better sculpt engine
the idea of a weld brush has been suggested on this website before but i have a better suggestion why not just make welding a part of the sculpt engine
like in 3d coat and other software when 2 different voxel bodies make contact they are automatically joined together/welded/stitched as like real clay , there is no separate brush for this process in 3dcoat and in other voxel based sculpting software
introduction of such a feature in blenders sculpt engine will prevent overlapping parts and remove the time wasted in the dreadful process of sculpting then remeshing to join separate mesh's and repeat , which slows the user's pc alot by magnitudes
this will bring blenders sculpting to higher levels , instead of adding new fancy but impractical tools the developers or developer behind the sculpt branch can focus on improving the basics and making small but effective and time saving changes that will make a massive difference in the long term rather than in the short term
10 Comments
Blender uses polygonal sculpting. It doesn't have voxel sculpting like 3DCoat, just a voxel remesher. If you want to join two meshes together you need to remesh or use booleans, and to my understanding this is the same within ZBrush. I do not believe it is as simple as adding a "weld brush" and I'm not aware of such a tool existing in any package that uses polygonal sculpting.
A voxel sculpting featureset within Blender would be great but this is a major undertaking. It's also not objectively better than polygonal sculpting: it's quite heavy and very demanding if higher detail is needed. Both techniques have pros and cons. I don't believe any of the sculpting additions since 2.8 have been "impractical," either. They have made Blender considerably more competitive.
They would be better off asking for SDF to be in main than to pursue voxel sculpt now.
Agreed. The addons for that have been promising and it would have more wide-reaching use cases considering how Blender is also used for traditional and hard surface modeling.
Both tools are more practical at the start/middle of the sculpting process regardless. If you want efficient fine detail you'd need to switch to polygons.
Voxel itself it's not a requirement, since a code to join or re-topo meshes can be "upgraded". The overall problem with re-mesh or dynatopo it's the big changes in overall mesh.
If you're doing a body for example, and split some parts like torso, arms, legs and them join and re-topo all the mesh will be changed since everything is recalculated, what means that small areas like eyes for example, will be messed up depending of how much resolution is used.
A "weld" option would be something less problematic, since only the overlapping joints should be changed, the concept can be something like the BRIDGE when in edit mode, but in sculpt mode.
The problem with proposing an "upgrade" of this nature is that it's similar to asking for Blender to efficiently handle 200 trillion polys on a laptop. It would absolutely make Blender better but it's not consistent with what's actually achievable. Is there open-source code out in the wild that will allow for this? Because otherwise it's asking for the 3D app equivalent of magic. Not even Zbrush allows for merging meshes this way without very specific use-case tools like Project Primitive.
Effort would be better spent on solutions that have an actual chance of existing.
Blender eventually will be stopped in time if this "there is open source code" thinking persist.
Eventually devs will need to create custom code and add new features to it.
At this moment, use things open-subdiv makes things easier for devs, but it also create a performance issue, since it's much slower than ZBrush subdiv, it also creates a scenario where devs cannot add or edit or create new features when they need something different of what open-subdiv offers, like a feature that I suggest long ago and requires a custom div math.
It also, make Blender dependent of Pixar (or any company that share OS).
The truth is: Blender need to do their own stuff to grow, so say something like "s there open-source code out in the wild that will allow for this?" is just a way to help to keep Blender stopped in time.
Since AI will be a rule for some scenarios like 3D rendering, Blender will need to find another ways to be attractive to industry and still existing, the better way is focus on get more artists and offer a better workflow and features, like the one suggested.
Few things on that:
You cannot expect for Blender to be more than it is. If you want the speed of development and innovation provided by propriety corporate-funded software, you will need to either pay up $millions yourself or build those solutions in your own time. They will not materialize in the current environment.
Something sad in Blender it's the lack of basic stuff inside sculpting option. So many years and "no layers" and a lot of "no this", "no that".
Layers can be safely implemented using Shapekeys
Do you mean the user use Shapekeys like a layer system or devs creating something with it?