Here is the problem: I would be able to make some studios set their workflow Blender-based, but I can’t because its limitations. So they will not use it, they will not support it, they will not be involved. Because - as an example - Max was able to handle more objects like 10 years ago in the viewport than Blender does today. And they hate AD, they are not happy with Max at all.
To be frank, Autodesk gave Blender a major opening by creating a near-monopoly in the 3D space and dramatically slowing down the rate of innovation (which it has benefited greatly from). If Maya, Max, and XSI were still being developed by their original companies, Blender wouldn’t even be close to where it is today in terms of professional and studio use.
You can also look at the highly competent development of Zbrush and Houdini, those apps. get new tools that are decently polished and feature-laden the first time around and see a lot of old tools enhanced with new releases as well (Houdini is also being developed at a faster rate than Blender at this point, meaning there’s a risk of a lot of people jumping ship if things like modeling become as straightforward as in a general DCC app. That being unless Blender continues to gain steam in development).
It’s more or less the same with Krita benefiting greatly from the Adobe monopoly. Godot is also reaping dividends in a land where the leading solutions have been known for making controversial changes in new versions, releasing new versions with lots of bugs to be fixed in service packs, and taking years to implement requested features. Then there is Firefox which became the world-leading browser amidst a relative lack of development from companies like Microsoft, but then fell like a stone when Google made a serious effort with Chrome (with a rapid development and innovation rate).
LMMS continues to struggle to find relevance in part because the area of music-making software is still quite healthy as far as competition goes (apps. like Mixcraft continue to innovate there with huge releases and decent stability, not to mention the benevolent and affordable licensing).
In short, the areas where FOSS is thriving the most are in the markets dominated by high-priced software with strict EULA’s, comprehensive DRM, and subpar development and bugfixing work.
One idea that people told was make a “pro” blender user to have better channels to talk with developers. Actually I’m tired to see people that don’t know model nothing except a box talking with developers about what it’s important because they have a lot of free time to see hundreds of emails, speak in IRC all day,…
Completely agree. That is the waste of their time. And here another thing comes: as an example I work on a huge project for 2 years, even for Max terms. It will be a commercial one, but I cannot share any project parts or details because developers are not contracted with NDA to handle it confidently (It comes from the OS I think). For the OpenCL error I reported in the ancient times and still unsolved I cannot support my findings (and the importance) with project details (which would show why it is important from the interest of BF, too).
I have reported few bugs of a project with NDA and allways find a option to send a few part of a mesh with the error.
I asked about it, they told they cannot guarantee anything. I had to spend lots of time to remake something similar.
Also in may times I’m not even sure they properly investigated the problem (and it is not their fault, there are too many hardware/software variations to reproduce an error - although related to the OpenCL error it was not the case).
But as an example, what clearly drove me crazy because of its usefulness: changing the display of mesh/material names in the 2.77 (if I remember correctly). It became completely useless just because someone never worked on any large scale project where naming conventions were important. There were no limits to implement it in the way I asked for, but the answer was NO… Sad. BTW it happened to may times for features I asked for (like choosing the ‘renderer’ GPUs and handle them separately from the display device, etc.). It seems that a critical mess is needed to convince developers from the obvious implementations sometimes (no offense, I’m just not happy).
If I remember back to the times when Unity3D started, they implemented requested small features like in 2-3 weeks.
There should be an improved bridge between developers and power users.
Blender has come a long way in the last few years and it’s getting more and more attention. At last year’s SIGGRAPH, quite are few technical paper presentations came with “…and we implemented this in Blender/Cycles”. At the MaterialX BOF session, they also showed Cycles renders. When I was job hunting around end of 2016/beginning of 2017, my contributions to the Cycles codebase were what got the most attention in my resume.
A couple of years ago, mentioning Blender just got you pitiful looks. These days, I see it often mentioned in line with Maya or 3ds (including job descriptions) and it’s taken for granted in hardware benchmarks or things like AMD product presentations.
It is very obvious that you have never worked with professional developers directly. I don’t have the time to explain you why you are wrong. If you had worked with developers just once, you would know how difficult even the communication is to be sure that everyone is talking about the same to begin with.
