Pablo Dobarro's master plan for sculpting and painting, development news

I need an asset manager and brush manager so bad. I would then be able to create a massive library of stuff I want to reuse to speed things up as well as share the brushes I have created. Been trying to recreate a lot of standard ZBrush brushes and in some cases I think I have been quite successful. Would love to see some of my work get introduced as standard tools for all Blender users. :slight_smile:

It was already confirmed that Everything Nodes will land this summer, so 2.90 sounds very plausible right now.


Did you tried with the multires as well?

Can anyone here confirm that the very latest build (not the one from the buildbot but a fresh compile after Pablo’s latest commits) instantly crashes Blender when adding a Multiresolution Modifier to a simple cube, subdividing 2 times and then switching to Sculpt mode?

EDIT: OK, the bug is confirmed. Pablo is killing it! (Like ALWAYS!)


I remember reading something about this, yes - but I believe they meant only the particle system. In any case, a new particle system is already christmas, and the underlying “functions” system lays the foundation for the other node contexts… so it’s sensible to think we won’t have to wait so long for them.

Asset manager is pretty cool as well, and will be extra convenient on larger productions. Hopefully overrides make it soon as well, those go hand-in-hand.

I just tried it with multires on a quadsphere. 6 million is comfortable I got to 25 million,which is the next level up, the brushes still handle will, whats not good is mouse movement and navigation(move,rotate,zoom) are slow.

With that said I don’t even need 25 million quads, I’ll be happy with 6-10 million quad sculpting for the fine details I can use a normal map.


Hans Goudey […] is working on drag and drop for the modifier stack.

Nice! I hope this can be integrated in the Modifier List add-on as well.


I think so too.

Looking forward to this very much! :+1:


Just had a go at the Sharpen mesh filter. Hmm… :thinking: I’m not convinced of its usefulness yet. It melts into a blobbish, smoothed result too rapidly, no matter how I change the Smooth Ratio value. What do others think of it?

I also wonder why there’s no Sharpen brush. It would be more convenient than having to mask areas, smooth the mask, activate the Sharpen mesh filter and applying it.


Maybe a little off-topic but here is my question anyway:
Do you think that changing the version numbering and adding LTS is a good idea (useful) ?

I think the version numbering is good. The increments were too small compared to the many changes per version.

LTS is a case of doubt for me. I can’t help a slight feeling of dismay from witnessing the big donating corporations creep their influence into Blender development. Blender developers will lose time to maintenance. I’d rather see Blender remain completely independent of the big commercial guys, but I realize that’s utopic.


I have always disliked “big softwares versions jumps” for marketing reasons (for example: the competition between Firefox and Chrome), but I do understand that sometimes, because of the number of important implemented features, a major version jump is necessary.

About LTS, I also doubt that studios will refrain to use the latest official version (2.90, 2.91, 2.92) just because 2.83 is LTS… Regular new features are enticing.

The people that want LTS have access to other software like Zbrush, Mudbox, 3DCoat, Arnold, Houdini etc in their pipeline I seriously doubt the are gushing over the current sculpts developments as much as humble broke plebs. :rofl:

LTS is such a nothing burger that I can’t see where all the worry is coming from. It’s not for joy average Blender user I also think that overall it’s better nomenclature than the usual a,b,c, release we get.

Back to gushing for me, Boy that layer brush looks super awesome :heart_eyes:


There is a crease brush. Pablo created a Draw Sharp.
He made pinch perpendicular to brush stroke.
He improved scrape brush and created a multiscrape brush.

I don’t think such brush will be useful.
If you create a sharpen brush based on sharpen mesh filter behavior. It will have same problems.
1 brush stroke -> good result. 2 brush strokes -> smoothing.

I am fine with limitation of using sharpen mesh filter only with one iteration. It implies for user to always use it with a mask or face set restriction.
Honestly, after quick tests, I did not find an utility to filter settings related to its name.
You can create ripples by disabling axis.
But except that, tweaking numerical values doesn’t seem useful at all.

It does not seem like a really useful mesh filter.
But seriously, Pablo already did his maximum to make brushes sharper.
Next to each brush specific improvement I already talked about ; he made a global change affecting all brushes by modifying FallOff management.
He created new FallOff profiles and added an Hardness slider to all brushes.

And he will complete that with trimming tools.

That should not be difficult at all to obtain sharp angles.


I guess you’re right, there are enough brushes to sharpen shapes already. I haven’t been able to make the Sharpen mesh filter work as effective as ZBrush ClayPolish though. You’d expect that increasing strokes / gestures would sharpen the shapes more and more, but in stead the result rapidly converges to smooth, blobby shapes.


Yes. That would be great if Pablo could make the tool more predictable.
I don’t know if second iteration is a pertinent behavior due to resulting geometry of first one ; or if when maximum of sharpening is reached, only smoothing is happening.

If it is the second case, maybe that is fixable by stopping filter to have any effect.
That could prevent to create a mask at each use of filter.


I think changing numbers is reasonable. I prefer current format but mostly for nostalgic reasons. I believe pretty much everyone who is familiar with Blender development doesn’t care about numbers but rather about the progress, feature, stability and overall improvement of the package.

Yet numbers are helpful for development to keep pace and high moral of developers with each milestone achieved. Having consistency in that regard is very helpful and healthy for Blender team and development.

But I believe we’re missing marketing opportunity with numbers. Since moving from 2.79 to 2.80 took so long it was a great marketing moment that created a huge momentum for Blender. With new numbering system 2.9 or 3.0 won’t feel like a huge achievement but just a steady predictable progress.

I know the biggest marking force behind Blender is it’s community but I would try to leverage numbers as much as possible in Blender development to introduce best and biggest changes at important milestones like 2.9, 3.0 etc. Again, these numbers doesn’t mean much to those who constantly follow blender progress but they are important to the outside world and can have great positive impact if leveraged properly.

I had similar worries if we’re honest. I don’t worry about it as much anymore. Basically because it’s out of communities control and solely on Blender Foundation leadership to leverage corporate connections well.

If it’s done right corporations are a great benefit to Blender. Funding is essential for healthy development. While I and many others (almost 4000 already) are supporting Blender Development Fund corporations can bring way more money for non-profit organization.

LTS is needed for better industry penetration. Stable Blender version with long term support can give so much piece of mind to big companies and allow integrate Blender in pipelines for longer projects. Right now I think it’s a good idea and in a way an experiment. If it doesn’t work out Blender Foundation can adjust or remove the offer for future releases.


Were you using dyntopo/dynamic topology while trying this method out?

Anyone had any success with sharpen filter? I’ve tried it on 4mil mesh. on max settings I can see no effect at all.

4mil is lot, you won’t see much effect on it…
Like ClayPolish, this tool is more for low poly stuff…