Nomad Sculpt is absurdly good, even on a budget tablet like a Tab S6 Lite. It’s from the same dev as SculptGL. People can try the browser version of Nomad here https://stephaneginier.com/archive/nomad_demo/
if only he could donate his “multires” code to blender… hehehe…
If I remember correctly this (or something similar) was Pablo’s plan, to take a look at the code of SculptGL and implement it in Blender’s sculpt module, but I think it was Brecht who said that wouldn’t be possible even if the code was available as open source to use.
I don’t really know the technical details but the conversation is somewhere in one of the many tasks created for the Multires improvement/redesign plan.
Edit: found it https://developer.blender.org/T73047
That would be cool!
I wish for a “blend in shape from a shape key” brush
It would be awesome for my workflow
first of all thanks for the answer.
The problem to me was that when you paint in vertex paint mode you can paint vertex and faces, this is done with the face selection masking on usually.
To paint faces is usefull to me because you can use the new face sets to mask and the result is sharp edge even in a low/medium poly object.
I know that it is something that you can do now but you have to use sculpt > hide > vertex paint mode > face selection masking, idk if this will be dropped in the sculpt mode or it will be a new “dyntopo” mode for this kind of things, I think it was discused here.
I have been testing the OCt13 build, and I see that the Snakehook brush acts nothing like before. It barely pulls or pushes, I put the strength up to 5 but then it acts unreliably in different directions, as if there is some randomness involved depending on the direction of the pull.
Also, the smooth brush setting is not reflected in shift-left click smoothing, and I am not seeing a way to change the shift Lclick smoothing in the individual brushes. I am just wondering if there is going to be a distinction between the regular smooth brush and the shift-Lclick smooth for the new implementations. it would be really nice if the user can set the shift Lclick smooth brush size and strength separate from the regular brushes.
Another issue I see is that the brush performance drops drastically with slightly faster brush strokes, it also feels like the performance drop is also not connected to the speed of the brush at other times. You can see in the video (especially after 00:12 ) how choppy it gets sometimes.
With the lastest build (win10 / 10/15/2021 ) instant crash if the model has a subdiv modifier
I’m wondering: did Sculpt Dev build features make it into the Blender Master lately? When I check the Master History it seems to be Geometry Nodes everywhere.
To be very honest I’m experiencing too many crashes in the Sculpt Dev builds lately. I’m also a bit overwhelmed by the large amount of new brush options, and the slowdown some options cause, such as the Topology Rake values. Personally, I’d rather see perfected what’s already there, before having new options added, although I can lively imagine the fun of adding new stuff versus the tediousness of hunting bugs and optimizing old code.
Something worth mentioning if you want to test the sculpt branch (sculpt-dev.313c6811c3ab), it is not fully backward compatible, if you save your file and reopen it in master you will loose your T and N panels.
The chat is your friend…
hehehe… yeah, it looks like there’s an excessive focus on geometry nodes…
if only BF could allocate more devs to work on the sculpt mode…
That is not excessive. Geometry Nodes are revamped to use fields for Blender 3.0.
They have to replace dozens of nodes using custom attributes. Aand they have 5 days left, to re-introduce a fields version of each one , before Bcon3 that should be bug-fixing only phase.
I suppose that they will make an exception and continue to add nodes during Bcon3.
And 2/3 of recent commits are about code clean-ups and fixes, Cycles X, Compositor, VSE and Asset Browser.
Should be possible with the new money from Apple. We really need at least one full time developer who takes care of the sculpt and maybe one for the texture paint mode.
Especial the poly/texture paint side is so god damn outdated.
Yeah yeah, probably it’s just me, but it feels like there’s just too many devs working on geo nodes right now…
I’d say at least 2 for sculpt and another 2 for texture paint… hehe…
The deadline I’m most concerned about for geometry nodes is #Nodevember!
In fact, Jacques Lucke is taking vacation, right now.
So, Johnny Matthews is helping Hans Goudey who mentored GSOCs about Geometry nodes.
Don’t forget that developers are helping in areas where they have sufficient knowledge.
So, several devs are just adding nodes related to their domain. Charlie Jolly is our expert in procedural textures stuff. So, he added same textures nodes to GN as the ones he added to Cycles and EEVEE.
And in principle, anybody can contribute to Blender how he want.
There were several calls to developers community to participate to GN and add nodes they wanted to see. So, developers who helped to expand GN between 2.92 and 2.93 are, now, helping to convert nodes to fields workflow.
So, it is logical to see several names in commits ; when author of the patch is not commiter, and don’t have commit rights.
I do not think that the corporate supporters would be happy to see their money is used in areas they are not interested in. We do not exactly know what Apple’s priorities beside the rendering on Metal and other Apple related ports and stability fixes. They have to get those areas implemented first.
I believe all the money goes into a pool and can be used on whatever ton decide to do. Correct me if i’m wrong.
edit: Development Fund - blender.org
“By assigning corporate members a board and developer contact, they have the opportunity to monitor more closely what will get funded with their contributions.”
Nonethless Pablo Dobarro isn’t on the pay list anymore so there should be money to hire someone else full time to develope sculpt and texture modes.
The main problem is finding somebody that can, and want to do development on certain area is hard, since BF can only pay a modest salary in comparison to any commercial venture.
So if nobody steps to take charge of sculpt mode, it will be efectively orphaned until somebody do what Pablo did. After all, the Sculpt module was orphaned before Pablo’s proposal to fix sculpt mode.
True, Ton in theory can asign any developer to work on the module, but without the “know how” you will end with what you had before Pablo took the mantle, and with some of the actual developers with the idea of “it just works” bethesda style. (Not to disrespect the work of the many developers that worked in the past in the sculpt module). Nowadays the most closer we have is Joseph, but AFAIK he’s not in the payroll so probably his work will be limited to do what he can on his spare time.
Time will tell.