaining. Blender physics is a joke.
Time to upgrade your hardware my friend
Please stop using Redshift. Use V-ray and Arnold instead !