Adding a new behavior as an option simply eases comparison and discovery of problems due to new behavior.
It looks like it breaks the use of Line Stroke Method.
There is no more line produced but just a little dot at start of line and a big one at end of line.
That may seems weird but the old behavior could be considered legit for curved surfaces.
This artifact is not always noticeable if amount of Input samples is increased and spacing is low.
In that case, the old behavior seems to provide a slightly faster fading of brush strength according curvature.
This world spacing option makes sense in many cases. But it looks like a bad choice for any brush using a setting based on View (sculpt plane or texture mapping).
I’d like to have a go at the Sculpt Mode Features branch.
A few questions, as I’ve never used a custom Blender build before, only Master builds of the main Blender:
Are the builds at Graphicall simply installable versions for anyone (no compiling or other tech expertise needed) ?
Does the Sculpt Mode Features branch include all features of the latest 2.8 beta Master builds? In other words: can you exchange the main 2.8 Beta builds for the Sculpt Mode Features branch compiles?
Yes, as @pixelgrip said, you just have to unzip, and you have it.
Not really, it depends on when the developpers merge master into their own branch, it is usually up to date, but it can be delayed by a week or two compared to master.
I wouldn’t switch completely to the Sculpt Mode Features branch because it makes some parts of Blender unstable, so it can crash sometimes, but you can have both versions, open your files with both of them.
Just tested, and this actually of kinda works, but since it has no UI yet, you need to assign a shortcut to it and play with the values there in the keymap…
This is a really exciting project that is pushing Blender’s Sculpt Mode much more towards a serious ZBrush and 3D-Coat competitor, but I do think one of the most important ZBrush and 3D-Coat strengths is still missing: a solid quad-poly auto-retopologizer. OpenVDB is an effective alternative for ZBrush Dynamesh, but there is no ZRemesher or Autopo yet. I really hope that will follow. Looking at Pablo’s achievements so far I think he has the capacity to create a good quad-poly auto-retopologizer as well, as that is the key to turning sculpts into clean, neat geometry without time-consuming manual retopo labor.
Zremesher dev I guess is making plugins for other software. Didn’t confirm anything for Blender, but seems to be looking into it towards the end of the thread.
Very interesting! Too bad the old 3ds Max and Maya dinosaurs are getting it first. I’d definitely be willing to pay for a solid Blender auto-retopologizer if it matches or surpasses ZRemesher.
IMHO the results from ZRemesher are really not that great most of the time. They’re the best on the market, sure, but far short of what a human can produce.
Now, faster manual retopology tools in Blender, that would be most welcome.
The new one in 2019 has improved a lot, especially coupled with the edge detect feature.
But yes, if it’s faces or proper subdiv hard surface it’s still not the way to go. Waiting for retopoflow for 2.8 and crossing my fingers for some serious speedup with high poly editing is the best I can do…
That’d be amazing! Zbrush, Substance and Rizom UV are the last commercial softwares in my pipeline, just waiting for a reason to hit that uninstall button :). Won’t be anytime soon I guess, but it’d be a big step forward.
Really glad to see a separate sculpt branch heading in a good direction as well.
The new ZRemesher 3.0 is pretty flawless indeed, although 3D-Coat’s Autopo also has improved during the past years.