Blender taking off?

@xrg - its fair point, but there has been low-end 3d apps around for a while, I recall GMax was some cut down 3DS-Max to be used with games… (free even) truespace also made a push in this direction (would often be included on magazine CD’s). both defunct now.
I don’t mean to say all low end 3d-apps are doomed, just that its not a new thing.

What I read into when you said ‘corner the market’, that suggested to me that Autodesk will be able to successfully erase all of the gains that Blender has made so far in that area (12 years of progress) because the majority will move back into their ecosystem and cease to touch anything made by the BF. The term ‘cornering’ really is when your company has such thorough control over a certain market segment that there is little room for competitors to try to eat their way into a larger share (they will either be outperformed on a consistent basis or crushed by monopolistic tactics and deal-making on the side of the larger company).

You’re reading too into what I said I guess. I don’t think Blender is going to get wiped out or anything, it’ll just be like it always is: too apathetic to compete for marketshare thus we cater to a niche. I don’t see any reason we won’t continue to grow though.

Here is the Senior Artist for Epic Games mentioning that he’s tried Blender but gave up pretty quickly every time. We have always (childishly) responded to those artists with the “Blender is for Blender Users” shtick. Since we never address the concerns of these artists, we ensure that those artists continue to never be Blender Users.

Well, artists in Small Studios aren’t necessarily this niche “Blender User” anymore than the artists working at Big Studios are. A lot of those artists are going to have identical issues regarding Blender. So, with Autodesk and friends jumping in with product lines targeting the Small Studios, I don’t see any reason that they won’t have the same broad appeal to artists in Small Studios that they have to artists in Big Studios.

Read the entire thread, and it does make a few things clear that Ton really needs to finally be serious about (if he wants his dream of many professionals using Blender fully realized).

In fact, Ton may really be in need to start reading that site, there’s a lot of people there who currently couldn’t care less about anything he produces because of interpretations of past comments saying things like ‘We make our own standards’ or ‘They’re not our target audience’. Now here’s a list of ideas that could help get Blender to where many on that forum will choose to run towards Blender and subsequently allow Ton’s wildest dreams to become a reality.

  • Upgrade the different control modes (Maya ect…) to include a full automatic change of preferences (so left-click select becomes default and the like while the Blender mode keeps right-click ect…).
  • (Possibly needed), start moving away from the GPL, more and more people and projects are starting to shun the license and many think it’s more about software ideology than keeping an application free and open. Replace this with a custom Blender license if needed, or add numerous exceptions to the GPL like for interfacing plugins with commercial software.
  • Improve communication links between users and developers, this already started to happen with the new developer website, but we could use further improvement through things like a special section here on BA where only those proven to make constructive comments on features can post.
  • Devote more time to patch review, this would encourage more people to contribute code more frequently and may inevitably lead to many more contributors with commit rights straight to Master
  • Make it a goal to include at least a few notable interface and general usability improvements with each new release.

Otherwise, you might have Blender being in a perpetual state where people look at it and say “It’s not there yet” no matter how much development is done. Now I’m sure no one high up in the BF would want to see a Blender that continually grasps for a larger audience only to fail year after year. Implement something along the lines of these guidelines and Blender may take off so fast it will make your head spin.

On the flip side this thread here about Blender shows the most recent pages of comments being a lot less critical of Blender on average than the starting pages. That would probably not invalidate the list of points I brought up though because not addressing them could stunt Blender’s increased adoption in the industry a lot more than how things could be.

Now that is what I call a post taking off. I expected three replies or so, but it has become definitely more. Thank you for your answers.

This cannot be emphasised enough, in my opinion. One of the worst things about introducing Blender to others in the game development industry is the attitude one gets exposed to if you dare start mentioning problems with or possible improvements to Blender.

It is one thing to have fanboys slam you for your opinion (you’re going to get that everywhere), but the “Blender is for Blender users” line from the lead developer? Yeah, that’s the line that keeps coming back like a bad penny. Sadly, it’s a line that would easily fade away if it didn’t strike a chord every couple of months. :frowning:

@BTolputt - re: Blender is for Blender users - I think the purpose of this statement, is to highlight that we are not working to make a clone of some other 3D application.

The problem we run into is some expert in other software talks to developers and tells us how Blender should work.
Without taking the time to learn how Blender already works, they most likely want some free-clone of their existing application.

Said differently, if someone wants Blender to be a clone of <some-other-3d-tool> then they can expect to be disappointed and we’re OK with that.

