Blender roadmap – 2.7, 2.8 and beyond

Two years ago, we scheduled the 2.6x period to have a “branch migration” focus. In the past years, we”ve seen the results of this work, with massive amounts of new tools and new editors in Blender.
All of 2.6x builds were meant to be (mostly) compatible with 2.5x though.

Since we”re running out of 2.6x numbers it”s really time to think of focus and projects for coming years. Below is a proposal I”d like to see discussed and reviewed here and on our mailing list(s).

The last 2.6x releases

  • For 2.68 and 2.69 we strictly keep compatibility and keep focusing on stability for Blender.
  • Anything potentially unstable or breaking compatibility should go to a 2.7 branch
  • If needed, we can do a couple of 2.69 updates (a b c d) to merge in bug fixes only.

Blender 2.7x projects

For 2.7x projects we will allow forward and (minor) backward compatibility breakage. That means that by default, the 2.7x .blend files don”t have to read reliably in 2.6x or older. Backward compatibility stays crucial though, and should only be acceptable for big and important improvements. Changes can also be including a revision of UI layouts, naming of options, themes, and shortcut defaults.

However, for as long as we add breakage in Blender 2.7x versions, we could also try to keep the last 2.69 release updated with essential fixes.

This is also the perfect period to move to git for Blender coding.

A summary of projects or targets that fit this period:

  • Move to OpenGL 2.1 minimal (means: UI/tools can be designed needing it, like offscreen drawing)
  • Depsgraph refactor, including threaded updates
  • Fix our duplicator system, animation proxy (for local parts of linked/referenced data)
  • Redesign 3D viewport drawing (full cleanup of space_view3d module)
  • Work on cpu-based selection code for viewport
  • Sequencer rewrite
  • Asset manager, better UI and tools for handling linkage
  • Python “Custom Editor” api (including better Python support for event handlers, notifiers).
  • UI: refresh our default

Blender 2.8x projects

I propose to not wait with 2.8x versions as long as we did for 2.6x and 2.7x. It”s not needed to release every digit. At some moment in the near future, even while we still work on a first 2.70, projects for 2.8x could already get started.

For 2.8x we can target projects for bigger rewrites, with a lot of compatibility loss. Examples could be:

  • New “unified physics” systems, using much more of Bullet, unification of point caches (Alembic).
  • Particle nodes (could co-exist for a while with old particles though)
  • Nodification of more parts of Blender (modifiers, constraints)
  • Game engine… (see below)
  • OpenGL 3.0?

Blender 3.0 projects

For all 2.x projects we will stick to the existing C core, Blender files and data structures (DNA) and Blender”s scene/object/data methods as much as possible. It has its limits though – it”s a design from mid 90ies that survived very well but never was predicted to work 20 years already.

During the next few years we can collect in our wiki the issues we have, and the wishlists and design ideas for tackling this topic.

Blender Game Engine

With work being done on threaded drawing and updates, viewport (compositing) effects, unified physics, node based animation, and everything that”s currently real-time in Blender already, I also propose to refocus the current game engine to re-use much more of this work.

Or more radically worded: I propose to make the GE to become a real part of Blender code – to make it not separated anymore. This would make it more supported, more stable and (I”m sure) much more fun to work on as well.

Instead of calling it the “GE” we would just put Blender in “Interaction mode”. Topics to think of:

  • Integrate the concept of “Logic” in the animation system itself. Rule or behavior based animation is a great step forward for animation as well (like massive anims, or for extras).
  • Support of all Blender physics.
  • Optimizing speed for interactive playback will then also benefit regular 3d editing (and vice versa)
  • Singular Python API for logic scripting
  • Ensure good I/O integration with external game engines, similar to render engines.

What should then be dropped is the idea to make Blender have an embedded “true” game engine. We should acknowledge that we never managed to make something with the portability and quality of Unreal or Crysis… or even Unity3D. And Blender”s GPL license is not helping here much either.

On the positive side – I think that the main cool feature of our GE is that it was integrated with a 3D tool, to allow people to make 3D interaction for walkthroughs, for scientific sims, or game prototypes. If we bring back this (original) design focus for a GE, I think we still get something unique and cool, with seamless integration of realtime and “offline” 3D.

Ton Roosendaal
June 2013

251 comments 126,685 Views
  1. What do I think? I think this is rather cool 😀

    • I really love the way and speed that we can develop a simple concept into a game in front our eyes with Blender… I use Blender for about 8 years now, and the most interesting part for me now is the BGE, every time that we need something done in the university in where I work (for no comercial) I usually use it, but I hate that for personal proyects like “Make and Android game” or simply “Make a PC game” I have to choose other GE like Unity when I feel very confortable with BGE…

      On the other hand my partners always feel insecure about the GPL licence and this is the real reason for change all… I really dream that something like the GameKit proyect (MIT License) :

      Was implemented in the core of my dear Blender. Some time ago I read that the GPL licence of the BGE can’t be change to (for example) MIT because the hard work of contact every single developer involved.

      Like a lot of persons I’m waiting for other BGE project direct from the Blender Fundation, right now I don’t know if one is in your list.

      The BGE like for many others users out there is my day and my night… the thing is that if you make a “render” or a “movie” is yours to make what you want without problems or having to do tricks, but with the BGE right now If you don’t use the trick to avoid GPL license well you know the terms, and that’s right when you are trying to make an open game or something but whe try to explain that to a client well…

      Right now the tech future is pointing to mobile hardware, more and more devs are trying to understand a way they can acces to that world. I was very exited about the Android integration of Blender but I don’t hear news about it in the forum or other places like the “Developer Meeting Notes”.

      Out there exist a lot of people trying to understand the BGE and the Blender Fundation can share its knowledge in the area of making games in the same way that they share the DVD’s trainings in the store.

      Well right now I’m waiting for the Mike Pan book (20 June) about the issue.

      Greetings, sorry for the bad English and size.

      • This 100%. The issue with the BGE isn’t the technology, it’s the licence. Like it or not, we live in a commercial world where proprietary licencing is necessary for most people to make money. Having your product constrained by a GPL licence isn’t a realistic possibility for most game studios.

        If Blender “Interaction Mode” can interact with proprietary, encrypted game files distributed over steam or something similar then well and good. If not, it’s probably never going to take off as a mainstream game engine.

        • First of all, I’d like to thank all the people who work on Blender. I’ve very much enjoyed using the product you have produced.

          I’ve tried the BGE and in a way I will be sorry to see it go. Unfortunately, though, it’s not viable for most projects because of the licensing. I know there have been discussions in the past about changing the licence, and people always seem to have concluded that it isn’t practical.

          Various people have suggested ways that the GPL issues can be avoided, but it’s really not enough. It’s not just about satisfying me, it’s about satisfying my potential business partners. I wouldn’t want to spend a lot of time and money developing a game, only to find that partners like Steam are scared to handle it because of the GPL issue.

          If it’s not viable to change the licensing then I think the BGE should be dropped. Otherwise the risk is that developers spend time on an engine that very few people will use, and game developers spend time developing games that no one will sell.

          People who are interested in games and Blender might instead spend some time developing GameKit ( and ). This project doesn’t have the licensing problems, and provides a game engine that is closely integrated into Blender.

          Both BGE and GameKit offer something that other game engines don’t: tight integration with the 3D design software. This is a very good thing, as moving models between the design software and the game engine is always a big waste of time.

      • I’m a game developer in USA (but I’m from Brazil) for 12 years now, in fact I eat thanks to games 🙂

        The GPL issue is not big deal to make a game work for you, but certainly as many Game Devs we have always a client (if we are not making in house devs), the thing is that like many clients out there, they will always pay the bills 🙂 , if our client don’t want eat the GPL licence because its terms and conditions, they simply choose other dev studio (to avoid complications), that’s the great truth… the clients always want full control over the output from the service that they paid for and that licence just works to open source minds.

        I have used BGE and surely the speed that we have to develop game concepts is amazing! even our artists can do this task with it and that’s is just… amazing!! if our artist can put to work their logic with the BGE and have the fun of their lives (they eyes shine like a children) that give me an idea of the importance that the BGE has to the persons that never develop something (like a far dream come true jejeje).

        I never hear about the GameKit before but if it has MIT licence and works closely with Blender (even has the Ogre render engine) … why this engine don’t substitute the actual task of the BGE? (you will avoid a lot of complications)

        -Ayrton Tebison-

        • When you say you’ve used the BGE to quickly output game concepts I’d believe you, but have you used it for any actual products that shipped to a consumer?

  2. I think integrating the GE is a very cool idea, as well as using the term “Interaction” instead of Game. Because Unity3D is the primary choice for many game makers, I see other applications more fitting for Blender.

    – creating animations for tech. manufacturers where one can interact with the product, such as light sensor, to demonstrate its usage;

    – virtual reality for architects and landscapes with interactivity like revolving doors, elevators, lighting, traffic lights;

    – creating a sim for ESA 😉 with all buttons and screens in the shuttle and provide training for future astronauts;

    – creating a sim for medical organizations and schools and have interaction with the operation room tools for training.

    • Unity Free (the next best equivalent engine) is crippled for free use and is incredibly buggy.

      If you want most things BGE users take for granted you need a lot of money as none of it is for free.

      Most other free standalone engines are great for seasoned devs but for the beginner there is nothing like Blender with logic bricks and Python.

      People often mention CryEngine or Unreal, but if you want to make a game that is not a FPS you are in for a long slog.

      • And if you add to this the fact that those development kits (which are the way to make game development “easy” for a non-dev) aren’t linux compatible, i’m afraid that we’ll lose something important. It would be cool to have again the stand-alone possibility.

