There are merge options in mirror, array …
At some point merge modifier code is already here, well tested and proven to work.
Why over-complexify other modifiers where making this one stand alone not only will reduce complexity of others but aslo allow more things ?
When you work daily with blender on professional jobs, waiting 2-4 years for an important tool is equal to not having it at all.
If for you that seems more than reasonable, for me that seems silly since it’s a pretty simple modifier and the code/patch already exist.
It is a bad excuse. If they make the modifier more efficient with what data is created it will probably be added. If a patch is created we can ask bf-admin and we can probably ask the artists in the blender animation studio what they think.
2-4 or 5-6, that can be maybe 8-10 if we wait to the everythings nodes
I do work with blender professionally daily as well. I’ve missed this feature as well. I just think that it seems like it’s on the road map and I think it may come faster than we think. Jacques is making particles nodes as we speak, and as I understand it, that’s half way done, or so. If I’m not mistaken the progression of nodes is:
Particles first, modelling second and constraints third. So it would be next in line I think. I wouldn’t be surprised if that only took 1-2 years.
I just feel we might be bemoaning the lack of this feature so much we might coax then into making it for us now, which as it sounds, will be kind of inificient as a modifier only to be probably re-written as a node later on.
If you seriously need this modifier at this very instant, you can do what I do sometimes and just create things in houdini apprentice version, then export out to blender. It’s free after all and does all that.
It’s on the roadmap since 10 years ago. Be care with this type of thought, because it could give you bad moments. 
And if you don’t want to use nodes?
It’s great for some kind of modeling but not all and it’s slower than stack.
You all talk about nodes but it’s not the only way, check houdini, nodes are great but people prefer using blender for direct modeling, it’s faster.
I also use houdini but it’s dumb to export when it could be in blender directly without breaking non-destructive workflow.
Nodes will create a modifier that can be added to the stack not instead of the stack
Yes. Nodes are not good for everything, and that’s something a lot of people don’t get.
There’s a big chance of blender become a very unattractive piece of 3D app if they go too deep with that everything nodes idea.
A stack can be seen as nothing more than a single chain of nodes, i think we’re getting way to heated over stacks vs nodes and which one may perform faster, none of the code has been written yet so it’ll just be a guessing game at best.
Oof, that long huh? If this makes any sense, I feel that back in the day it could have been implemented, but right now I definitely understand the hesitation before modelling nodes.
Yea, that’s definitely fair. Directly modelling is always faster, but I don’t think that’s because of any virtue in the modifier stack. Nodes would just work in the same way as applying a modifier. You do an action and it is automatically added in a “stack”, so it would still go just as fast. The issue with Houdini is that modelling in the viewport, that is to say not explicitly typing out the node name in the tree and connecting it, is that it’s not convenient. It’s a pain to be perfectly frank. I’m hoping that since blender is coming from a background of being not fully nodal that we will maintain a good 3D viewport workflow. But if it’s flexibility vs speed, personally ill probably take flexibility in the long run.
I hear that.
It’s not about performance, it’s about usability.
I think we can say that everybody understand the need of this modifier.
Some are willing to wait for everything nodes, other would prefer to have it right now.
Yea that’s pretty much where it stands. And that’s fine I think. It’s just up to the devs.
A stack is a simple linear node system, I don’t see the problem, is like edit materials in the panel of materials or in the shader node. You will have a modifier stack that if you want you could see in nodes like a chain in the classical node editor.
If you could “link” modifiers together so that only one set of mesh data is created, act as one modifier. This would be useful for limiting that amount of data that is stored if you need to use a bunch of modifiers
This is a problem that is not “they do not want to give users what we ask”, it comes from far. It is an implicit problem to the blender and Open Source development model that the foundation has never seemed to intend to solve and has always preferred to avoid when we have talk about that. And that is the main impediment that blender has to grow in the real world.
The problem is the Open Source development model, its financing and everything that has derived from it in the case of blender. And trying to open the eyes about this is very difficult.
The first problem is that blender is free, and it is not free the problem itself, is what this means for development. Someone who sells a product tries to improve their product for their users because he wants to keep selling it.
Since it is free, the foundation does not need to satisfy the users. At least in the same way that other company will do. That means what we can see in this thread, that a simple tool takes there a decade when it saves thousands of hours of work.
