I quit C4D because the developers are developing the product while ignoring user requests.
I decided to go back because geo-nodes are not as friendly as mograf. without mograf it becomes more and more difficult to work. the modern world requires ultra-fast construction of models and scenes.
Let’s be honest, it’s hard to clone a group of objects in Blender even in a circle.
I notice how things that are specific to Blender that C4D doesn’t have are slowing me down.
The two things I still use Blender for are Linux compatibility (I love KDE) and Luxcore.
That is weird to hear somebody who has used Blender during 5 years without geo-nodes, losing patience, only after several months of geometry nodes existence.
I don’t deny that Blender’s workflow may be slower.
But it is not hard to clone a group of objects in Blender even in a circle.
- Use 3D cursor as pivot point. Duplicate Rotate with wanted Angle value and Repeat.
- Use a Circle as a Dupliverts parent of a Collection Instance.
- Use a nodetree made of 3 nodes ( Circle, Align Rotation to Vector, Point Instance) with geometry nodes.
None of these 3 methods is complicated. They are slow. But they can be automated, turned into a script, an addon or a reusable nodegroup.
I understand that you tried to illustrate your pain. You are probably struggling with lots of other stuff.
And if you have to conform Blender to your habits for each little thing, you may be fed up of that.
I was asked “why?” and I said.
I am not pouring out any "pain.
I know how to work in Blender. I also know how to clone objects.
I’m saying that it’s noticeably faster with blender.
I’m saying that geo-nodes couldn’t replace the blender for hard-surface modeling tasks, which are close to CAD.
I’m not spoiling, I’m making money, providing for my family. This is a very deliberate decision that is under consideration.
It’s a process where every task has a time record. Every project is analyzed for weaknesses. I’m a freelancer with an hourly wage. It’s a very, very deliberate decision.
3 Likes
I don’t want to persuade you either way, but I think hardops is a good addon that might be helpful for this kind of workflow if you still want to use Blender. What I do recommend though is using multiple tools if one isn’t sufficient. So rather than switch, add.
I too think that scene and object managing in Blender is still bad and your criticism is valid, especially if compared to how C4D is doing things. I miss duplicating whole groups with modifiers attached to the whole group, something that still isn’t possible in Blender. (Like mirroring a whole group of objects with one modifier or adding a subd modifier to a group with tons of objects) 
I am using more and more Maya these days, but that is because I am interested in pushing into the high end segment and Blender isn’t getting the performance regression fixed, while also lacking in certain areas without progression in sight.
That’s a good point, too. Why are you loosing patience now, shouldn’t you already have lost it years ago? 
Speaking from my own experience (also coming from C4d 6 years ago). The reason why I don’t get too disappointed with Blender is because it enables me to brute force through large amounts of modelling work that I wouldn’t have been able to do faster in C4d (quite the opposite actually). So the overall level of energy i spent is still in the plus in comparison.
What makes me angry though is knowing that if Blender would get some shit fixed, it could be even much more faster and efficient than it already is and seeing that lost potential is painful.
The fact that the BF isn’t looking at the competition and trying actively to adapt by copying more, rather than trying to reinvent the wheel out of ideological reasons is extra painful. Like really facepalm inducing painful.
In some areas Blender performs like a Ferrari, in some disciplines its driving like with a pulled handbrake.
Combining Blender with another tool is like jumping in another car when it gets to slow and if the other car is fast in these disciplines the overall speed is optimal.
I can totally relate. The good thing is that Blender is free and will always remain so. So you don’t have to choose either or because of money, you can concentrate purely on efficiency, you have invested time, but you shouldn’t have buyers remorse or other psychological biases.
3 Likes
It’s amazing how free Blender has taught you to pay for software products.
In C4D I have always used pirated versions for the reason that the license for it costs too much. In Blender, all the addons are affordable and easy to buy.