        A new blender player, without the licence problems?

      • There are other options regarding open-source permissively-licensed game SDKs. OGRE3D comes to mind.

  3. I really like the renewed focus for the GE. Make it suitable for rapid prototyping and game compatibility testing, and you have yourselves a winner. Making it a full blown game engine is just too much of a stretch.

    It’s for the same reason that you don’t see any CMS projects with an equally successful forum/wiki component for instance, because even though they’re closely related, a separate, more laser focused tool will always get the job better.

    Here’s another idea for the GE: Include a few stock “game templates” with pre-made game cameras and controls for common genres like first person shooter, top-down real time strategy and third person MMO. No need for any assets (though I’m sure the Blender community could quickly fill in that end), just a solid camera and movement system. That way I can do quick preliminary tests with my animated model as I keep working on it, without having to deal with a constant export/import.

    I also hope the “Interaction mode” will be equivalent to “Anything you make inside the boundaries of this mode can be safely imported into every popular game engine”.

  4. It’d be awesome if blender GE become asynchronous; as currently python scripts will pause other scripts and game logic till its finished.

  5. Seeing a roadmap is really nice;however I cannot believe there’s no subjects for existing limits of blender.Why don’t we see the following problems fixed first? * there’s no smoothing groups or vertex normal editing , edge split is just not enough:/. * how about a projection cage for texture baking, multires baking is good but doesn’t solve issues for multi-part complex models. * how about hair collision support current work arounds don’t solve anything and not useful in production. * how about a particle mesher I’m aware some people worked on it in the past and some people are still working on it any plans?. I wish these issues were resolved first.

    • I think this post is intended to address the bigger picture, rather than specific features.

    • That’s mostly part of (possible) regular ongoing development. But for particles and hair physics we need a massive project (which got listed).

    • There is an edge split modifier, which, as far as I know, does the same thing as smoothing groups… or does it? Are there advantages of smoothing groups, compared to the modifier?

      • The edge split modifier just separates edges (optionally based on angle). Currently, it’s not possible (to my knowledge) to have one mesh that is partially subsurfed.

        • you can use edge crease to do the same thing as smooth groups, edge crease is actually more flexible. select the edges where you want sharp edges on a subdivided model and press CTRL-E and select edge crease. if you want a perfectly sharp edge set the edge crease to 1. to remove it set the edge crease to -1.

  6. Am I understanding correctly that the proposal is to release 2.7x and 2.8x at roughly the same time? Sort of like a ‘mostly compatible’ and ‘most incompatible’ branches?

    Will the software updates alter the minimum hardware specifications? My PC is at the current minimum specification and an advanced warning of this would help with purchasing a new system; I’m in need of a PC upgrade anyway.

    Looking forward to Blender’s future. 🙂

    • No – as soon as 2.8 projects are release ready, it becomes ‘the’ release too. Such 2.8 work stays in branches meanwhile.

  7. I m really happy with the projects that Blender Foundation is having, I m pretty sure it will improve Blender. I hope that this big migration will not confuse artists.

  8. Those plans for the BGE are absolutely the right direction IMO. It’s correct to admit BGE will never be a major “game engine”, and that in many ways it doesn’t need to be. There are other open source engines dedicated to that stuff, so it’s kind of a waste of effort to try and make BGE the same.

    An in-built “interaction mode” for easy creation of simulations and prototypes on the other hand is a very useful, and unique, concept. So that’s a great focus for the future of BGE.

    • Thing is, the BGE could become a major game engine (as in, used by lots of games). Just look at how Unity has spread, and it has nowhere near the features of top tier game engines.

      If the Blender Foundation offered a free and open source alternative which people could use to create proprietary contend (just like we do with models and animations) I could see it overtaking Unity in a heartbeat.

      • That would take a lot of effort though, and doesn’t play to the strengths of Blender or the Blender Foundation. As I say, there are many other FOSS game engines about, so it seems to me that it would be better to focus on those.

      • Also, see Ton’s responses to similar posts below. I think he makes a lot of sense.

  9. Hi,

    I am a teacher in an arts school and I teach 3D animation in Blender and creative programming. In one of the subjects, since other colleagues taught Processing and Pure Data, I tried to teach Blender as a tool for creative programming which proved a bit tricky, particularly regarding libraries available between different Python versions – the Blender upgrade to Python 3 and the slow adoption of Python version 3 broke a lot of possibilities in this regard.

    I still managed to get Blender to receive information from an Arduino board, OpenCV and Kinect to object instancing and Audaspace, but none of the students chose to make the final project in Blender due to the configuration complexity (they preferred Pure Data which still seems strange to me because i think is harder).

    Therefore, a repositioning of the BGE as 3D interactive digital arts to be extremely interesting for me and for my students.

    I do not like to mention other projects as an example because I think that every project should be different, but something like Cinder, maybe optimized for 3D would be great, so not newbie friendly as Processing, but accessible enough for an artist with some programming skills.

    Perhaps could be a discussion session about this topic in the next Blender Conference where the community can express their opinion.

  10. I think a rename from GE is good, it could adapt a VVVV philosophy or other realtime graphics that’s done mostly with node logics.

    But the naming, interaction mode.. I would consider “real time”.

  11. Thank you so much for all of the work that everyone has put into Blender. I’m currently finishing up my BFA in computer animation and I was wondering if there were plans for muscle systems or rigging influence objects in blender? I understand that one can use blend shapes with drivers to achieve a similar result but it would be wonderful if there were a muscle system that could show realistic mesh deformation.

    Thank you again for creating such a fantastic program!

  12. I think its true that we cant compete with Cryengine and Unreal. .but we can make blender game engine an engine that a small team can create games for mobiles,tablets,consoles,pc’s ! So i think its better to upgrade it so that indie game devs can use it !

    • This!

    • I think you forget how Blender ended up on the map at all. People don’t have money to “experiment and learn” (especially in today’s “times). Blender offered 3D Rendering for free and a unique UI. Blender Game Engine has helped drive desire to use Blender (as least it did for me, otherwise I would’ve stuck with my Paid license for Newtek’s Lightwave).

      Blender is extremely modular, which means, if written / programmed correctly, the GE can be modular too (along with everything within it, Bullet, Particle systems, Rendering systems (shaders), etc).

      Unity Free is a joke and there is no Linux option. Crytek..ha, who has 30k to spend on soemthing that might not even pay off (you are looking at 100k – 200k for a “standard” license).

      I believe Blender trying to “wipe itself” of BGE is a mistake and goes against everything Blender stood for and how it got here. The BGE is outdated, so lets in-date it. The license is bad, so lets find a way to fix it. Instead of shooting for the easiest approach, why not try to achieve the best approach. I believe Blender selling itself “short” on BGE will have a lasting affect on the quality perception of Blender in 3D Rendering. It sends the message of “if we can’t figure it out or if we don’t want to, we will aim for the easiest solution”. IN all honesty, is that what is happening with BGE?

  13. Sounds very exciting. I especially like the nodification of more parts of Blender (modifiers, constraints). I hope this also means that modifiers can be instanced data blocks.

  14. I dream about AutoCad features in future Blender for very precise modeling and drafting architectural. And also Luxrender renderer features in future Blender like caustics, editing and control of lighting or lighting groups real time when rendering a scene.

  15. this sounds scary,

    I love the BGE….

    • don’t be scared 😛 I think many would love to both make games. but do interactive art installations with blender I think can boom out of this!

      and finally merge bge with blender will get more c++ into blender i hope 😛 and maybe more bugs fixed. (I recall much in bge being c++, and almost everything in blender source is C)

      • It would also be nice if we could interact directly with the C++ GE core rather than using python scripts.

  16. I suppose if this roadmap comes to pass Blender could fork and we all then get what we want.

    I would rather not have the BGE assimilated into Blender as some interactive walkthrough. Despite it being the lesser child its demise would cut out a valuable entry point into game creation that is not served as well by other programs.

    Sadly I think the lack of success of the BGE is mainly due to the BGE not having enough support or focus placed upon it (despite great efforts from many contributors)- how many collaborative projects have been games? Look what happened when one game was made (Yo Frankie) and the developmental quantum leap that caused.

    • If you look at the activity in BGE code the past 11 years, there’s no reason to expect a fork to become successful. In contrary, by designing it to re-use much more of Blender’s code itself, it will get quite a lot of more support from everyone who works on Blender. We also will make sure that the current GE coders support the plans too. So who’s going to code a fork? And why didn’t they then help out in the past?

      The original design decision (back in 2000) to make a fully separated engine is just holding us back now. Similar – potentially much better – functionality you’d achieve if you choose to integrate it.

      You should also not treat my definition of “Interactive 3D” to be a step back. That’s the best you can get from current BGE now as well. We should just be honest about that.

      I also want to bring focus on tools, on creation, focus on artists. Not on technology or making something “you can use to deliver and sell your games” like Unreal, Ogre or Unity3d.

      The reason why Yo Franky could happen was solely because a single volunteer developer started to fix a lot of bad bugs that were open in GE, and help adding substantial new features. He was doing that for 3-4 months already, which was the reason why we could decide to complete Yo Franky in BGE. And even then, it was still more a prototype than a real game.

  17. I hope the BGE will still be available whatever its future form.

    It’s a great tool to experiment 3D games, especially under Linux where I don’t know of any other alternatives to engines such as Unity3d for instance. (which is Windows and Mac OS X only for design 🙁 !)