However these users may be right at times too, its just they typically say things we already have heard many times… from existing users, its not like we’re close to running out of TODO’s :slight_smile:

Also I get a bit wary of experienced users of other apps, who aren’t really that invested in learning a new tool.

  • I tried GoLang, a while back, it was pretty nice but I prefer C for now, does that mean Go’s a bad language? Of course not, I just don’t have any good reason to switch at the moment.

It’s not usually about cloning as much as following conventions artists are used to. MODO and Maya aren’t clones either, but they do share similar conventions that artists are used to.

To put it in a programmers way: it’s like PEP8 style guides in Python. I don’t know C, but I assume on that side there is also some set of standards for formatting as well. Why do coders follow conventions like these? To try to make it easier for developers to read, debug, and maintain the code. Patches that don’t follow conventions that Blender devs are used to I imagine tend to not get accepted for this very reason.

Like coders, artists also have conventions they’re used to. Since we break them, an artist with 10 years experience has to start at basic how-to videos on selection and navigation. If we followed conventions more closely, they’d be able to get up and running quickly just like Python devs or whatever.

@xrg, pep8 is just a code-style (being self-consistent), but I see your point that Blender isn’t following many conventions.

Nevertheless there are highly successful projects which don’t follow industry-standard-conventions… blah blah across many areas in technology (Check Erlang/Vim/Git/tiling-window-managers/nosql :slight_smile: )… standard != {ergonomic or efficient}. This is just my opinion, I don’t expect others to agree.

The ship is steering dangerously close to left-right-mouse button territory… how about we avoid that one for now? :slight_smile:

To my understanding Zbrush too broke conventions yet it was accepted as an industrywide standard.

Though it is not essential for established artists to learn Blender so they give up, it`s not revolutionary like Zbrush was. I guess some key mapping presets and further UI usability improvement could remedy that to an extent.

I’ll never understand why it has to be left or right. How about a compromise and pick the middle mouse button, such that everyone is (un)happy…

Reading this, I’m thinking about this exchange of opinions at Ton’s reddit AMA some months ago :

*After this it gets into RMS vs LMS - and I agree with ideasman42, lets leave this one out as long as possible. :slight_smile:

Not sure I agree with your analogy.

Yes, but their own version of formatting. Their own styles, guidelines and ways to do things. There are equivalences and similarities with other languages, but style guides are mostly for legibility within a language - not cross-language legibility.
Likewise, Blender has similarities and equivalences with other DCC softwares, but it is in no way “wrong” for not having the exact same shortcuts/control mechanisms etc or having it’s own unique take on certain things.

Why do coders follow conventions like these? To try to make it easier for developers to read, debug, and maintain the code.

Right. Developers using the same language and probably working on the same project. Not for random developers using some arbitrary other language.

Like coders, artists also have conventions they’re used to. Since we break them, an artist with 10 years experience has to start at basic how-to videos on selection and navigation. If we followed conventions more closely, they’d be able to get up and running quickly just like Python devs or whatever.

I’ve used a lot of DCC’s and have not run across these conventions you guys are talking about. So the assertion that we “break them” rests upon this premise that they exist, which seems to be at best subjective. To me blender is as much the same and as different from 3DS/Maya/Modo/Houdini/Lightwave/Softimage/etc as these are from each other. Trust me 3DSMax artists are just as lost when transitioning to Maya.

I have over 12y industry experience and I’m currently starting to learn Houdini. I’m lost. It’s weird. The interface seems odd. I have to start learning at the bottom with beginner how-to tutorials. But that’s OK. It’s to be expected and I’ve done it before with other software.
My experience still counts a lot, but you have to differentiate between practical and abstract knowledge. In Houdini i have my 10+ years of abstract knowledge. I understand modeling and topology. I understand particles and simulations. I understand animation and f-curves, etc. But i have no practical knowledge of the software and that is to be expected.

Like in programming. If you know Python, you already have an abstract understanding of objects, vars, classes, etc. That might not help you write a class in C, because that is practical knowledge. But it’s very valuable knowledge because you at least know what a class is and where and why you’d want to write one. After that, learning how in a specific language is just a detail.:slight_smile:

I don’t really want to debate how well individual programs do and don’t follow conventions. My point is just that the complaint from those artists isn’t necessarily to make Blender into a clone.