Unfortunately some of the not copying other software isn’t necessarily always an ideological reason. Sometimes it’s of the threat of possible legal action. if this devtalk thread is any thing to go by.
https://devtalk.blender.org/t/copyright-guidelines-for-devtalk/17331
"Did the Blender project ever get in such legal troubles in the past?
Unfortunately… yes. Only very few times. I will not talk about that here though. It is in the best interest of the Blender project and its contributors to keep that information confidential. Luckily the cases faded out and never lead to actions. We were also well protected by legal support from the Free Software Foundation and the Open Invention Network.
However, these situations did help to become better aware of how to share information. The first guideline I added was:
- Always post your ideas and designs on a blender.org website, and create context to which current project you contribute.
And the second:
- Stay away from copying screenshots from commercial apps on blender.org .
The first guideline is to ensure you transfer the responsibilities of sharing the ideas to the Blender Foundation, which is much better suited to act legally than an individual coder.
The second guideline avoids a future conflict with a third party to claim “we steal their copyrights”.
But then what is possible?
Possible is:
- Talk about other programs in general terms based on public available knowledge. For example: “Maya’s layer system offers overrides”.
- Share links to videos or articles endorsed by the rights holders. For example “this promo video shows well how Modo offers character rigging”.
Do not:
- Include screenshots in posts from commercial / non-open products.
- Copy texts, diagrams or other content from websites without their permission.
- Indicate in any way that we should copy external ideas or methods without permission."
Although I’ll be the first to admit we can’t just blame all the bad decision making on possible legal repercussions. Especially when it comes to the lack of focus that is given to performance at times. There’s still been highly questionable decisions made throughout blenders history. But what software is perfect?
And since we have the source code and previous versions downloadable. We can make changes or stay with the version that worked for our use case.
Can’t say that about commercial software. Which can suddenly disappear if the company decides to change business strategies or go bankrupt.
Right, but often you have to buy a ton of those to get on the same level as the vanilla versions of commercial software products provide and the amount of money spent gets in a similar ballpark.
Some guy I know who switched to Blender recently lamented how he had to spent a similar amount of money for addons in Blender than the Indie License of Maya would have cost him, to get on feature parity with it. Of course these addons are permanent licenses (but have additional risk of maybe won’t get developed into many future versions).
Funny how things even out (risk, stability, learning curve investment, time budgets).
That’s what makes it so difficult to compare stuff, it’s all so very individual taking personal preferences into account.
3 Likes
I totally agree!
And this fan of addons makes my head spin.
Every addon needs a little bit of this and a little bit of that.
The result is a stack of addons, buttons, and a cluttered sidebar.
I partially save myself with the pie-menu editor by creating such “sets of favorites”.
But there are too many of them to just work. And the work slows down. There’s no feeling that you’re like a Jedi waving a sword - whack, whack, and the result is ready. There is a feeling that you, like a car mechanic, spend the whole day turning small parts with a huge number of tools in an attempt to start a capricious engine.
1 Like
That argument is valid, but it seems to me that its more an communication problem. I am not well versed in License and patent stuff, but afaik most of the things I have in mind are common software UI interface standards that are so general that there is absolutely no legal ground for legal action.
Like for example, grouping systems with pick walking, selection previews, modifiers that apply to whole hierarchies as well as other very simple industry standard behaviour. The fact that it is a standard that is present in many software’s basically proves that there can’t be legal consequences for following it.
Not practicable for many. I am not a programmer, I’ll never touch code to fix stuff.
Staying with older versions gets complex really fast. I have like 8 versions of Blender installed, with all the different addons for each versions, my Blender preference folders are getting huge and I am completely loosing oversight where stuff is.
Only because of the recent trend of software as a service 
Usual practice in movie studios for example was to stick with the version you started the project with, so some of them were like 5 years behind the newest version.
And since they bought perma licenses they could stick with them forever.
Up until recently there where still some shops using XSI years after its demise.
This limit is technically gone with geometry nodes.