    • If this is going to happen, at the very least this ‘interactive’ session needs to be realtime in nature, and have a standalone player available. Otherwise its a betrayal of all the effort and passion that many people have invested into the BGE. With this news, its going to pretty much kill any enthusiasm for the BGE as it is (along with GSOC projects) without some solid clarification.

      • A player would still be made and possible. I also count on that a refocused BGE would be much more feature rich and usable than it is now. All I propose is to focus on tools for game artists, not on building a platform for (final, market ready) game delivery.

  18. maybe a stupid question: will blender get some security options?

    i mean options to restrict python scripts to read/write data from/to disk and/or internet and restrict the auto execution of scripts in a .blend file?
    [x] restrict auto execution scripts on loading .blend file
    [x] restrict disk IO
    [x] limit disk IO to path of .blend and sub folders only
    [x] restrict intranet IO
    [x] restrict internet IO
    [x] restrict execution…

    and so on, i think you know, what i mean.

  19. Interaction mode – Sounds OK. Sounds more like PureData and Processing. For some fun stuff. Nothing for serious gaming.

    As a gamedev I would have more liked to see a rework of the GE, seperation of work and development from the Blender code and dev team, good interfaces, a competition to Unity.

    See you gamedev-people at another GE!

  20. I would like to see in the “main” view a mode in which you can walkthrough the model. Currently a model of, say, a house is hard to see.

    I would love to see more incremental improvements – what has happened in the 2.6x series has bee really great. I get that internals do need changing sometimes, but I would counsel against incompatibility unless there really is no other option.

    Finally, many thanks for a really GREAT tool 🙂

  21. Removal/re-targeting of the BGE into a “realtime preview” or “interactive mode” sounds very reasonable to me for much the same reasons you posit, Ton. Lack of features/visuals compared to the ‘competition’, the inability to use GPL engines for distribution of games on many platforms, and lack of development effort. There are open-source engines out there compatible with Blender (some loading straight from the .blend file itself), so I don’t see this killing off games through Blender.

    With that said, can I ask if/when the focused code & patch tracker cleanup that keeps being bandied about is going to occur? New features are all well & good, but sometimes a concerted effort to clean out the cruft and apply the contributed patches becomes necessary. Blender 2.7’s “incompatible changes” are going to make a lot of the contributed patches unusable (for those that aren’t already) – is there going to be a push to get this sorted out before then?

  22. I’m ok with the new direction for the BGE, but I’d like to see the image editor mentioned somewhere in the roadmap. Specifically developing it more so that Blender could support image layers. The texture paint mode is great, but it can’t really shine without layers (I use it for stuff like hand-painted characters).

    Also, I teach Blender at the Academy of Fine Arts in Warsaw, and I see how image and texture management is confusing and non-intuitive for students. For instance assigning an image to a mesh to paint on – select mesh, choose image in uv/image editor – is not intuitive because usually nothing happens when you choose the image to indicate it was assigned. The render result sometimes replacing a texture is confusing. The task to manually name a texture, an image it uses, and the file that image uses, is daunting. Etc.

    So that could use some work IMO.

    Other than that, the rest of the mentioned planned features sound yummy.


    • The proposal above is just the broad outline. Just because it doesn’t mention layers, as it didn’t mention every other minor detail, doesn’t indicate that it’s been ruled out.

      • I would say mentioning the image editor would belong to the broad outline.

        • Check “get involved” on

          Such feedback we can never get enough. Usability and streamlining tools is work we do all the time, and can just be marked on our active todo list.

    • This article is more about the “big picture” of Blender’s future.
      But I agree with what you said about texturing. My students have the same problems. The whole texturing system could really be more intuitive and. I dream of a kind of UV-Texture assistent with interactive features and more live preview capabilities. If there is a talented coder anywhere who have ideas how to manage such a system, he could become “my heroe”. I think, it would help Blender to make a big leap forward.

      • This is planned *Now*, on this GSOC’s paint branch. But it’s not clear to me how you would envision this UV-Texture system.

        Currently there have been two solutions that came up:

        * Reuse material options (which textures assigned materials use? Paint on diffuse/specular etc layers of material, use assigned material UV layer)
        * Use a combination of UV layer + image selector to create the painting mapping instead of going into and out of edit mode (well, may be required for unwrapping), image editor etc.

  23. It would be great if Bullet was integrated much more cleanly. What about a function that bakes everything (physics, particles, water… everything) at once and handles interaction between them? It shouldn’t be that difficult… 🙂 but it would be very handy.

  24. The roadmap for GE is indeed radical. But I think it’s going in the right direction.

  25. I have a project in the works

    Blender open project:Wrectified

    I have artists, and art already, and a roadmap but I need some help coding. I use logic, and python and physics to solve problems, but a fresh take on everything is welcome, as this is a open project.

    3d logic node system for assembeling vehicles, robots, puzzles and “Map logic” like briges and doors etc.- Adressless comunication via rayCast() –

    “Torque Doll”-Keyframeless animation system -

    ??Terrain LOD??

    ??Object LOD??

    ??MAP EDITOR??

  26. Continuted-> “Project:Wrectifed” –


    Modular architechture designed to drop unexposed faces,

    PS- We need some one who writes GSL shaders

  27. So, I guess this will be the end of the GE as we know it.
    I think this is a step in the wrong direction!
    Converting BGE to “Blender Interaction Mode”..c’mon , this means the end of BGE. The great community will be gone instantly, no one will use it as it is used now.
    Blender will become just a modeling tool and this so called “Interaction Mode” will become something like “Blender physics”- it is there, it is used, but noone cares and no one talks about it.
    Sadly, but I guess this is the future for our beloved GE.

    • A plan to fork the BGE is being talked about over at Blenderartists, have a peek and support it.

      No-one who is using the BGE wants this merging so it does look like a parting of the ways.

    • Right now, I use both the GE and the built in Bullet physics. I think they could both benefit from this as they currently are similar (but with different features). I do see your point though, but I think that BGE will never be to the point of Crystal Space et. al. This doesn’t mean that the BGE shouldn’t be developed; it just means that for people who are developing games, Crystal Space et. al. will probably have more features and speed.

    • I understand your concern, but the last years have shown that the BGE as we know it will forever stay like this because no one develop it seriously.

      Honestly, I’m glad of the end of the BGE as we know it which has always be more of a burden than an asset. The BGE cumulates only the negative aspects : not integrated enough to benefit from blender viewport, APIs, physic sytem, etc…. but not independent enough to have his own level editor with dedicated tools like other game engines.

      Convergence of BGE and Blender will boost both the BGE and Blender. Features developped by a regular Blender dev will also directly benefit the BGE community.

    • It really doesn’t look like this is the case. In Ton’s other comments he has said “a refocused BGE would be much more feature rich and usable than it is now. All I propose is to focus on tools for game artists, not on building a platform for (final, market ready) game delivery.”

      This doesn’t sound like leaving BGE behind, or leaving it as it is, it just looks like Ton’s not interested in turning Blender into a publishing tool.

      • Blender is a tool, so what we can do well is to focus on creation. That’s where most coding efforts go to now, if we can make sure the BGE shares this well, it’s a complete win-win situation.

        Creating a game-delivery platform is a very different job. We could acknowledge that it’s a specialized project to handle, for which we can’t attract developers really (nor does any open source do that now).
        The open source engines (Ogre, Gamekit) are really far away from what commercial products offer. That’s not because they’re “not good” or so, it’s just a huge venture, which currently only has been proven to work for commercial environments.

        I really love challenges, and always like to do things nobody thinks is possible. But I’m also realistic enough to accept our limits, to make sure we cooperate and focus well – for Blender’s future.

        • Ton, we just love the idea of Blender being capable of publishing games, however this will be not possible as you said. Good, ok, so if some people wants blender to be a full game publishing tool, it could be but in combination of another tool in the workflow, maybe as Gamekit (even though Gamekit lacks of Python support and such), what do you think will need Gamekit or another engine to be very tight with Blender 3D and make the full-game-publishing-tool dream comes true?

  28. I once saw an awesome game scripting environment called NEMO from virtools france, soon after It became part of AutoCAD I think, now maybe it’s at Dessault, but discontinued. Rad Node scripting I guess. I’m a newbie still.

  29. I think the BGE should stay. I really enjoy it MUCH more than Unity. Granted that it has less features than unity but with good knowledge of scripting and some shaders we can “hack” it to look as good. The BGE provides fast workflow for game developing and especially for those who don’t know how to script. Please don’t remove it!

  30. Personally I’m a bit worried about the potential of parallel development of 2.7x and 2.8x.

    Scribus team has been there, and it slowed them down *a lot*. E.g. v1.5.0 has been in the works for years. I recall myself demoing some of the new features at an exhibition as long as 4 years ago, and it’s still under development with a stable release probably a year ahead.

    You’ve been doing a great job at exterminating bugs and keeping a low amount of reports. Imagine doing twice as much work.

    • Yep, I’m well aware of the risks. But for some parts of Blender (physics, point caches, particles, hair, cloth, etc) we can only really fix it if we can completely rip it out and assemble back again.

  31. @Ton:

    This roadmap sounds promising. I also like to see the physics systems being merged, for example. But I wonder:

    -Will game development still be possible at the level as it is today?
    -Will games run slower or faster with this new engine?
    -Will there still be the option to export to Runtime?
    These are the questions I have for now.

    I like Blender so much because it’s a whole in one package. I really appreciate all the work that’s put into it. I hope I’ll return the favour by one day finish my project. My goal is to create a descent platform game and animation feature combo.

