Open-sourced alternative to zBrush - updates [it ain't happening] + personal advice

I’m not really understanding what your frustration is. That you want an open source ZBrush alternative, and you want it right now?

Because, you said yourself…

And you don’t think a brand new, from scratch sculpting project is going to take vastly more time to be planned, researched, implemented, and at the level of ZB?

Sooo… either way, even if you got your wish and a project started this second, you’d still be looking at several years of waiting. :slight_smile:

4 Likes

Now, I would be fine to wait a long while for an OS version of zBrush. I was trying to collaborate people to start something that would likely take a long time. The way I see it, even if it isn’t happening soon, the sooner it would’ve been started, the farther ahead it would be compared to if such a project started at a later date.

What I meant by the saying,

is that the latest stages of Blender’s development are still working out bugs in things zBrush had worked out it its early stages. Blender is far behind in sculpting, even as a maturing program.

Another thing is that I don’t mind the idea of years of waiting for something that will eventually be capable of what zBrush is currently capable of.

Developing a new program from scratch would take a long time for all of those reasons, but at least it would be developing. Also, zBrush will likely always be ahead because they’ve had a long head-start; I’m more just hoping we can eventually get an open-sourced program that can function at zBrush’s current (2019) capacities some day.

The problem with working on Blender is that Blender is not a sculpting application; it’s a CG application with sculpting capacities. You can improve the sculpting a fair bit but that’s topical; the foundation under the changes is still the same and it will limit what you can do.

For example, lets say a sculpting program like zBrush handles its high poly counts by storing vertex data in a way drastically different from Blender. If you were to try to change the way Blender handles vertex data all together, it would cause problems for the rest of the program, built on the premise that vertices are handled the original way. At that point, you’d basically have to overhaul and rewrite a lot of Blender; at that point you might as well start a program from scratch. So even if Pixologic published were to publish a paper on how they handle high poly counts, you still wouldn’t be able to easily implement it in Blender. OOP can’t stop all dependency issues.

If the project had gotten started, we would be able to build a foundation with the intent of handling millions of polygons, and we wouldn’t be limited to trying to accommodate for other areas since sculpting, painting, and retopologizing would be the main focus. If you don’t have things like physics, you don’t have to compromise to make them work.

Sorry to see you so deeply hurt by having your daydreams put into perspective. Now you started a new thread. With what intent? Do you really think blaming the community for not being consent with your ( again sorry to that) unrealistic, short thought opinion would somehow help you? What did you expect to happen here?
Now here’s my personal advice ( btw. thank you very much for the revelation of those deep insights in your opening post. If only earlier anybody would have pointed that out to … Oh wait… ):
Do the Elsa! Let it go… It isn’t worth the sweat and tears to continue this uphill battle against our ignorance. I am sure you will find another project where your energy will be put to a good cause and your sleep will be fluffy and refreshing again.
Seriously now, get over that anger and get a little time between your current disappointment and this little incident. Maybe then you will see why you got all those contrary stances.

1 Like

Yeah but the truth is, no one knows how to do it like pixologic, not even autodesk. lol
Nowadays people rely on gpu tech to speed things up, so it will never match zbrush’s performance, not with the current tech. :wink:

1 Like

You’ve been looking for the wrong people. Projects don’t start with programmers, they start with project managers (code is the end product, not the starting point). Programmers aren’t going to sign on without an idea of what needs to be done, and most open source projects fail without good planning and strong guidance.

2 Likes

Who says sculpting mode in Blender can’t transfer geometry data into a more optimized format for sculpting, like an internal import/export between different architectures? Even ZBrush leverages that concept to transfer between 3D and 2.5D.

Sculpting and retopology are obviously hot topics around Blender today. So it would stand to reason optimization will be seriously looked at in the near future, if not already.

And even if optimization is currently not planned, you’d be better off lobbying for THAT instead of a new project. Not only because it has a head start on sculpting tools, but the chance of eliminating additional fittings in your pipeline is a MAJOR plus. When that happens, I’ll gladly sell my beloved ZBrush license.

5 Likes

Okay, first off, I admit I was thinking in an unrealistic, optimistic way, but deriding my failed goals is just rude. I get that you disagree with me, and that’s your right, but contesting my opinions with sarcastic condolences is disrespectful.

