Hi! In previous video you said that soon come some news about Viewport Compositor...)))
3
j
jeroenbakker
Today I was helping out Omar with Nvidia support. We are currently not sure if we want to proceed the current implementation and desired to write an updated design. No eta can be told right now.
r
rouellenicolas
Hello Pablo, hope you doing great today, i just want to know if you have any news from the physics nodes project, thanks for your time and the show
3
m
moko
Hay Pablo!
I'm having a big problem
When i finish a project, the way i back it up.
i pack resources in blender. And then unpack resources,
the problem with this is if i have 2 different materials that have images with the same name as "diffuse" when i unpack resources it will just save one image, and in a big project i will not know till it is too late
2
n
nubnubbud
it would be nice to have an autorename, internally, like diffuse, diffuse_1, but that's a really, really bad idea generally. it will also mean unpacked files will differ from packed files, which can mess things up!
organize your files, use standard naming conventions! you will thank past you.
m
moko
i download many textures i cant remember all the names to know if i have a duplicate
b
bestelix
I was confronted recently with this problem, and I'm forced to keep identical names because of the workflow. The solution I'd like to have would be automatic folders for each material inside the "textures' folder. I think this can be an addon.
m
moko
that is a problem because you will have 100 folders
b
bestelix
As long as I can identify what texture goes to a specific material without changing the strict naming convention I work with, I would be happy with many folders instead of hundreds of textures.
hey pablo
why was the blender Particle Nodes abolished and is geometry nodes a replacement for that? how did the blender decide on the vast server?
thank you.
1
n
nubnubbud
nothing with particles has been removed yet, just, it's been largely abandoned for now. geometry nodes will include particles, and probably physics too. Just, there's no reason to upgrade it if it's high on the list for being included in a different area.
t
thinsoldier
Helloo Pablo!
Do you ever get annoyed by the editor list closing because you put your mouse in the wrong place too fast?
When this menu is open there should be an even-on-all-sides dead zone so that an inaccurate mouse movement doesn't inadvertently close the menu before you can make your selection.
1
s
slowk1d
Dude it's the second time you post 5 different questions here.
I understand you want Pablo to answer to more than one question, I do too sometimes, but in this way you over-fill the Q&A with posts, taking away from others the chance to have their only one question to be read. I politely ask you to limit your questions to just one, or at least if you have more than one to write them in a single post, thanks.
1
t
thinsoldier
Having never had a question read, I don't see the problem. Especially when so many people post 3 and 4 questions in a single post.
If nobody votes my individual posts then they will not be read.
That is fair.
If somebody votes a post with 3 or 4 or 5 questions in a single post, that person will get all 4 of their questions answered.
F
Fred
double thumbs up for the auto smooth! Thanks a lot for this! :)
F
FreeMind
Hey bro. Question. Ever since 2.8, everything that has anything to do with grouping objects together was replaced with collections. Now that Light groups are a thing, why aren't they based on collections like everything else? Wouldn't that be consistent?
The workflow I imagine: You place all your lamps and emissive objects in a collection. You mark that collection as a Light group. And boom, ya have a light group What's the downside of that?
t
thinsoldier
a single light can exist in infinte collections, this complicates everything.
I really wish they would stop everything and reevaluate the collections / real groups situation, fix/improve that and then go back to working on other things
F
FreeMind
A single light can exist in infinite collections, but you may use only one of those collections as a light group... So what's the problem?
Also, currently, a light can't be in more than one light group at a time. Why? I don't see why this can't be allowed.
Yeah, merging the two light groups with the same light wouldn't work... but so what?
n
nubnubbud
oi, Pablo!
is there any news on geometry nodes presets/premade node groups or possibly, dare I say, a direct way to share node groups as files? it can be a little clunky appending them from files manually. (maybe "bulk append" could be an addon to work on...)
also on that note, I may have asked already, but what is your take on having multiple solvers with overlapping uses in blender? powerful? better left to addons? what are your thoughts on uber solvers like Vellum, that are inter-material and matter phase? how about mantaflow updates?
also, any news on the baking workflow?
sorry about the potpourri questions, but I'm really curious!
p
pxy-gnomes
Hi, Pablo; ¡Fantastic Youtube show!