Professional artists use all sorts of software all the time, and hardly any of them are clones of each other. If you’re an experienced user and pick up Mudbox for the first time you can be up and painting in less than an hour. You try to do the same with Blender, you have to watch several hours of video training, and probably will need a week or two before you’re comfortable enough to paint with it. That’s what I was getting to.

Imagine if you made tax software, and a significant chunk of tax preparers found your tax software too difficult to use. Wouldn’t that be an indication that maybe there is a fault in the design somewhere that should be addressed?

Blender is designed for 3D Artists but a significant chunk of professional artists don’t use it, not because it isn’t capable, but because they simply find it too much of a hassle to figure out. Again professional artists; not 3D newbies. Rather than identifying this as a design issue to even potentially fix, we brush them all off as just not knowing what they’re talking about.

@Muffy - I just mean if you have a dev with 10 years Python experience, they can probably get up and running with Blender’s Python code pretty quickly because we’re not doing anything wildly out of line with what Python developers are already used to. That isn’t the case for artists.

I don’t mean to nitpick, but you’re now comparing Python to Python. So its reasonable to assume that standards and styles are consistent if you are comparing one thing to itself. You could follow that Blender is perfectly fine because Blender is consistent with Blender.
Before you were comparing Python to C, assuming there was a style standard between the two, and concluding that this should be the same for Blender and other DCC applications. There was also the assumption that a style standard between DCC applications exists. I was just pointing out that this isn’t at all the case.

Autodesk has made some efforts to unify UI in the softwares they purchased, but this has in practice resulted in little more than gadgets like the viewcube, which mostly everyone turns off.

Blender is designed for 3D Artists but a significant chunk of professional artists don’t use it, not because it isn’t capable, but because they simply find it too much of a hassle to figure out.

You also have to factor in learning motivation. Artists put in more effort transitioning from one DCC to another usually because they are looking for a job at a studio that’s uses X software over Y software, or they have a new job and are obliged to learn the new pipeline. Id say this is by far the number one reason experienced artists learn new software. I don’t know a lot of artists who change their personal art pipeline for the sake of it.
For blender, the professional job market is very small compared to Max/Maya/etc. So the motivation to learn it is low, which gives the perception that learning it is much harder than it actually is.

Take the example of the lead artist from Epic. Unreal has a long history with 3DSMax. They have money so the licences are not a problem. Their art pipeline with Autodesk products is up and functional and production tested. They have a good business and development relationship with Autodesk. That artist has made a great career using the pipeline he is currently using and used to. He has zero motivation to learn blender other than curiosity or a manager who asked him to test the Blender-Unreal pipeline.

Here is my take on conventions and originality -

(I broadened this section, so if you read it already, it now contains some new argumentation, and a link to an insightful talk by Matt Ridley)

Luckily smaller studios are more flexible when it comes to switching or implementing tools in their current pipeline. So the market is more “in flux”. I think, as long as blender is able to offer a plus in workflow or feature wise, things will move on quite happily. Especially if it really learns to “talk” correctly with other tools (export/import) and is able to compensate for some restrictions other software have.
So imho MayaLT or Substance Painter is actually an opportunity than a thread.
And yes my pov might be too fanboyish and cross-application thinking.


Sure but one reason we brush them of is, that blenders workflow is quite coherent once you get it. And its really hard to communicate that, especially with forum posts. Also you rarely get to the point where you could explain to a user why a certain design decission was actually made.

To get users of different software using blender it actually needs two things:

  • Hastle free import/export (so using a feature is less time intensive than using a workaround in its existing software)
  • Easy to communicate features and workflow implementations wich actually rocks other software implmentations-

By doing so you would lure other-soft-users to learn blenders interface in a task oriented way. Wich is imho the best way to “get it”

About standard/nonstandard issue.

Now that Blender has layers in texture painting mode, this struck me more than usual : in every image applications (or at least most of the one i ever used) , layers come in that order :

  • bottom layer is the base layer
  • above it are all the various layers with various blending modes

But Blender has layers using the same order as in the texture panel , meaning that :

  • top layer is the base layer
  • below it are the various layers with various blending modes.

Basically the ordering of the layers is inverted in texture paint mode in comparison to image editing/painting applications.

I mean, by example in Gimp , look at the order of layers used :

If in Blender i replicate this order in the layers, i don’t obtain the correct result :

To obtain the correct result i then need to invert this order of layer :

I have no idea and don’t care much about how other 3D applications are doing with their own painting layers, but it strikes me odd as for texture painting Blender is doing it the inverted way as how you do when painting textures in image/texture editing applications ?