    Greetings from Belgium

    • – I think game development would get to much higher levels, a focus on good tools and full support of turning everything in Blender interactive, will unleash a lot of creativity.
      – Speed is something we want in in Blender everywhere. I mentioned we’re recoding our viewport and animation system to be fully threaded. That’s something the current engine doesn’t have.
      – Players or runtimes can simply be kept supported.

      • This is great, Ton. This is only good news! Thanks for your reply and many thanks for your commitment and heart into the Blender project.

  32. Sorry, what I meant was: “all in one”, of course. 😀

  33. @Raco
    I wanted to ask the same questions. I am not making money with BGE. But i do make some games for nephews and little sister.
    BGE is the only engine that makes it possible to someone like me(non programmer) to make games.

  34. You see? BGE sounded like it was not important. Now everyone talks about it!

  35. Yes, everyone *talks* about it. Of course, people *always* talk about features when someone mentions they might be removed and, more importantly, if *talking* about something were all that was required to make it work in the real world, we’d already have our flying cars 😉

    The issue is there is only so much development effort & focus to go around and BGE isn’t getting that effort/focus.

    • Then a fork is what is needed. The developers who are improving the BGE do what they want, and regular Blender gets is ‘interactive walkthrough’ mode that the artists apparently desire.

      Its just a shame games like Dead Cyborg will never be made with Blender again. You know, that commercial game made with Blender.

      • I rather see people join forces than forking. Also remember this is posted as a proposal. It’s just good to have this discussion about the future for BGE. Solutions or decisions might well be many months or even years away. 🙂

  36. But is it even certain that this new engine will have less functionality then the current bge? I would like to hear this from Ton first before giving a fork my thumbs up.

    • What is clear is that the standalone is dead, even though players exist that sidestep license issues.

      Its the standalone which is my biggest problem- I have no issue with merging the BGE with Blender if similar functionality (i.e. export) is
      still there. From what is proposed games will be difficult to make as all activity is within Blender, despite the visuals being hopefully better with tighter integration.

  37. This is great, but just a quick note to @jeremaya, no serious game studio will use Unity as their game engine for the one very simple reason – Unity uses C# and serious AAA class games simply cannot be made with it for performance reasons. There are some great game engines which use C++, “CryEngine 3” is one of them. And if you are using it in non-commercial way this superb engine is completely free!

    • And completely impenetrable to beginners, unlike Blender GE and Python/ logic.

    • Mojang, Zynga, Ninja Theory, Cardboard Computer, among MANY others have used Unity to release highly successful commercial games. Going just by profits alone, I’d call all of them “serious game studios”.

    • Erm, writing this from a serious game studio that uses Unity? Granted, we’re not triple A, but mobile, online casual, and console/PC indie all have a better ROI than AAA, if you’ve got what it takes. In these spaces, the quality of your ideas, agility and ease of development are the bottlenecks, not “performance” – which is only marginally improved by going to C++. Not to mention that C# is so easy that even the artists are scripting stuff here. All in all, we don’t have the manpower to even approach the AAA levels where this is remotely a factor, so much like Blender, C# makes total sense. No reason to run out and buy an expensive, commercial 3D package that has muscle simulation when you’re tasked with creating and animating several characters inside a single month’s work. You’ll never get to the point where that’s even needed.

      • @SBN, @Gerald, @Matt Heimlich
        Guys, I’m not interested in having language war – I’m far too busy, so the only point I’m making that if you (one) is going to tackle highest mountains, C++ is the only viable and sensible choice.
        And @SBN, couple of points specifically to your post:
        “agility and ease of development are the bottlenecks” – It has been proved zillions of times over the last decade that ease of development is virtually identicall where it comes to Java, C# and C++ – last report I’ve read on the subject was couple of years ago while I was doing my degree.
        Second point:
        “not “performance” – which is only marginally improved by going to C++. ”
        C++ is faster than C# in the range of 4 to 13 times (even MS admits that). Now, to put that into perspective and give a meaningful example, lets talk about driving a car. When you drive with the speed of 10km/h, I, with worst case scenario drive 40km/h and best case scenario 130km/h. You moving to 20km/h which is still crawling, I’m worst case scenario 80km/h and best 260km/h. You going to 30km/h which is still, you have to admit slow even for city drive and I’m: worst case scenario 120km/h best case scenario 390km/h.
        If that not convincing than I don’t know what is. And whatever you may say, gaming industry IS the industry where every CPU cycle counts.
        Last but not least, in case you didn’t realize all AAA games are made in C++, most MS products (including Windows, Office and what not) are made in C++, they moved Visual Studio to C# and you know what happened? People started complaining about sluggish performance.

  38. I’m glad to see that sometimes idea posted on BA goes up to the Blender Team (I’m thinking about the convergence between BGE and Blender).

    This convergence of the BGE and Blender is a big challenge, a lot of potential issues at the beginning, but this gives a future to Blender not even commercial apps have ! This is awsome.

  39. I remember the days that there was a business model around the BGE…..the good old days. Now its the child in the family causing th problems…..

    I think its strange in a world of growing Indie Studio’s, mobile gaming, upcoming webGL to think that a BGE or whatever tool would not be usefull. Small studio’s dont need or can’t pay Crysis/Unreal visuals, however having the option to is nice….

    For me it sounds like a surrender. Never there has been much attention for the BGE, it’s the third wheel on the car (Sutch frase). No funding projects, no game projects(franky came with the short)….NOTHING

    Greets and luv for this piece of software…

  40. Ton, i think the overall roadmap is great and almost ensured me that Blender is going to be the 3d package i always “dreamed for”.

    Though, i’m wondering if you have any idea (or at least considering) about recruiting more paid developers (and how/where get money). The roadmap you wrote implies lots of core stuff, to be developed side by side with features missing atm, improving the current ones, stability etc…

    So, any idea on future funding topic? Lets say the job can be accomplished by the current devs in ‘n’ years, but what about (finding money) to hire the devs who contribuited deeply to some Blender specific topic, each one on its own preferred task? Example: BroadStu on Cycles, Nic-Bishop on sculpting, psy-fi focused on texture painting, Jason Wilkins on sculpt or viewport performance, Miikah on fluids/volumetrics…hope it makes sense.

    These guys will be hired (luckily for them of course) in no time by competitors, loosing forever their contributions due to market conflicts, as already seen for some of them. Can we be able to hire them, have you got any idea on how to do it? Or are you fine with current method? (you + 3 devs, the rest volunteers)


    • We work on many levels to ensure our talented volunteers can also get a future job. Check on for example, and the Blender Development Fund. I also intend to organize a next open movie project with as target to provide longer term jobs for active coders.

      • you could also try a kickstarter to raise funds for development.

      • “Ensure good I/O integration with external game engines, similar to render engines.”

        So I am assuming that the game engine will merge with Blender and plunder will have a export feature to unity, ogre or crystal space?

  41. To integrate GE in Blender – doesnt this mean we get access to many more functions that we already have in the renderer/3D-View? (Particelsystem for example)

    It reads like: With integrated GE you benefit from all upcoming features that come to Blender, and we dont have that much extra work to apply it to the old GE.

    I cant find the big negatives in Ton´s post and i´m surprised how many people pick out the GE-roadmap.
    I am more concerned about 2.7. ^^

    But also, i´m impressed once again to see how Blender rushes forward and that the team has no fear to take risky overhauls in attack – in the end, it will make Blender easier for users and programmers.

    This update quantity and frequency is outstanding.

  42. MG > “So, any idea on future funding topic?”

    The current funding system isnt used to its full extend; thats what is causing the problem! 5 euros is the minimal donation base and people are not even kind wnough to donate that.Lets say that the minimal donation amount was 1 euros; and lets say the number of people using blender in the world was only 50.000 users , if each donated 1 euro each month 50.000 euros/per month easy as cake. or 10.000 serious users which donate 5 euros… We need better promotion of blender and of course donation rallies, people pay ridiculous amounts of money for small things they dont use why not pay ridiculously small amount of money for a good cause ?

  43. I’m chewing on this. With GE integration all you’d need is some stronger collaborative tools and asset management and you’d be 90% there to replacing Opensim and SL. Snowcrash here I come!

  44. I think that we never get a professional Game Engine in Blender!
    My feeling is that Blender eats the GE.

    Make the BGE better than Unity, Unreal and Cryengine.

    More power, online mulitplayer support, export to android,ios,
    ps3/4,xbox and so on.

    Why not?

    • Unity, Unreal and Cryengine are backed by millionar companies with thousand of people working 24/7 on it.

      So can we directly compete with them, using the same scheme (i.e a classic game engine)? Of course not.

      So what can we do? Well use the only yet HUGE advantage we have which is Blender itself. Unity, Unreal and Cryengine don’t have any of the modeling or animation tools of Blender. They’re constrained to a constant import/export with Maya or 3dsMax.

      A convergence between a 3D suite and a game engine has never been done before even though we all feel this is the future because movies and realtime graphics will (or have already? just look at UE showcase) obviously converge, and so must the softwares.

      Other companies are already making research on it (eg cinebox from crytek).

      Just think about all the posibilities of this project !

      -You want a live studio ? Set up a green wall in your room, add a green key in Blender, switch to this “interaction mode” and here you go ! Ready to broadcast.

      – Imagine blender gets mocap (with several kinect for e.g). Press record if you want to make a movie, or switch to interaction mode and you have a full kinect game (and you can even play with the unified physic system).

      -Someone makes a crowd simulation ? Use it to make epic battle in your next movie, or switch to interaction mode and you’ve almost done a strategic age-of-empire-like game.