On Rigging topic, ¿is there any plan for improvements of Shape Keys' functionalities in Blender —or maybe, even, new alternative Rigging solutions for Shape Keys?
For example, ¿wouldn't it be nice if (a mesh Object's) Mirror Modifier could be Applied to a character with Shape Keys?
Another example: what if Shape Keys could have in-between (A→B→C… "Keys"), instead of just linear A→B; so that 3D Animators could develop (inside Shape Keys already) easier and "Non-Linear" Shape Keys (like if "B" where a "Breakdown" animation frame in-between "A" and "C"; "Breakdown" as the frame concept from 2D Animation, like in the image below, plz check the red, middle Breakdown drawing).
PS: I've heard Maya's "Blend Shapes" would have that last functionality btw.
Thank you.
j
jamesa
Hi Pablo!
Do you think Animation Nodes will ever be implemented into Blender?
Thanks!
t
thinsoldier
It's a long way off. I think a lot more curve and hair specific nodes are being worked on first, then something else, then physics nodes, and then maybe scene/object nodes and finally stuff similar to Animation Nodes would be built on top of all of those.
o
ocular.evolution
Hola Pabz! Hope you're havin' a swell time over there! :)
My question is - why isn't the state of the outliner saved when saving a file? Every time we open a file after starting Blender, all of the collections have their dropdown state as 'active'. This seems like a minor issue, but having to 'deactivate' the dropdown state of all the collections every time a file is opened gets annoying lol. It almost defeats one of the aesthetic importance of having collections - for better organization and reduction of clutter, visually.
Have a great week and weekend! :D
G
Gary
grash@Pablo / @Blender you need to check the buildbot 5/13 was last build
j
jeroenbakker
Yes. We are looking into it. But during weekends we are mostly doing other stuff ;-)
t
thinsoldier
Hi Pablo.
Amaranth Bug Report or maybe a feature request.
Lighter's Corner is not mentioned in the description of the addon so it took me forever to figure out which addon created the Lighter's Corner.
Lighter's Corner does not list meshes using materials that use the emission property of the Principled BSDF shader node.
r
reanimate
Hi Pablo,
I got some question about blender physics simulation regarding upgrade or performance improvement.
I read somewhere (i forgot where actually), that some of blender physics codes or algorithm that currently used are never been updated (i guess only bugfixing?) and some of them are actually still using the algorithm from the beginning of blender i guess except mantaflow?(sorry if i might be wrong about it) Is there any plan to update them and using more modern optimize approach? i guess for example are maybe RBD, RBD Contraints, Cloths, and Softbody?
Is there a plan for some kind of Simulation/Physics Nodes for object based or scene based simulation for blender?
Baked/Cached High density( grid? ) fluid simulation playback seems very slow, is there any plan to improve the performance for that?
Sorry for a lot of question, i'm really wondering about blender physics simulations since the performance and workflow are quite slow, and sometimes kinda limited.
-Thanks-
t
thinsoldier
See what I mean @slowk1d ?
m
meszarosmate0510
Hi Pablo!
With the new hair tools being a main focus of the Heist project, can we hope to have simulations, animation and all currently available hair options available with this new approach, by the time the movie comes out?
The progress is amazing, just hope it doesn't stop before it's "finished".
t
thinsoldier
Hi Pablo! Feel free to skip this question if you don't have time.
Collections pull double duty as both a tool for re-using things and for organizing things whether or not they will be re-used.
To keep such things separate I add a $ to the names of collections intended for reuse as collection instances. This has the added benefit of floating such collections to the top of the "add collection instance" list.
