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

Sorry for the delayed reply. May the 4th you know. :wink:

Really, does it? And these are all directly quickdraggable?

But however it is…I really don’t want to downplay the quality of PME addon itself. To me it looks like a really well done tool and If you are happy with pie menus, than thats how it is.

Seems we argue towards the same point from starting from different sides.
My argumentation for the brush panel is quite exactly that. I just think that these favorites could/should be handled overall by the same panel in prioritized way, many solutions come to mind here, and as a panel dedicated for brushes a better visualization would be possible. And overall there’s no need to think that being like zbrush is the only solution.

To be honest, I’m not crazy about Pie Menus myself, as I prefer traditional floating dialogue menus. :wink: You can create all sorts of options in PME, but I’ve only just installed it recently and haven’t had much time to explore.

Indeed, this is a VERY important aspect of Blender’s sculpting workflow and essential that it gets it right going into the future when there will eventually be many, many brushes including custom user creations. Zbrush is always going to be held up as a mirror to Blender Sculpt mode and I suppose it’s what I know most and will naturally compare to, but you’re right of course and it shouldn’t always be looked to as a solution.

In the future I would like to see a global asset management system for sculpting/painting brushes, textures, brush alphas, etc.

3 Likes

Yes all true. I fully agree.

Me too. Seems like William is against changes coming to the brushpanel in smaller pieces. So it might be branched out. (https://developer.blender.org/D7616) I am not convinced that this is needed, as it is already marked as experimental and we at the beginning of the alpha phase, but currently it looks like changes will be added solely there until the overall concept fits together. :pensive: But nonetheless. There’s movement and movement is good. :slightly_smiling_face:

2 Likes

You can create all kinds of custom menus using PME. Here are the features:

  • Pie Menu Editor
  • Regular Menu Editor
  • Pop-up Dialog Editor
  • Macro Operator Editor
  • Modal Operator Editor
  • Property Editor
  • Side Panel Editor
  • Toolbar Editor
  • Hiding panels
  • Extending panels
  • Floating panels
  • Pie menus with 8+ buttons
  • Sticky Keys
  • Stack Keys
  • Custom Icons
  • And more…

Informative videos can be found here.

2 Likes

Thanks much @Metin_Seven. I put quite some time already into looking at the addons features, so I know that already. I was more stunned that it shall not be limited to the eight octants as I’ve never read about it being different here and didn’t want to let it sound that the tool is bad just because I think that pie menus themselves have some problems. (And yes it really is listed in there. I overread that.) If you could make a check with eg putting sculpt masking to a sticky key and tell me if that works flawless, that would be a real decision support for me.

Hmm… I can’t seem to get sticky keys to work like in this video. I’ll post a question in the PME thread. If you keep an eye on that thread, the answer will follow soon.

1 Like

Thanks much for that. I’ll have a look on that.

1 Like

My little wishlist :

  • a symetrize tool based on topology, that works with multires modifier
  • a mask face sets border tool
  • An eraser brush to erase multires data locally (useful when you need to change topology, and avoid peaks, and corrupt sculpt)
4 Likes

The sculpt vertex colors patch has been updated

  • Fix smear brush falloff
  • Fix smooth brush falloff
  • Fix smear smooth mode
  • Make smear faster, only store previous nodes
  • Fix skip surface sampling optimization for paint
  • New curve preset for smear
  • Remove experimental code in the paint brush. This was affecting performance a lot.
  • Fix mesh restore for color data, should fix anchored and drag dot storkes
  • Fix wrong data update in flush updates
  • Fix rendering when switching to Workbench from EEVEE
  • Rebase

What is starting to give this a higher chance of being committed to master, in my opinion, is that feature accumulation appears to be slowing down a lot in favor of fixing what is there.

11 Likes

here’s a case for that feature :innocent:

https://developer.blender.org/T76302

I think it is not correct to look as an example always on cube. Here is I did some comparison on more real examples.


For me in 90% reconstruct works better then apply base , it try to be more close to high poly model then apply base. For example look on lips - in apply base lips getting too much close to each other and overlapping too much.
But If you want to add subiv( not multires) to it maybe some times apply base works better. I think it can be used in more specific cases.
For me by default apply base should have the same result as now have reconstruct algorithm. And maybe make some checkbox to turn old method of apply base for specific cases . and of course apply base by default must be automatic, and only with checkbox it turns on to manual mode

9 Likes

I strongly suggest you post this in a comment on https://developer.blender.org/T73317 just so it’s seen because yeah, you’re on to something here. Maybe @ pablo in it too.

2 Likes

For me in 90% reconstruct works better then apply base ,…

What is this reconstruct you are talking about?

I mean rebuild subdivision . you can apply multires - add it again - rebuild subdivision. result on level 0 will be better then if apply base ( for my taste )

2 Likes

Made the same observation while testing. The current Apply Base algorithm is pretty bad when compared to reconstructing subdiv levels with MultiRes. Once Delete Lower is added it is more effective to just delete level 0 and than rebuild it. I hope they reuse the new algorithm so everything will be consistent.

2 Likes

I am waiting for the same :grinning:

Good work guys. I haven’t touched Multires much yet, but your feedback is valuable for fixing the modifier and becoming a force to be reckoned with. Please keep submitting your observations to the issue tracker, thanks! :+1:

1 Like

To me this is a fundamental flaw in how the system is supposed to work. Far from a feature, I would call it broken.

So this is in 2.90 now already? The full rebuild/project feature he showed a couple of days ago?

yes) I am using 2.9 only because of this feature and because multires levels back.

I have posted a comment on https://developer.blender.org/T76302
with my opinion. Make tests only on cubes is totally wrong