      -SPH fluids, rigid body, soft body (and potentially smoke sim but not with the current implementation) are all suited for both realtime and movie. So why bother make to different implentation not share the tools we develop for them (such as rigid body constrains tools)

      Also I can tell you that Blender will become n°1 in Universities for AI (and even robotic) research/simulation. Blender could easily replace OpenSim.

      The is the most ambitious project since Blender exists. Blender has always implented features other had already (sculpt mode like zbrush, modeling tools like 3DS&Maya, etc…) But here, it is the first time Blender targets something that competitors don’t haveat all, that’s true innovation :).

  45. Hold the phone! Blender GE getting integrated does not mean the end of the road i am exited for the lack of ideas there are coming from everyone that dioesnt understand this IS the correct direction innevitably even if you look at unity 3d they have developers going the oppositr direction there trying to embed animation and modeling tools i mean theres no way unity can catch blender and now that our GE will be embeded doors will open we didnt even think making blenders GE waaay suerior to anythingthat can even exist. I personally think universal scripting is what we need asap. If something gets developed to be able to build to anything, java, c#,html5, or w.e and vice versa if it is possible to import things for example i dont exactpy know how this works but in maya there was this plugin craft designer studio to simulate real car engines and what not for car commercials and it used scripts for controllers so it was like a game and to be able to import and use those scripts would be ridiculous especially if blenders game engine was integrated in the middle of a scene you pick up a vcontroller and drive through a brick wall that has fracture and physics all embeded perfect stuff!!

  46. What I would really like to see, would be a focus on that “Interactive Design” sort of idea of using the Game Engine. For example, using it for things other then playing and making games. Such as live video mixing and realtime data-driven broadcast graphics. This can already be done to an extent, but imagine if you could use all of Blender’s tools and interface for live production work – integrating everything needed right into Blender. From live video mixing to editing to recording, streaming, every part of the workflow – all automatable with nodes. The ability to put the input of a live capture device onto any material and have it playback in realtime right in the viewport on a surface, and combine that with traditional vision mixing of multiple sources. I think something like this would be incredibly powerful and anyone using it would see their capabilities and production quality skyrocket to television levels, all for free – they just have to have a good PC and capture devices/cameras.

    A project like this would be about as big as the BBB re-factor of the game engine I think, and I think rather than being confined to the game engine (though it should definitely leverage those features), it should be done as a sort of new feature set to Blender as a whole – Blender Live or something, with things like a few new window types to handle that live functionality and switching, including updates and a new mode to the Sequencer for editing clips of video quickly (even automatically with dummy/placeholder strips that could pull in video as it’s recorded/streamed for doing things like instant replays and automated highlight reels. A chat and network feature to enable team collaboration on live projects, a simplified web application for remote operation and control of the entire live feature set. This sort of thing is kind of a dream of mine, but I’m no programmer. I think Blender would gain a lot of users by having unique and deeply integrated, high performance live video features. Pre, live, and post production all in one powerful, fast, cross-platform program.

    Sorry for the length, I just get so excited talking about this stuff.

  47. Great! With more performance in general and in the viewport, it would be fine 🙂

  48. Python “Custom Editor” api? Sounds like it’s what I’d need to turn Blender into a fully featured Level Editor for just about anything using Python, right? Looking forward to that.

  49. Finally, clearly sane thinking towards BGE. With that, any chance of add tool that would allow to create and author assets for any other video game engines? (batch exporting, better animation/rig handling, better scattering tools, more advanced texture baking tools, better decimation tools, UV/vcolor/etc. preservation, vertex normals manipulation and preservation, etc.)

  50. Thanks for the responses Ton!

    If a player exists like today then its a good thing. If tools from the non BGE side (like particles) are available to the interactive side, that would be an interesting turn of events too.

    One thing, I am learning Python within the BGE- in the future, how do you see the API changing? Are you looking to start from scratch or clean up what exists?

    • I really don’t know how APIs would evolve. That’s for the teams to sit down and look into. Having two completely separated APIs in Blender just doesn’t seem to me sensible really. Imagine the cool extras if you could use the full blender py api in BGE now, including all UI control?

  51. Personally I believe that the Blender community has made efforts to develop and showcase Blender as professional tool that can be a credible alternative to Maya, Cinema 4D and others. And the target of that message is the professional artistic community.

    If this is true, then I consider it important to understand how a professional artist can use 3D in real time for professional work and work from there.

  52. I love the roadmap… was hoping to get the SPH fluids more refined, allow for smaller particles, more along the lines of RealFlow…

  53. Might the idea of more highly integrated interactive logic allow for creating interactive tutorials?

  54. This sounds great for my project to create a presentation program based on Blender Game Engine(“BlendShow”, linked to from my name). As long as no game engine features are missing from 3.0’s “interaction mode”(yes, I’m a little concerned, but mostly optimistic), it sounds like the version that I base on Blender 3.0 will have MUCH more flexible mesh generation. Currently I have to generate meshes outside of the game engine.

    Judging by the 2.49-2.5 transition and your mention of “During the next few years”, I am guessing it might be a year or two between 2.8 and 3.0? This will give me plenty of time to catch up with Blender’s version numbers(my variation is still in pre-alpha) and sync version numbers with the Blender it’s based on after I pass my initial 1.0 and 2.0 feature goals.

  55. From my blog:
    In the new Blender Roadmap, it is admitted that not much time or effort has been spent on the Game Engine. Well, as with governments spying on ordinary people, many people suspected that it was happening but it’s still sad to see it confirmed. Starving the Game Engine of development and resources hasn’t killed it though, just annoyed its users a lot.

    Another paragraph in the Roadmap serves to illustrate why this happened:

    “On the positive side – I think that the main cool feature of our GE is that it was integrated with a 3D tool, to allow people to make 3D interaction for walkthroughs, for scientific sims, or game prototypes. If we bring back this (original) design focus for a GE, I think we still get something unique and cool, with seamless integration of realtime and ‘offline’ 3D. (quote) .”


    “Instead of calling it the “GE” we would just put Blender in “Interaction mode”.

    If starving it of resources doesn’t work, I suppose you can always try trivialising it and then absorbing it.

    Again, I for sure and many others, thought this was probably how the Game Engine part of Blender is regarded – a neat little toy for artists, a frill, a grudgingly maintained prototyping tool for game makers – but I’m afraid seeing it written down makes me realise all over again that I think they are chasing the wrong doggie.

    The Indie scene is very lively indeed just now (as I predicted at begining of year) and showing no sign of slowing down. It is a bona fide growth industry with a growth outlet (apps and handhelds) as well as more traditional outlets, now not niche at all. It is painful to watch Blender deliberately missing this particular boat. This is a field it could easily have dominated if the effort had been put in. Not bothering with the Game Engine was a bad choice. I am glad for the inroads into film, but film just isn’t growing as fast and big as development for handhelds and apps. The Ouya should be along soon too.

    Not dealing with the licencing issue didn’t kill the Game Engine either. Enthusiasts will go to the lengths required to sidestep this limitation, though in all honesty it should have been addressed ages ago.

    Plans to subsume the Game Engine into Blender and redefine it as “interactive mode” has caused some debate as is evident in the comments on the link I gave and on various forums. Something of a hornet’s nest. Apparently this surprises some people. Well I think it’s good to get the issue out into the open at last. It’s about time the people at the top had a look at user demand, user experience and saw how their plans and outlook are regarded by Game Engine Users. I don’t see any need for vituperation, but honestly…. “interactive mode???”. It remains to be seen whether this will kill the Game Engine at last.

    I hope not. I can offer some insight into why it hasn’t died yet, despite all efforts not to support or sustain it. It hasn’t died because there is a void, a market gap the size of the San Andreas fault, for just such a thing. Here is a magnificent, accesible graphics editor and creation suite, there is a vast crowd of independent game makers, and in the middle are some engines. Blender Game Engine sits in amongst this group like a little island. The others grow bigger and prosper whilst the one that fits best with the graphics stays small and ineffectual under a weight of apathy and quite possibly dislike.

    The other engines, (I know, I’ve looked), are buggy/expensive/cumbersome/unsupported/untutorialised/undocumented/don’t integrate with other applications – sometimes all of the above. But most important, they aren’t in pole position to take advantage of Blender’s massive graphic power. What is happening is that game makers who like the BGE use it as far as possible then seek out their own solutions for all the things it does not address (and there are many). This means that the BGE never gets to be in the final credits, never gets to be the big name. It also means there are as many solutions as there are people looking for them. I also will be following this route. Another uncapitalised advantage being that the Engine’s source is accessible. That is such a plus for an Indie. At the moment, it’s a mess, but all this hasn’t killed the engine either, just fringified those who still use it despite the limitations and incredible annoyances of doing so.

    It’s needed. That’s why it survives.

    In my view, the Blender Game engine, with it’s old code needs a complete rewrite which would be a perfect time to sort out a serviceable licence. It should be integrated more closely with the graphics as planned, it should become feature rich, supported, sustained like the rest of Blender which already operates beautifully as a modular system. What has happened is that one of the modules hasn’t been worked on and has fallen out of step. This can be fixed. But the will clearly isn’t there.

    If the Blender Foundation does manage to kill off the engine, something will spring up to fill it’s place simply because something is needed in the position it currently occupies.

    Film is all very well, but it doesn’t have the kind of big future that games do. It’s amazing how well Blender otherwise covers the 3d field. It is easily used for visualisation, teaching, a million other things. But on games the Blender Foundation is dropping the ball badly.