Similarly, groups in the material editor and geometry node editor pull double duty as both a tool for re-using things and for organizing things. The dozens of groups I made just to make the node networks easier to read make it painful to find the 1 node group in my file that I need to re-use, unless I impose some kind of naming convention and stick to it.
There need to be
intended for reuse entities
in the system that are
separate from
the
organization helper entities
.
What do you think?
m
maxedge
Hi, how are you?
d
dino107
hey pablo when will blender add vdm vector displacement maps to blender vdms are basically 3d alphas would love to see them in blender
t
thinsoldier
It's at the top of the list of the sculpt roadmap:
For people with multiple large monitors I think an easy way to open Editors in new windows would be useful. What do you think?
d
dino107
when will blender add quad dyntopo support for multires because that can lead to great sculpting . being able to add new geometry without messing up the uv map and being to subdivide to avoid problems that dyntopo users encounter can shorten the worlflow and can lead to better results and will make blender a powerful sculpting software
t
thinsoldier
I know I saw quad dyntopo mentioned somewhere but not in relation to multires and it's not on the roadmap:
developer.blender.org/T93226
But I know I saw someone somewhere talking about dyntopo + multires and separately dyntopo + quad generation.
Hi Pablo, I hope you are doing great! Thank you for the shows!
I want to ask you is Hair Nodes a priority now? Will the new Hair System be more perfomant than the current one?
Also what happened with the performance improvements in Sculpt Mode Pablo Dobarro did a time ago. Did it make it to master branch?
t
thinsoldier
I don't think it did. The person working on sculpt currently made some DynTopo speed improvements and I think they were working on combining all the speed improvements and polishing Pablo Dobarro's unfinished improvements. See the sculpt roadmap:
developer.blender.org/T93226
2
s
slowk1d
Hola Pablo!
I wanted to share a thought about the current development of Blender. This is a personal impression after following the Blender development and its releases since 2018.
As much as I love and admire how Blender has progressed in the last 3 years, I'm starting to notice a tendency, especially since last year: many new features have been introduced, but along with them every new release comes out with bugs or regressions that sometimes are quite important and stick for an entire release (most recent example:
developer.blender.org/T95428
). I feel, despite lots of people still argue that Blender development is going slow and we're missing a lot of basic features, that actually the development is rushing a lot, to the detriment of stability. Some newly introduced features cause regressions that then aren’t solved even after 3 new releases; sometimes they feel not totally polished, and sometimes they change face (UI and workflow) even more than once with every release. Now, it's good that features can be improved even in the span of one release, and of course the devs have to give a higher priority to bugs that cause crashes or make impossible to use Blender, but sometimes I feel it would be better to have maybe less releases per year (or wait more before merging new stuff in master), and focus more time on polishing features, closely watch out for bugs that they could cause, and more in general give more time to new stuff to be tested out. As much as I would love to have light linking or multi-object property editing right now, I would prefer to wait more for something that is stable and won't need a fix in a new release the exact time is merged in master. Hope I didn't annoy to much with this long text, just an impression I hope can be food for thoughts.
Thanks for the time and for the hard work you do!
t
thinsoldier
Then there are the people who post whole essays and eat up question time just by having to read the wall of text.
n
noname
as blender team do now - it is better way.
They add feature fast and if you or users find bugs in new feature - you can freely report a bug.
They fix bugs - again - fast.
This is much better than NOT add new features. After add and fix than NOT add at all.
I say more - i think that blender developers for my taste add features - SLOW - my example is vulkan - words about implementation appear 3-4 years a go and even now we use silly openGL instead of enjoy realtime RTX GI, "just-in-time" shader compilation and all those features what appear in "unreal 5" engine - we can have some of it in EEVEE but devs add new features - SLOW.
44 Comments
Hi! In previous video you said that soon come some news about Viewport Compositor...)))
Today I was helping out Omar with Nvidia support. We are currently not sure if we want to proceed the current implementation and desired to write an updated design. No eta can be told right now.