If you have the resources, the source code is that bridge. In contrast to other, closed source applications, you can always implement missing features yourself or hire someone to do it for you. I understand that this is not an option for most individuals or small teams, but since the starting point of this thread was big studios, they certainly can fix problems on the spot and quicker than they would if they had to beg a software vendor to do it for them.
I worked with several developers in the last 20 years and I know that communication is hard ands starts with finding the ‘common language’. So please be careful what you say about others. But it clearly has nothing to do with the fact that effectiveness is a key for spending time/money better.
Yep, on small scale I pay for small features. But there are areas what cannot be resolved without the help of BF, also ones what are too big to finance them alone (it’s cheaper to pay a license for a commercial tool what have the feature set already).
I agree with Autodesk - even Carl Bass mentioned that teams love to run after new projects yet not really focus on finishing the job which is less attractive. Alias Fusion ReMake I crash their apps on such a high basis … AD I think lacks good leadership to work on such problems and help guide teams and probably is simply too big.
However I am shocked to hear the same about C4D - that is pretty sad.
That’s why blender foundation does these open movies to find the bugs and workflow problems.
Then I have no idea how they were not able to find and solve the CUDA limit (and tons of other things). The most cost-effective way is not to find these problems alone with own projects, but making a tighter collaboration with studios and fulfill their request in real production environment IMHO.
No offense again; I really appreciate the work BF does, but this is my opinion; there is no need for reinvent anything, just listen to the users work in these environments/pipelines.
If you have worked several times with developers and you did not have the same issues every professional software company is encountering, you can sell that knowledge for millions or even billions of dollars!
Well, the industry won’t suddenly change to Blender while their workflow isn’t fit for blender to do the job.
But, with Blender Masters spreading out all over the world into different companies, I can see more and more companies implementing Blender into their workflows.
It’s only up to Blender Foundation to become a sustainable company with its own service department for those companies. Until this business 2 business channel is open, Blender won’t take over the large studios. The liability is simply too high.
About bugs, I think Blender Foundation don’t have enough people or simply can’t afford to maintain and repair all the features that are either limited, outdated, or bugged.
I believe that a lot of early design choices of Blender are like deep rooted diseases at this point. If BF would like to cure those diseases, it’s probably going to be a very far step. 2.8 seems to have a lot of new features and exciting stuff, but I guess the old parts are all going to remain exactly the same.
Autodesk, work on so much things at the same time, it is a big entity… when they want to push 3Dprinting, they open 3 labs + 2 open research and developpement fabrics and in 3 months they have come with some of the best and innovative 3D printing machine never made (industry wise ) . guys who have participate on it have surely have the best of their professional life during this period ( invent, create, rush, conclude ) … At the same times they have developps a tons of tools and show projec for the industry for transform thoses research for port it to the industry… and now you can can design airplanes part using this technology ( organic ) .
I could - theoretically - but I can’t. As is you worked with developers and companies you know that they always want to play ‘safe’; as the company grows as safe as it is possible, with saving themselves from any responsibility in all level. Also the size of the company determines how easily you could break through the walls made by the average quality of human resources and tradition.
A number of crufty systems plagued by those early design decisions are actually going away and/or being re-done for 2.8 (Viewport, depsgraph, the layer workflow, and BI). The Particle system is also going to be replaced with something modern, but that was delayed due to Lukas Tonne’s departure (which is one example of the BF not being able to address everything Blender needs due to resource limits). There has also been talk on completely redoing the way Blender handles simulations (in part due to what is needed to properly implement Scorpion’s fracture modifier, but I haven’t seen any recent news as to where they are at).
The last thing is that the UPBGE implementation, if it actually goes through as planned, will act as the complete overhaul of the game engine part.
There’s also a core rewrite planned when it’s time for Blender to hit 3.0, but that is likely many years away yet.
Make a movie is not make a game, they made a few models with a clear and easy style, without any pipeline for games… is like try to know how to be a cheff in a restaurante making sandwichs in your house. If they want to know problems of workflow they only need to give a users a channel to talk with them. Because meanwhile they have only their pov we have a lot of pov from differents projects, perspectives, styles, pipelines,…
all companies make this, you want real feedback? You talk with your real users, you pick them, they made models, see problems,…
the problem actually is that blender is for Artists, but freelance artist are the people with less power to improve blender, they dont know to code, they dont know production,… but they are the people that use blender and speak of blender to other users.