    As in any gold rush. the people making most money (and publicity) from Indie gaming are not so much the developers, but any company that integrates with their efforts. Publishers, platform makers, tool makers, publicists, lawyers etc. But apparently… not Blender.

    • May I remind you that “Blender Foundation” is not making Blender? It’s a real community effort. All I can do is to help focusing, align forces well, and try to be a good facilitator for projects.

      Blender’s target is to provide a complete open source 3d creation toolset for artists and small teams. In that respect we do great, including in game creation pipelines (similarly to other 3d tools).

      However, where 3d tools are common to to deliver a full finished short film, there’s no comparable product that does it for games. Game delivery is still a market (and method) dominated by specialized engines, optimized for platforms and for highly specialized constructed content to play. Even mid-tier engines like Unity3D focus on that delivery aspect (and not offer it as creation tool).
      Also all the other open source engines are really far away from becoming viable alternatives for the commercial engines. That’s not because they suck, it’s just a huge specialized area demanding enormous time investments.

      Realistically we just cannot expect to be able to realize ever anything that is even close to the gaming experience and appearance we know from mid or high-end games studios. Simply because they also don’t use a tool like Blender to do it. The entire process is too specialized still, with a lot of middleware and tools, and teams with plenty of full time coders working on achieving the results.

      If your analysis of *huge* future for games was correct, it would have shown and happened in the past 11 years for Blender too (and for other open source engines). Its even why my company got so much investment in 2000. We crashed badly, and for reasons that are valid today still. It’s just not so simple.

      So – let’s do first what we know we can do well. And when that works, build on top of that further. Who knows this refocus then leads to what you (and I) dreamed of.

      • Not that I am proponent of keeping BGE alive, but this statement “Also all the other open source engines are really far away from becoming viable alternatives for the commercial engines. That’s not because they suck, it’s just a huge specialized area demanding enormous time investments.” is not true.

        There are 3 complete open source engine, which one can grab and make a game using those. They are Darkplaces engine, idTech 4 and Doom 3 BFG engine.

        I have been proposing to focus on creating tools in Blender to help developers using those engines, instead of pushing BGE. No one really listened.

        • Do those game engines allow me to CREATE, Modify, Script, texture, light, and have all my stuff in one space? Or do I need to import FBX objects and then teir those objects to the Engine’s specs? (my guess is the later). Unity, Crytek, idTech, etc None of them allow the creation of 3d content within their game engines, they just manage / manipulate the 3D content imported into their packages. Blender literally has something unique in that Middle ware isn’t NEEDED (it might be desire for faster productivity or better results but it’s not NEEDED).

  56. Another addition to the Blender roadmap would be the addition of three editors or sets of tools.
    1. The ability to use a sequence of images to create a 3-d model with as a means to obtain virtual objects from real world images.
    2. The ability to create bump maps (and other texture/material maps) from real world images to enhance the realism without having to resort to outside programs.
    3. An audio track composer that would allow the creation and control of music/sound effects, and audio placement (by moving speakers or linked to objects) with support up to the current surround sound formats.

    These tools would enhance the usefulness of the Blender suite of editors.

  57. Hey Ton, what you call momentarily ‘Interactive Mode’ is something similar to update the rendering in real time using the Cycles? Type, we changed the lighting and it automatically reflects in the render?
    If anything, it will be incredible mess in BGE.

    My only question is: Will there still be able to export the project developed the new GE as executable, type via the add-on added in later versions to 2:58?

    I understand your explanation as prototyping, the do not believe this to be a problem for those who like to deal with GE.;

    Anyway, Blender as a whole is moving in a surprising manner. Who knew this years ago?

    • At the moment this is just a review and discussion. Answers about what would work, and how, is really in the hands of those who will actively work on it.
      Export something as executable to autorun would well be possible though, I wouldn’t know why not to support it (apart from security reasons).

  58. IMHO new Blender needs a large API for plugins. Not only for Python addons and plugins for rendering but also for Modifiers and even third-party GE plugins or OpenGL rendering ones. And why not to use such GE plugins for this new “Interaction Mode” ?
    As for me: i want to make some commits to BGE but i could do this easier by commiting not to vanilla Blender but to some plugin(s) or even by forking such plugin

    • And i think it will be good to make realtime renderer a separate replaceable library (.dll/.so) to allow end users/developers to modify and build it by themself apart from Blender.
      p.s. Waiting for reply from anybody of Stokes…

  59. The road map is looking good. Great to see that the sequencer is getting an rewrite. Would be great if there was a way to exchange information between the compositor and the vse. Keep up the good work ton.

  60. The new goals for the blender game engine are perfect! I think blender shouldn’t be a professional programm with a laborious interface. In my eyes blender should get smarter and more intuitive

  61. Hello Ton,

    As far as I can tell, the plan looks excellent. This is how I understand your BGE section.

    – Code sharing (drawing, physics, shaders, lights? … ). Indeed, this should we a win-win. If the same code paths can be used for both worlds it will certainly result in very robust paths (more testers and maintainers for them).
    – As many activities are common to the non-live creation and the live (game) creation (texturing, lightning, viewport rendering), the current workflow won’t be degraded, the code merge can be totally transparent to the user (Pkey still starts the interactive mode). With time, however, the workflow may evolve as new tools can more easily be integrated and maintained.
    – Game Player == Heavily ifdefed blender where main thread skips UI and some modules are skipped (eg: VSE)?

    Is this the idea?

    • Basically yep!
      But – maybe someone wants to make VSE game, who knows 😉

  62. Regarding our beloved GE:

    As long as we have bullet, logic bricks and priority for GNU Linux builds everything will be fine.


    • Blender stays fully cross platform, not any OS will need priority.

  63. It is great to see how Blender is developing and becomes mature application.
    About developing, I understand how much effort it is to make a good product in open community, but I am an architect and I would like to see these features developed further:

    Must have:
    – Cycles speed improvements
    – Physical Sun and Sky improvement
    – IES files support
    – Colour temperature
    – Physical correct camera controls for Cycles
    – Light/sky portals for better/cleaner interior lighting

    Nice to have back:
    – Preview of particles in particle paint mode

    Will be great:
    – Better viewport performance with big scenes and particles
    – Improve Scene Outline (select multiple object, group them by categories,rename set of similar objects in one go)
    – Nodes for modifiers and for modelling (like grasshopper, but for mesh)
    – Further develop NURBS

    I know there are ways around all these, but I think it will improve usability of Blender and not only for Architectural rendering.

    Thanks for great job.

  64. It all sounds like a good plan. I just to lend my voice in positive feedback. Blender is a good tool for artists. When It comes to the blender game engine im looking forward to the improvements that will be made in this design refocous.

  65. Just dreaming, but if the BGE could be incorporated into browsers as WebGL has, that would be wonderful.

  66. I like it… Particularly the sequencer rewrite, that is placed *before* the game engine focus.

    I would suggest that the sequencer rewrite should have a similar idea behind it that the blender “interactive mode” has, that is, it should become a fully integrated part of blender, and should have a similar structure and flow.

    Currently, for example, the sequencer uses the term “group” this function is extremely different than the groups in regular blender.

    Features that could be implemented are things like parenting. Clips could be parented to one another. More complex constraints could simplify video editing actions like making sure that clips end together, or resize relative to one another. An Array modifier could be used to loop a looping background clip.

    It would be more consistent if, compositing could happen *after* the sequencer for example. Every blend scene would have a default sequence which includes just the 3d scene itself, go into the sequence editor the select the scene and check “use compositor.” whenever you do a render you’d actually be rendering that sequence. I think this would reduce ambiguity of turning the sequencer on and off. Calling other scenes from a sequencer would use the sequence from that scene, allowing the layering that clip groups currently has, and the node editor would have full charge of effects, instead of the effects being handled in layers or in clip properties.

    • You’ll have to forgive me, as I’ve just begun to use the sequencer as an almost replacement for adobe premiere, and I really love it. The ability to video edit from home without having to purchase another license from adobe (or windows for that matter) it makes me passionate about this little piece of blender.

      Keep up the good work

  67. It’s kind of sad to hear that the BGE won’t be supported in the future as a “real GE”.

    Why sad? Unity and Unreal are quite difficult to customize in order to get a not FPS-like game, in BGE on the other hand it takes only a few hours to get any game-core mechanic working nice and smooth.

    Logic bricks makes life a lot easier even for programmers like myself, the fact that I know how to code whatever I’d like in Python doesn’t override the simplicity and speed of logic bricks.

    Camera control, simple physics and ruled-based animation are within minutes in BGE while in Unreal takes over two hours just to get the camera point somewhere else.

    What I wanted to hear for a Blender roadmap was completely different, I wanted to hear that particles and complex physics will be available, that realtime camera effects will be integrated (blur, lens flare, realtime fov changes, etc…), that rigid body joints will be available for armature bones, and so more.

    I am a software developer, I’m not really interested in blender as a 3D designing tool but as a game design tool, a really good one and easy to use.
    I don’t really know how many BGE users are out there, but as one of them I hope this tool gets really reinforced and not only “merged” into the design tools.

    • I don’t see your point, you know RODOLFO MORA?
      (Not only your, i would say, all yours, who blame about BGE integration!)

      What you’re saing about, that’s just what Ton in proposing 🙂

      Integrating BGE into Regular Blender a lot more than now, so that BGE could use all of the Regular Blender cool features out of the box (after the integration of course).

      What’s even more, after the integration, if someone implements some – i don’t know, let’s say – new animation system, than it will be auto-magically availble in BGE, with no waiting for years. In most cases, i can see, that almost every new feature will be availble in the exact same day, as the new blender realse.