Hello Pablo, hope you doing great today, i just want to know if you have any news from the physics nodes project, thanks for your time and the show
Hay Pablo!
I'm having a big problem
When i finish a project, the way i back it up. i pack resources in blender. And then unpack resources,
the problem with this is if i have 2 different materials that have images with the same name as "diffuse" when i unpack resources it will just save one image, and in a big project i will not know till it is too late
it would be nice to have an autorename, internally, like diffuse, diffuse_1, but that's a really, really bad idea generally. it will also mean unpacked files will differ from packed files, which can mess things up!
organize your files, use standard naming conventions! you will thank past you.
i download many textures i cant remember all the names to know if i have a duplicate
I was confronted recently with this problem, and I'm forced to keep identical names because of the workflow. The solution I'd like to have would be automatic folders for each material inside the "textures' folder. I think this can be an addon.
that is a problem because you will have 100 folders
As long as I can identify what texture goes to a specific material without changing the strict naming convention I work with, I would be happy with many folders instead of hundreds of textures.
Hey Pablo,
In the last two years, from time to time I ask you about this idea, could somebody implement it?
StylusUX - pressure sensitive & pen-optimized interface elements & tools
While recently Blender has been improved with some touch features, this one might be a timely one now.
hey pablo why was the blender Particle Nodes abolished and is geometry nodes a replacement for that? how did the blender decide on the vast server? thank you.
nothing with particles has been removed yet, just, it's been largely abandoned for now. geometry nodes will include particles, and probably physics too. Just, there's no reason to upgrade it if it's high on the list for being included in a different area.
Helloo Pablo!
Do you ever get annoyed by the editor list closing because you put your mouse in the wrong place too fast?
When this menu is open there should be an even-on-all-sides dead zone so that an inaccurate mouse movement doesn't inadvertently close the menu before you can make your selection.
Dude it's the second time you post 5 different questions here.
I understand you want Pablo to answer to more than one question, I do too sometimes, but in this way you over-fill the Q&A with posts, taking away from others the chance to have their only one question to be read. I politely ask you to limit your questions to just one, or at least if you have more than one to write them in a single post, thanks.
Having never had a question read, I don't see the problem. Especially when so many people post 3 and 4 questions in a single post.
If nobody votes my individual posts then they will not be read. That is fair.
If somebody votes a post with 3 or 4 or 5 questions in a single post, that person will get all 4 of their questions answered.
double thumbs up for the auto smooth! Thanks a lot for this! :)
Hey bro. Question. Ever since 2.8, everything that has anything to do with grouping objects together was replaced with collections. Now that Light groups are a thing, why aren't they based on collections like everything else? Wouldn't that be consistent?
The workflow I imagine: You place all your lamps and emissive objects in a collection. You mark that collection as a Light group. And boom, ya have a light group What's the downside of that?
a single light can exist in infinte collections, this complicates everything.
I really wish they would stop everything and reevaluate the collections / real groups situation, fix/improve that and then go back to working on other things
A single light can exist in infinite collections, but you may use only one of those collections as a light group... So what's the problem?
Also, currently, a light can't be in more than one light group at a time. Why? I don't see why this can't be allowed. Yeah, merging the two light groups with the same light wouldn't work... but so what?
oi, Pablo!
is there any news on geometry nodes presets/premade node groups or possibly, dare I say, a direct way to share node groups as files? it can be a little clunky appending them from files manually. (maybe "bulk append" could be an addon to work on...)
also on that note, I may have asked already, but what is your take on having multiple solvers with overlapping uses in blender? powerful? better left to addons? what are your thoughts on uber solvers like Vellum, that are inter-material and matter phase? how about mantaflow updates?
also, any news on the baking workflow?
sorry about the potpourri questions, but I'm really curious!
Hi, Pablo; ¡Fantastic Youtube show!