You need to add a supplementary object to handle geometry nodes modifier and as a result, you have a duplication objects of your scene, forcing you to adapt display.
So, in terms of modeling simplicity, result is there.
Now, that is just really a question of improving presentation in outliner and display in viewport.
I don’t know if that would be mentioned in UI workshop.
But a focus on that could facilitate use of geometry nodes.
Technically yes, practically no. Its a whole new level of complexity with how many clicks?
Compare that with C4d where you create a group (1click), then drop a modifier into/onto the group (1click).
Done! That’s it. And the group is like an empty that you can move around.
Blender makes it a complex thing with more than 10 clicks that is spread through several editors.
Its possible, but also kinda horrible.
Yes it opens the door for future greatness, but considering that C4D has this functionality since 15+ years makes my head spin.
You have any idea how many clicks spared that is over the years?
And if I would think about it (I wont!) I probably find a couple of these things that are really REALLY simple concepts, that a 10 year old can understand in one minute.
Blenders way of doing it resembles a reinventing of a wheel, only that its a cube, but there is an tool hidden somewhere that lets you make it a wheel temporarily.
Don’t get me started, I might come up with even more brutal allegories. I hope at least somebody who reads this has a good laugh about it.
I too am NOT too serious right now. I am grinning hard - but come on man.
1 Like
Even if there’s no legal grounds. it still won’t stop a company from suing. Even if they know they won’t win the case. It’s a intimidation tactic. Bully the competition into stopping something with the threat of legal fees. So the Blender Foundation is erring on the side of caution. At least that’s how I understand it.
I’m in the same boat. But having the source there allows the community to decide to do their own thing. Or continue to support features that get removed. Say the game engine for instance. Tangent animations studios has heavily modified cycles for their uses. Malt is shaping up to be a new interesting npr render engine. There’s many examples of just having the source makes for a much better user experience long term. In commercial apps addon developers will usually be limited by the addon scripting language and some attributes not being exposed.
Same. And yet I’m still always running the bleeding edge nightly alpha builds. lol
Yeah this so much this. I hate it with a passion. And now that’s leaking into the real world with cars (tesla) and other stuff. Yay, Ownership as a service.
Which is great. But eventually the software will no longer be able to be run unless you have an outdated os (security risk) and computer. There might not be x86 architecture computers in 5 years. Everything might switch over to arm. Technology changes at a rapid pace.
I do get it, but parts of me believe that they are too cautious.
You now what the BF would need is support from like EPIC games or some other big company.
A public announcement that Epic would send their lawyers if somebody tried to play that game with the BF.
There is this ancient saying: The only thing that stops bad guys and violence, is good guys who are just better at violence.
In regards to your comment about the open source code. I do get this too, I am just saying it doesn’t apply to the single artist that don’t know how to code, which are the majority of artists.
Stupid idea and I hope it flies in their face. If people don’t own stuff, they won’t care about it either.
Communism has proven that without a doubt.
1 Like
Well. You create a collection. You add a mesh. You go to Geometry Nodes workspace. You click on +New button. You append a custom nodetree corresponding to a modifier. You replace datablock.
That is still a lot of extra clicks. But not more than 10 clicks, if you are using a recurrent nodetree.
With nodetree presets similar to current modifiers, we could imagine an automation of that.
A short key opening a menu list of Geometry Nodes modifiers presets.
And if selection in outliner is a collection, choosing one preset could automatically add a mesh holding the modifier.
I meant simplicity compared to what you had to do without geometry nodes (using an empty or armature bone as parent/modifier target, browsing all objects of group to be sure that did not forget one when copying modifiers from active and solving modifier stack order problems).
c4d has a great answer to geo-nodes. I’m talking about “Scene Manager”. At first glance it looks the same, but it’s geared towards people who don’t have a programming mindset.