      In fact, this integration is best what could happen not for the artists, but for the BGE game makers.

      There is this whole holly war, about BGE. Is it commercial-production ready, or not? What’s the point? If someone makes AAA games, then he will never use BGE for that purpose, but still maybe he would like use it for fast prototyping. If someone makes Indie-AAA games, than still he won’t use blender for that, cause Unity is far more advanced, but of course he too will need a tool for prototyping. If someone makes total indie games, than maybe he will use Unity free, or maybe he will find a way to use blender for his commercial product. If someone makes free games, then there is no problem, to use BGE. Right?

      Only two thing changes really with Ton’s proposal:

      1). Fact, that you will be able to use new features “next day” after new blender realese, and
      2). The name. Cause as i can get it, from professional point of view that’s really silly that we call BGE a _GAME ENGINE_.

      So the coders make a new feature once, and it is “shipped” at the same time to “normal” blender, and to BGE, because there is only Blender in fact, in which you can enable and disable – by pressing some magic button – a [live/interactive/whatever you call it] mode, code this live mode via bricks, or via python, and then save it as a standalone 🙂

      Would be cool if Ton could confirm my interpreation (or not).

  68. I agree with Rodolfo Mora, saying that the BGE must not only be a another “tool” for designing. I actually see that Blender has all the potential it can get for a physics engine. So, why not a game engine? It already has massive physics, accurate and precise, it has python scripting, it has modelling, sculpting, painting, rigging, Blender has everything it needs for a great and awesome GE! The only thing left, is to fully integrate the BGE to all the features and power of Blender. We will be having a much better Physics engine than even Unity!

    I personally wanted to create my games with Blender, but as seeing it was not fully built in physics, I decided to go with Unity! Come on, make the Engine powerfull!

  69. If i could add anything to the wish list, it would be better realtime scripting (or expressions if you will). Coming from a motion graphics background, those parts of blender’s API have always disappointed me a little.

    Currently there is a way to “cheat” and have a python script run on every frame, but even that has some pretty sharp limitations. Like the fact that the only way for the script to get the current time is an integer, and will crash the script if you activate any subframe options like motion blur.

    What i would like: some kind of easy way to apply expressions/realtime scripts to objects or animation channels (possibly a modifier), and a float variable for the scene’s time starting from frame 0. Seems like something that would be fairly easy to implement.

  70. It’s great to see Blender madurity but I find it very disappointing that there is no mention of NURBS in this Road Map. From a point of view of modeling it’s has a lot more priority to me than the game engine.

  71. With regards to the BGE, the biggest issue by far isn’t the technology, it’s the licencing. If there were some way to separate the Blender “Interaction Mode” Server and the gamefiles themsevles in such a way that the game can be sold in a proprietary manner over a platform like steam I think this would do much much more for adoption than any advances in technology.

    Alternately make it a completely separate project based on an MIT licence, like Ogre, but still supported by the Blender Foundation.

  72. The only absolute important thing for the gameengine to shine out like a supernova that never ends is to give the uses a possibility to “klick to make Exe” that creates a property of the owner with 0 licences.
    Just like, klick to render, image is yours.
    Klick to animate, movie is yours.


  73. Ton,

    You should really consider getting Blender on Steam since they now have a Software section. Not only would this expose Blender to 5 million Steam users ( )but we’d be able to do automatic updates across multiple platforms while giving the option to install/update plugins via Steam Workshop which features ratings, filters, etc…

    And as a personal note, some serious work needs to be done to getting rid of the redundancies to theme editing. As of now, it takes several days just to come up with a consistent theme…I hope this addressed by version 3 as I’m a bit of a theme junkie! =]

  74. @Everyone

    My personal oppinon is that we try to build a strong relationship with Valve Software. Since out of all the game importers/exporters that I’ve tried out, the ones for the Valve Source Engine seem to be the most feature rich.

    If we can get Blender on the Steam store as the exclusive mod tool for Source Engine based games, then I believe this will concurrently address many of the issues we’re facing with Blender game development, most especially if Valve takes the initutive to garner a special focus towards improving compatibility between Blender and the Source Engine…

    What are your thoughts?

    • The source engine is very feature rich, like most top of the line game engines out there. Also like most top of the line engines it requires a commercial licence if you want to sell anything you make with it.

      The whole point of software like blender is to develop free tools for artists. Tying it down to a non-free and extremely costly tool would kinda defeat the point.

      • Peter H.

        Completely disagree.

        A few things that are important to note about Valve.
        1) They donate to and have helped in the development of open source software, especially where Linux is concerned.
        2) They publish games for Linux.
        3) Gabe Newell (Valve’s glorious leader) has pretty much said that Open Source is the future, or that it will play a huge roll in the future tools.
        4) Steam Workshop allows users to create content for commercial games and even sell them on the market place thereby empowering artists in both via exposure and financial areas.

        That said, having Blender be used alongside something like the Source engine (which may very well go the way of Unity or UDK soon as they have a new version in development and based on their current philosphy) would never hurt Blender itself. What FMORTIZ is suggesting in the second part of his post is that Blender be available via Steam. This means that anyone interested in making content for the steamworkshop or just having a tool tied to Steam (easy updates) will find Blender as part of the marketplace (which includes both commercial and free games/tools). A number of tools are already there including but not limited to 3D Coat, Silo and Art Rage.

        I am not sure how you are thinking that any linkage via pipeline to a commercial package would some how lead to it being “tied down” and “defeat the point” of Blender.

        • Valve are a great company, but the source engine as it currently exists isn’t an option for most small game studios. The licencing terms are too opaque.

          I think steam distibution of blender is a great idea, but that’s just a distribution channel, not official support for one commercial game engine.

          “I am not sure how you are thinking that any linkage via pipeline to a commercial package would some how lead to it being “tied down” and “defeat the point” of Blender.”

          FMoritz proposed that, instead of keeping Blender as a general content creation tool, it should be aimed at a particular pipeline: Source. This would inevitably tie blender down by hurting the artists who work with other pipelines which in turn would indeed defeat the point of Blender (which is to provide a F&OS general tool for any small creative studio, not just studios making games through source).

  75. Sorry for the triple post, but I think this it’s important to note that Valve has expressed intrest in targeting multiple platforms including Linux. ( However in this day-&-age, many 3D authoring software programs (mainly from Autodesk) are dropping multi-platform support. Which is most evident with their 2014 iteration except for Softimage…

    Just recently it has been noted that they have finally seen the potential in OpenGL:,21876.html

    However, John Carmack (The creator of Doom) made a famouse speech (or rant nearly 20 years ago) about the inevidble downfall of DirectX:

    Since Valve Software has become aware of this fact and is trying to target multiple platforms, this may be the oppertune moment to create a relationship with them most especially since Blender is the ONLY software out there right now that is available on all platforms…

  76. One of the interesting things we are considering in Second Life is the advent of the Oculus Rift. Linden Lab, the creators/owners of SL are devising an interface for Oculus Rift users. Several of us are seeing this as something that will change how we build in SL.

    Architects are talking about how using the Rift is changing how they and their clients perceive scale. I think that will be very true of SL building too.

    As more of SL is built in Blender, Maya, and 3DS I would think more Rift owners will be Blender users. Will having a Rift UI for Blender make it more usable? Will architects and modelers find Blender with a Rift interface more useful? I don’t know. I would like to see it, but you developers will have to make the call.

    Thanks for your work and efforts, Nal.

  77. Hi,
    one more time : Thanks for all… heu for Blender 🙂

    “Fix our duplicator system, animation proxy (for local parts of linked/referenced data)”
    is that mean fix armature duplication of linked objects ? for example several instances of a linked object with an armature for each one…

  78. I miss openCL accelerated physics.
    It can be a realy big improvement in the future.


  79. Cycles – SPEED



  80. Interaction mode and Nodification, you guys are so awesome!

    Can you describe python “custom editor” api a little more?

  81. I haven’t read every reply, of course, but it seems to me that no one mentioned the improvement of the network rendering functionality in Blender.

    Let’s face it: PCs are being phased out. Smartphones and tablets are taking over. And with them, comes the emphasis on cloud computing. So, as a beginning, I would suggest stabilizing, and simplifying the use of, the network rendering in Blender.

    The next step would be, coming up with a couple of ideas about turning, say, a smartphone into some kind of additional pointing device, like a virtual trackball, and a simple UI for a tablet that lets you do some basic operations on the viewport, while rendering is done on the server side (yes, a ‘good old’ PC, initially). That way, it might be possible to work on the materials, at least, using a tablet.

    And later, when the cloud computing and mobile technologies mature, how about offering cloud rendering capacity? It may be based on SLAs, with the initial level being free (for beginners/hobbyists), the next level, say, pre-paid (for indies) and finally, some paid subscription (for professionals and studios that make money).

    Another suggestion about the use of mobile devices: how about a combined smartphone/tablet and PC app that turns, say, a couple of mobile devices with cameras into a MoCap system?

    These are just examples: the sky is the limit when it comes to the ways the new technologies could be used. Regarding the development platform: has CLR/Mono/.NET been considered? Yes, I know it’s Microsoft (evil, bad, the enemy of all things free and open source — or so they say 🙂 ), but it *is* powerful, multi-platform (.NET for Windows, Mono for almost everything else, Xamarin for mobile devices), and it supports multiple programming languages, including dynamic ones, like IronPython or IronRuby…

  82. How about actually fixing the audio not playing the the video sequence editor on the linux platform? I have been attempting to animate to audio for six months now and I can’t sync anything to sound. A little love for linux users would be amazing.