On Rigging topic, ¿is there any plan for improvements of Shape Keys' functionalities in Blender —or maybe, even, new alternative Rigging solutions for Shape Keys?
For example, ¿wouldn't it be nice if (a mesh Object's) Mirror Modifier could be Applied to a character with Shape Keys?
Another example: what if Shape Keys could have in-between (A→B→C… "Keys"), instead of just linear A→B; so that 3D Animators could develop (inside Shape Keys already) easier and "Non-Linear" Shape Keys (like if "B" where a "Breakdown" animation frame in-between "A" and "C"; "Breakdown" as the frame concept from 2D Animation, like in the image below, plz check the red, middle Breakdown drawing).
angryanimator.com/pic/breakdown_01.jpg
PS: I've heard Maya's "Blend Shapes" would have that last functionality btw.
Thank you.
Hi Pablo!
Do you think Animation Nodes will ever be implemented into Blender?
Thanks!
It's a long way off. I think a lot more curve and hair specific nodes are being worked on first, then something else, then physics nodes, and then maybe scene/object nodes and finally stuff similar to Animation Nodes would be built on top of all of those.
Hola Pabz! Hope you're havin' a swell time over there! :)
My question is - why isn't the state of the outliner saved when saving a file? Every time we open a file after starting Blender, all of the collections have their dropdown state as 'active'. This seems like a minor issue, but having to 'deactivate' the dropdown state of all the collections every time a file is opened gets annoying lol. It almost defeats one of the aesthetic importance of having collections - for better organization and reduction of clutter, visually.
Have a great week and weekend! :D
grash@Pablo / @Blender you need to check the buildbot 5/13 was last build
Yes. We are looking into it. But during weekends we are mostly doing other stuff ;-)
Hi Pablo.
Amaranth Bug Report or maybe a feature request.
Lighter's Corner is not mentioned in the description of the addon so it took me forever to figure out which addon created the Lighter's Corner.
Lighter's Corner does not list meshes using materials that use the emission property of the Principled BSDF shader node.
Hi Pablo,
I got some question about blender physics simulation regarding upgrade or performance improvement.
I read somewhere (i forgot where actually), that some of blender physics codes or algorithm that currently used are never been updated (i guess only bugfixing?) and some of them are actually still using the algorithm from the beginning of blender i guess except mantaflow?(sorry if i might be wrong about it) Is there any plan to update them and using more modern optimize approach? i guess for example are maybe RBD, RBD Contraints, Cloths, and Softbody?
Is there a plan for some kind of Simulation/Physics Nodes for object based or scene based simulation for blender?
Baked/Cached High density( grid? ) fluid simulation playback seems very slow, is there any plan to improve the performance for that?
Sorry for a lot of question, i'm really wondering about blender physics simulations since the performance and workflow are quite slow, and sometimes kinda limited.
-Thanks-
See what I mean @slowk1d ?
Hi Pablo!
With the new hair tools being a main focus of the Heist project, can we hope to have simulations, animation and all currently available hair options available with this new approach, by the time the movie comes out?
The progress is amazing, just hope it doesn't stop before it's "finished".
Hi Pablo! Feel free to skip this question if you don't have time.
Collections pull double duty as both a tool for re-using things and for organizing things whether or not they will be re-used.
To keep such things separate I add a $ to the names of collections intended for reuse as collection instances. This has the added benefit of floating such collections to the top of the "add collection instance" list.
Similarly, groups in the material editor and geometry node editor pull double duty as both a tool for re-using things and for organizing things. The dozens of groups I made just to make the node networks easier to read make it painful to find the 1 node group in my file that I need to re-use, unless I impose some kind of naming convention and stick to it.
There need to be intended for reuse entities in the system that are separate from the organization helper entities .
What do you think?
Hi, how are you?
hey pablo when will blender add vdm vector displacement maps to blender vdms are basically 3d alphas would love to see them in blender
It's at the top of the list of the sculpt roadmap:
developer.blender.org/T93226
Hello Pablo,
Do you think mesh objects should have an Active Element tab in the Properties Editor just like Metaballs?