Seriously, I can not do Animation Nodes or Geo Nodes because I am a designer. I don’t even have a mouse, only a stylus. I think that’s a very good move by Maxon.
I’m sure there will be an addon for Blender that will do the same thing.
But guys. One more addon. Which will also conflict with Python and other addons. Ubrbrbrb…
Yeah. Blender is cool.
But. There are nuances.
1 Like
That’s all fine and dandy, since it comes with the geometry nodes workflow, but as a general workflow for modelling its pretty unintuitive and way to complex.
Here is my counter proposal: the ability to drop an modifier on a collection and have every object in the collection automatically linked to that modifier.
Under the hood every object basically gets its own (linked) modifier (that is invisible in the modifier tab) and the moment you take an object out of the collection Blender removes it.
It’s simple, intuitive and easy to understand.
Of course you could do it all in the geo nodes network, especially if you want more control and more complexity, but if you just want to mirror, solidify, bevel and subdivide a bunch of meshes, this would be IMHO much better, also more user friendly.
I also think it would be very simple to develop. It just works like an override would.
What do you think?
3 Likes
Had a look at C4D’s “Scene Manager” because of your post and it looks awesome indeed! It appears to be very intuitive. That’s pretty cool!
1 Like
The intention of developers is to make “everything nodes”.(modifiers, particles, constraints…)
Currently, there are only 3 nodes under Geometry category, 5 nodes under Mesh category in 2.93.
There are a Subdivide Surface node, a Boolean node and ability to mirror stuff by combining Transform node with a negative scale and Join node.
But Bevel is a lot more complex operator/modifier that would probably take months of work before being translated into node.
This summer, there is a GSOC about porting of popular modifiers to Geometry Nodes.
So, work about Solidify is in progress.
In master, there are 4 more Geometry nodes and a new Curve category.
Between 2.92 and 2.93, Attribute category was filled to make geometry nodes more useful, more capable than just duplicates of old modifiers for animation.
But now, categories with few nodes will expand and they will become more capable of replacing current modifiers for modeling.
Original Idea was that, at end, there should be a node corresponding to each operator available in edit mode, making modifiers nodes having a spectrum larger than current modifiers.
So, modifiers stack was supposed to be condemned to disappear.
On the contrary, it became a mandatory stack of nodetrees.
So, I don’t know how long old modifiers will stay in place.
But if in the end, we could have only modifiers referring to nodetrees.
A linked modifier would only be equivalent to modifiers using same linked nodetree.
I don’t think there would be a need to work on modifiers tab.
Future brainstorming for UI devs should probably be more about making a distinction between linked/single user modifiers, at creation process.
It seems 2.93 finally removed the blatant lie that was “link modifiers” and it now actually says “copy modifiers”. 
I’m not holding my breath for true linked modifers, though, since as you say, for some ungodly reason the new node system has not made them obsolete as was intended, but boy would that be a fundamental thing to work on for the next major Blender release!
That’s all interesting and I know most of it but it doesn’t answer my question.
How about a workflow where I can add an modifier/nodetree-modfier to a large group of meshes in an collection without diving into the geometry nodes editor and referencing them into another object (which is not the same workflow and can be done right now)?
After all we would still need a place in the property editor where we can access and edit exposed parameters of those modifiers/nodes.
So why not have a modifer/node (stack) on an collection to apply it to a bunch of meshes simultaneously.
I think this is missing in the concept and it is IMHO a great and proven idea to make working in Blender smarter and not harder which it is right now.
A 3 click solution to apply something to a large group of objects that is dynamic and easily controllable.
The fact that this concept is not talked about, never mentioned and absent from future plans is making me kinda irritating.
Am I somehow wrong in my applied logic, am I the only missing this and thinking about this?
It can’t be me being selfishly wanting an arbitrary feature that nobody else has any us case for?
This is an exact example of what I meant earlier with what should be copied from other software, C4D in this case.
And on top of it it would be actually kinda trivial to implement since everything is already there.