Secondly, I wanted this thread to be clean and conclusive. The intent of this was to act as an update, so if there was anyone particularly invested in this idea, they can at least get closure. The other component was to just give my advice on the matter, based on the past few months. I just wanted to be helpful but it looks like I’ve just made people angry. I’m sorry for anything I’ve written that you’ve taken offense to.

I didn’t think about it that way. That’s a really cool way of looking at the situation. I’m still somewhat worried about the possibility that stuff put in place for non-sculpting related things might cause dependency issues if the algorithms to do some of this stuff require alterations to them, but I’m gonna be optimistic about this and trust your opinions.

I wasn’t aware of that; I wish I’d known that sooner. To be fair, I’m not familiar with how these things go, I’m not a professional programmer, I’m an art major with programming experience.

I suppose. It’s still quite a piss off that there are these amazing algorithms that we use in zBrush but we aren’t allowed to know them, and that there would probably be a ton of new, improved variations on them if they were public; having said that, c’est la vie, the world’s falling apart so it doesn’t make much difference anyways.

I probably should have commented on this sooner, but I figured that either someone else would mention it, or you’d realize this yourself. There is no reason to assume there’s anything in blender’s “core” that holds back sculpting. The fact of the matter is that until recently blender didn’t have a full time developer working on its sculpting module. Before the relaunch of the dev fund and the code quest last year there were only 4 full time devs who had to carefully chose what areas of blender to work on, so it’s only natural that areas like texture painting and sculpting would fall behind what the alternatives offer.

However, things are now different. At the time of writing this the dev fund is at 95k a month. The foundation was able to hire a talented programmer like Pablo Dobarro to work on sculpting full time. If he started his branch before the increase in funding, then we would have probably just seen a handful of new features merged into master before pablo moved on to work for some big company.

They say that sculpting in blender is a separate mode because it handles the data differently from how it’s done in edit mode, so blender already uses optimizations that are unique to sculpting. There is no reason to think that the sculpting module can’t be further improved in order to get even better performance. At most we’d probably only see a rewrite of that specific part of blender. There are already plans to rewrite the multi-res modifier in order to fix the long standing issues with it. Maybe this rewrite in particular will bring huge performance improvements.

And finally, I should mention that if sculpting in blender is impossible to improve, then why did pablo choose to work to work with it? He was under no obligation to make the tools he wanted in blender. He probably would have made something new from scratch instead if it really took a complete rewrite of blender in order to get what he wanted done.

3 Likes

Everyone seems to forget sculptris existed.
https://pixologic.com/sculptris/
Its owned by pixologic now, because it was a free alternative to zbrush that was quickly advancing and actually posed a threat to them. They stopped development on it after purchasing it, but they still let you download the last version for free.

Been a long time since I’ve used it, and been a long time since it was updated, so I dont know if its any good nowadays but if you really dont want to use blender, and you dont want to buy zbrush, it is a free alternative.

This makes me very happy. If the limitations Blender currently has on its sculpting are due to lack of development, as opposed to a difficulty in developing a sculpting module on the foundation of what Blender has established for its other modules.

What I was, and still am a little concerned over, that Blender’s foundation is already made. You can add stuff to it and change some things, but you can’t change some of the core things that a bunch of the modules use without having to rework all of them to ensure they aren’t adversely effected by the changes. It would incredibly time-consuming and difficult to adjust Blender if you were to try to change how things like polygons are stored so that you can ‘compress’ a lot of them into individual data structures.

So far as I can tell from your comment, this is less likely and Blender’s current foundation is actually conducive to sculpting? We wouldn’t have tons of functions that need to be rewritten to allow for a massive increase in the poly count?

What are you trying to sculpt? Post an image of some 2D or 3D thing you want to sculpt or have sculpted. People probably would be able to help.

Maybe all you need is to learn the proper way to use the sculpting tools in Blender. YanSculpts seems to be doing just fine.

As is you sound like a troll saying Blender is a free open source sculpting alternative to Zbrush that has been started by people and I could help with, but I’m not because I didn’t start Blender. If, the way Blender handles polys is really the reason for it not working for what you want to do, and you have the coding experience to help with a whole bottom up sculpting program, then you should be able to easily go in the open source Blender code and change it so it handles the polys differently. Millions of people will thank you deeply for you time. It’s easy to blame the tools and hard to admit it’s your own artistic talent that is not good enough.