The developers work on what they like and, unlike a job, here nobody wants to do the worst tasks. And anything that is not interesting is less probable to be done. For example, the API to be able to make addons that use modifiers, for the interface,… We have seen a lot of times that “we only need somebody that wants to do it”.
There is no need to take users into account. Until the creation of this forum, and even so, ideas are not really shared between users and programmers. Only Blender animation studio artists seem to have some voice over what they need. And besides being few are mainly animators and do not do the work that interests, games, architecture, VR. And there is nothing more impervious to other points of view than a programmer, sorry, is the truth, we live in completely different worlds. It’s really hard that a programmer could see the importance of a tool in the same way that an artist rarely will understand what a coder needs.
That is to say, what there is is a lack of connection between the development of blender and the needs of its users. Exactly the same thing that happens in Linux for desktop and so it is not almost used.
Linux and other free software are very different from blender. Most of the free software that works its developers use it for their personal use or a company pays them to develop it. The fact of being free is the least important, without money would be equally dead. Android doesn’t work because it’s Open Source, it works because it has Google and the whole mobile industry behind it.
Also we add the problem of the GPL. As @BD3D said very accurately if blender is something is closed, in reality because of the GPL you get just the opposite to what you want to get with it. To make a change in blender is simply a long process, difficult and that can lead in the oblivion.
The GPL forces everything to be GPL. And blender programmers can work for the love of art, but the rest of programmers in the world can’t. The GPL forces everything to be GPL. An example is the branch of unwrap, which was basically finished and the developer got a job and took off. As no one wanted to take care of the changes was forgotten and there continues. A branch that works almost perfectly and makes unwrap much better than the current ones. This has happened countless times.
Instead of trying, no matter how difficult, to change the license to LGPL is not done. When it would be undoubtedly the best thing that could happen to blender. Allow third party software for payment. I understand that between some and your love for free software this won’t be important. But I would like to be able to have simulations of fluids, hairs and pay for addons.
This software of payment would be in charge of replacing, at least temporarily, all the lacks of blender. But apparently the philosophy of blender is that the shortages have to endure all, for years, and that will allow somebody be happy.
In short, that the current problem is the disconnection between user needs and what programmers do. If we actually see almost the best of blender2.81 is going to be all the improvements of Pablo dobarro in sculpt, and these are because he uses blender for his models, so the same implements what he needs. But is that a programmer and good artist, is very rare to find. And sooner or later, especially living in Spain where there is no technical artist, someone will offer him 60-80k annual salary and disappear from our lives.
In short, blender has to solve the development problems it has to allow the development of third parties, to be able to be also paid, and that the development of anything minimally more complex than a python script can be done without going through the monolithic framework that is the core development of blender and annoying developers.
And to solve the problems of coordination between the needs of the users and the development of the program. That solutions have already been given and have never been accepted.
One think I like on other software are tags, you add a modifier to a tag and all the object with this tag have this modifier.
But it’s off-topic, let’s focus on Merge modifier!
So, devs ( @ideasman42 @Ton ), it’s up to you, I think you can see why we desperate ask for the merge modifier, it’s not a trivial request for something we will not use.
There is several possibilities:
I don’t know if the previous patch can be used for 2.8, sadly I don’t know how to make a modifier in C, I only know python.
I know some people who made it in forks, but they don’t seems to want to make a patch.
So, if you are ok to add a merge modifier, even if it will not be as good in term of perfs, we can try to find someone to make the patch or you can make it dirrectly yourself.
I see this modifier with one slider for distance and vgroups (+ invert vgroup)
As simple as that.
So please, hear us, make blender the best software and give the users more possibilities to make it append.
Pablo and other always say, make it yourself, but it’s not always possible, Making patches for blender isn’t like making addons, it’s for experienced coders, not for people like us.
So, I really hope you will hear us.
Thx for reading this!
I have nothing against this modifier personally, I won’t use it, but i’m convinced other people will make good use of it.
However i am slightly worried that giving the loudest users their pet requests will set a rather unsustainable precedent for the next time someone wants something *NOW*
In my previous message I commented that the problem is that right now, because of the development model, you have to be the one that shouts the loudest so that it is heeded. Instead of what one does in any other program, which is to see how someone implements it and buy it.
It’s going to stay that way as long as the only ones who can implement things are the blender developers.