Mesh objects have the properties for the currently selected vert/edge/face-edges only available in the sidebar of the 3d viewport.
Metaballs have properties for the currently selected ball in both the sidebar and the properties editor:
I think it would be consistent to have such a tab available for mesh objects also.
Hello Pablo.
I noticed that very many people making tutorials do not know about the essential "Set Offset from Cursor" feature of Collections.
It would help greatly if it was also in a more logical and easier to find location like the Collection Properties tab.
For anyone wanting to know more about "Set Offset from Cursor" and Collections: blenderartists.org/t/is-there-anything-else-you-should-know-about-collection-instances/1302683
Hello Pablo. Happy Monday!
For people with multiple large monitors I think an easy way to open Editors in new windows would be useful. What do you think?
when will blender add quad dyntopo support for multires because that can lead to great sculpting . being able to add new geometry without messing up the uv map and being to subdivide to avoid problems that dyntopo users encounter can shorten the worlflow and can lead to better results and will make blender a powerful sculpting software
I know I saw quad dyntopo mentioned somewhere but not in relation to multires and it's not on the roadmap: developer.blender.org/T93226 But I know I saw someone somewhere talking about dyntopo + multires and separately dyntopo + quad generation.
Try lurking in blender.chat in #sculpt-paint-texture-module and following the activity of developer.blender.org/p/joeedh/ + youtube.com/user/joeedh/videos
Hi Pablo, I hope you are doing great! Thank you for the shows!
I want to ask you is Hair Nodes a priority now? Will the new Hair System be more perfomant than the current one?
Also what happened with the performance improvements in Sculpt Mode Pablo Dobarro did a time ago. Did it make it to master branch?
I don't think it did. The person working on sculpt currently made some DynTopo speed improvements and I think they were working on combining all the speed improvements and polishing Pablo Dobarro's unfinished improvements. See the sculpt roadmap: developer.blender.org/T93226
Hola Pablo!
I wanted to share a thought about the current development of Blender. This is a personal impression after following the Blender development and its releases since 2018. As much as I love and admire how Blender has progressed in the last 3 years, I'm starting to notice a tendency, especially since last year: many new features have been introduced, but along with them every new release comes out with bugs or regressions that sometimes are quite important and stick for an entire release (most recent example: developer.blender.org/T95428 ). I feel, despite lots of people still argue that Blender development is going slow and we're missing a lot of basic features, that actually the development is rushing a lot, to the detriment of stability. Some newly introduced features cause regressions that then aren’t solved even after 3 new releases; sometimes they feel not totally polished, and sometimes they change face (UI and workflow) even more than once with every release. Now, it's good that features can be improved even in the span of one release, and of course the devs have to give a higher priority to bugs that cause crashes or make impossible to use Blender, but sometimes I feel it would be better to have maybe less releases per year (or wait more before merging new stuff in master), and focus more time on polishing features, closely watch out for bugs that they could cause, and more in general give more time to new stuff to be tested out. As much as I would love to have light linking or multi-object property editing right now, I would prefer to wait more for something that is stable and won't need a fix in a new release the exact time is merged in master. Hope I didn't annoy to much with this long text, just an impression I hope can be food for thoughts.
Thanks for the time and for the hard work you do!
Then there are the people who post whole essays and eat up question time just by having to read the wall of text.
as blender team do now - it is better way. They add feature fast and if you or users find bugs in new feature - you can freely report a bug. They fix bugs - again - fast. This is much better than NOT add new features. After add and fix than NOT add at all.
I say more - i think that blender developers for my taste add features - SLOW - my example is vulkan - words about implementation appear 3-4 years a go and even now we use silly openGL instead of enjoy realtime RTX GI, "just-in-time" shader compilation and all those features what appear in "unreal 5" engine - we can have some of it in EEVEE but devs add new features - SLOW.