2 Likes

Based off of what I’ve read yes. The majority of the time, things don’t get improved simply because of funding. As an opensource project there are a lot of outside devs who contribute, and that is what brought us a lot of the features we have now, but those guys don’t maintain the code in the long term, and the paid devs have a lot of other things on their plate so areas often get neglected. However, I’m not a developer, so I can’t say beyond the shadow of a doubt there’s nothing stopping sculpting from being improved. It is just that from what I’ve seen the main obstacle is simply having enough developers around to continuously improve all the different modules. I would need to hear statements from a core developer in order to believe that sculpting is held back by whatever core component in blender.

For example, there was a big bottle neck like that before 2.8. The old dependency graph didn’t support copy on write, so there were features that couldn’t be implemented because it involved different things modifying the same data at the same time. That’s why we didn’t have multiobject editing before 2.8. I think the worst of those issues are over now that we have a new depsgraph that supports it.

They bought it just to stop developing it? I thought it was just one of their early developments. Considering how, so far, they have a fairly positive reputation for free updates and permanent licenses, that kind of anti-competitive conduct seems surprising for them. I could see a company like Adobe doing that, and Autodesk already did that with Maya, but Pixologic is fairly ethical for a software company owning an industry standard. Even if it doesn’t explicitly break any antitrust laws, that’s not exactly something I would expect from a privately owned company.

Anyways, my issue isn’t about the software being free, it’s about the access to the source code. I’ve emailed them and asked about Sculptris and they have told me that they have no intention of releasing the code (even though it’s free, and they aren’t doing anything with it). Actually, that part makes sense, if Pixologic did purchase Sculptris to stop its development, they probably don’t want to allow the possibility of it being developed further, even with zBrush so far ahead of it.

These “scaled up” rants are getting waaayyy toxifying. You guys read the post where Ton was sent that other guy’s email saying he took 1h and some minutes to figure how to do bevel on a cube in Blender? He said Blender is a “carpy” software. Regardless, this is not a proper way to communicate with developers or community.
I don´t see this thread’s poster going over to Houdini forums to call them “they are years behind of what basic Maya could do in modeling”, etc. Some nerve to come up with that attitude here.

I learned to sculpt with Pablo Dobarro, YanSculpts, Zach Reinhardt and other colossal inspiring people in the community in just 4 months. And it took me a while because I had to watch really closely to what they did with the timelapses and brushes. I invested time and money to get their sculpting trainings and was not dissapointed. Flipped normals have added youtube videos to the arsenal to start sculpting pro levels. You don´t see them bashing the “carpy software”.

I guess these people come frustrated because they haven´t experience any better. I know currently there are rising new studios that are piping Blender into their core production and getting awesome 2D+3D results, realistically and also NPR.

Maybe this guy needs to start here?

I didn’t intend on offending anyone or stressing anyone out. If I did that to you or anyone else, I’m extremely sorry.

Also yeah, you’re right. I am pissed off because I look at zBrush and I just want to take it apart and know its intricacies but I can’t. I want the open-source community to have something as powerful and versatile as zBrush. I was ranting a bit, and I realize this is not the place to do that. That was my fault, and I take full blame for it. I am not trying to insult Blender or its community; the fact that they exist is awesome.

Having said all that, I will say that my concerns (until they were assuaged by some of the informative replies) were genuine. I was legitimately concerned that Blender’s sculpting would be limited by having to avoid creating dependency errors from editing the more fundamental parts of Blender. I don’t feel I was wrong to be concerned and stressed about that.

I haven’t read into any specifics on the matter, nor have I ever used Sculptris, but I don’t think it was on the open market. So it’s not a case of Pixologic buying out the shares than burying it. The Sculptris devs were free to choose, and I believe they were even hired by Pixologic.

@DavidRivera, so far this conversation has been very level headed and the OP has been open minded. No sense in adding fuel to a fire that doesn’t exist yet. :slight_smile:

From what I heard they just hired the guy who made it to work for them. He simply sold the rights to sculptris to pixologic when he started working for them.

1 Like