  83. I am a game artist and sadly Blender gradually falls off of my radar.

    It’s still lacking some features for basic in-game content creation. Not a kind of cool advanced animation and dynamic features but something for usual everyday work.

    No vertex normal editing – you can’t model a decent trees, foliage,grass for a game scene, outdoor scenes in general.

    No smooth groups – ok, you can workaround it but extremely inefficient.

    No projection cage and you can’t solve a number of problems with texture baking on extremely low poly environment models where you need full control over projection direction.

    No anti-aliasing in amazingly convenient Blender’s texture baker and you can forget of quality hi res textures. Especially if you textures include something hair based.

    No attribute transfer and you are doing the same routine work again and again for a typical huge game scenes.

    Blender is really cool and have a lot of great advanced features but for basic routine you still need Max or Maya. Honestly I would like to get all the game specific tools before any BGE advancement.

  84. That would be great, about interaction mode, have a similar workflow in Blender

  85. Hi,

    This is all verry exiting. The only “bad” think that I can think of is about the 3D view and the BGE RT part. In the roadmap they seam completly separate but having a shared openGL rendering module have advantages, mainly:
    -code reuse
    -realtime and 100% accurate view of the game assets, scene, materials and lighting in the 3D view.


  86. Since Blender already has ‘interactive’ editing, emphasis on developing a high quality run-time for interactive applications is already implied. With such a large community of talented developers, the resulting run-time can be quite competitive with other game engines.

    Adopting BGE performance techniques by the interactive Blender application is exactly what I choose to interpret Ton’s proposal to mean. I am 100% behind this effort and I intend to contribute source patches along these lines.

    Blender’s current window type and GE export features are already extremely useful for prototyping interactive applications. When rigs, animations, meshes, materials, physics properties and scene layout are complete, exporting to engine X is the challenge that development staff spend a lot of time on.

    The editors for engine X are always limited compared to Blender’s ability to create content. Blender and a standalone BGE run-time make a great combination for competitive games, especially in the indy space.

    Only the license slows progress of the BGE run-time as a deployment tool. If the license was more flexible, the step that translates from Blender lingo to engine X lingo could be removed, resulting in rapid adoption of the BGE run-time as a deployment tool.

    The community has asked for help with the license issues related to using the run-time. The run-time is developed for Blender, and hence uses that license. I do not see a way to avoid the license issue with reusing the run-time. The license for the Blender’s run-time, although powerful for deployment of games, should not change.

    However, I believe that a license and software design that promotes plugins and addons for commercial use will help the community flourish and Blender thrive. Studios (movies and games) will integrate their run-times with Blender, and will contribute significantly to the core Blender features. A license and software design promoting commercial integration will result in a much richer community and superior Blender application. Blender will remain free, open-source, with the same license – plus an exception for plugins and addons.

  87. I think that for the game engine, that a separate RUNTIME module would be great ! just export the needed stuf in a kind of GE xml-format and
    load it in the runtime module to play it off

    Thanks for this great program that blender is !!!

    Have a nice day,


  88. Quote:
    “•For 2.68 and 2.69 we strictly keep compatibility and keep focusing on stability for Blender.”

    Blender 2.6x is still not fully compatible with Blender 2.4x in an important regard.

    Animation drivers were given the ability to reference “self” as a PyDriver expression as of around 2.43 if I remember right. In 2.6x I can no longer use scripted expressions such as “self.LocX” for example. This is long-standing break in compatibility. Will attention finally be given to this after nearly 20 releases without it being addressed? Is it trivial or non-trivial? I’ve scoured the net for information on this to no avail.

  89. Lets get simple.
    1 the License of the player is not liked. Does raise question what license should it be. But this is not the big issue.
    2 Exporting from bender node logic to other engines is problem. Nodal Logic would be great for non coders as a fully functional thing. Big issue is you have game in BGE and you want it in some other game engine. There is no direct transformation path. Would blender GPL license be a bother if the games for final commerical release could change engines most likely no.

    Please be aware Nodal Logic has been look at as just game developers thing. Could Nodal logic also been good for people who cannot code in python very well to get stuff done like model transformation tools. Yes it could be.

    Maybe the way forwards is a programable complier that takes .blend as input and rebuilds into C++, C#, python… with different object formats for different game engines. Advantage of this solution blender can safely remain GPL.

    Just like blender having plugable rendering engines because each engine has different advantages. Plugable game engines is also required.

  90. BGE

    A fork is inevitable, but that does not have to be a bad thing, if done right.

    By a fork, one could literally gut Blender out of all the art tools – and concentrate on a Game Engine solely. The main selling point would be that the GE tool would be link to blender itself.

    For me a good fork idea would be a tool like the Construction Kits used for the Elderscrolls games. The tool does no modeling, but runs the games. Creating an environment like this, based in C++ with python scripting for the npc interactions etc, and C++ libraries for physics and other hard code, network etc can be done.

    Then Blender and its interactive mode can be used to create everything but the logic, from the particle effects to the simulations etc, basically what it seems like you want to do with “Interactive Mode”.

    The key here would not be that the fork be a bad thing, but a focus on coding, and the only thing that be needed is that the two guys talk to each other nicely and ‘possibly’ in real time.

    That is my take on it, I truly wish I knew more coding to help with such a project.

  91. Ton,
    It’s an excellent décision. Mix gameblender with général blender développement.
    Please, intégrate if possible all the new features discuss on the gmaeartist forum, like bake general lightmap in one click, ocean research by martin’s, area shadows ect…
    Please make it HTML5 / android / PS4 / IOS etc… compatible, it’ll be absolute very good for indiegames maker like me. Blender is a very interesting part of workflow to make games.
    I use it all days!

  92. Well, there are a lot of comments, so i’m not sure anyone will read my post. But will try to reach to minds behind Blender. I am a 3d artist with main interest in animation. I also do static 3d’s(like product shots). Well, i started to use Blender and now i am using it’s modeling/uvw features quite a lot. Up to the point that i want to dump all paid apps and focus on Blender only. BUT, there are clients out there, and even though i create in whatever package i want I NEED to deliver final file into a format compatible with client format. Now, for export there is FBX(witch works very well) and not to mention the BVH(outstanding, lots of thanks for that). But for import??? I think you guys should focus on Alembic as it is open format and it’s wide spread across industry. This is the best chance Blender will ever have to integrate into pipelines. If i where near developers of Blender i will tell them nigh and day: alembic, alembic, alembic, until they get it…. Wright the best alembic import export plugin to work with skeletons, skin(envelopes) and all types of animations(including particles) and you just give the Blender a bright future. Sorry for my long post, but i am a practical guy. I don’t invest effort and time into something that can’t communicate well with other apps. You need to jump all over collada and alembic if you want this brilliant app to have a future.
    Thank you,

    p.s and maybe, find some time to improve(like a lot) the weight paintings/envelope deformations.

  93. um in GE i think it will be more of a problem than a solution if it were to split off from the main program but to improve it i do support. unsure how to actually contribute coding to the project so make it easier to find out how to volunteer.
    Things which i would like to see are premade modules for the BGE for networking between computers, a way to use CUDA in the .exe export when and if supported with a way to turn it on and off when making it including a way to toggle it on and off before making the file .exe export, a way to integrate templates into blender and be able to add to them for example to have one for when making a static 3D image (like product shots) another for when designing a interactive environment (alike to games), and a preset which cannot be changed for just the basic cube, lamp and camera.

  94. Focusing on blender first is important. I don’t understand why so much work goes into Cycles (render engine) or the game engine, when there are other engines available with huge teams that are far better.
    Blender should focus on the tools that people keep complaining about, such as the weight painting etc. And playing nice with industry formats.
    Engines take a long long time to do, and there’s no way you can do it better than the ones out there. Also, with the BGE, coding deployment is way out of the question. Adding PS4, iphone etc support as if.
    Just use Unity. If you are making a game, make a good one. Unity is better than the BGE.

  95. can’t wait!

  96. Blender is becoming really strong because of your efforts.

    If I can suggest something, I will ask you to open blender to other softwares (Realflow, Rayfire, External Render Engines, etc) , I mean, give a chance to blender to enter in the production world by making it more compatible with other production-ready softwares.

    We believe in Blender! Thank you for all you’ve done guys.
    (Sorry for my English)

  97. Please sort out the X,Y and Z coordinate system so that I matches that of everyone else. It is a disconnect when you get used to an x,y,z system but find that Blender wants to make you use something completely different.

  98. I don’t know about other things but I would certainly love to have realtime virtual render with better quality of rendering than the GE now has.

  99. I like the new concept of future of Blender . But I think we should update the BGE with some ,tree generator, ocean ,realistic sun ,Lens flare ,bloom ,glare ,& others little things like that ! Gradually with little improvement also in realtime & logic brick ! Because I don’t understand ,everybody know that Blender has a lot of potential in modelign,painting,sculpting,rigging,animating in Cycle as Blender render. Blender game can be a great for game for kind of improvement like I said .I think it’s really important because we are in new generation of technologie as PS4 & we can’t export or game on smartphone & PS3 , Xbox & other like Unity .GE is a really important thing in Blender for all who use all day .

    I think we need to focus on whole of Blender to complete what isn’t perfectly finish compare Max & Maya !

  100. I would love to see the Game Engine being more integrated and enhanced in general!

    However, this is scheduled for the 2.8 release which might become true when, in fall 2015 maybe?
    Until then the rasterizer will be hopelessly outdated. The future of gaming graphics is real time pathtracing, as the Brigade engine does it:

    The downside by now is the requirement of a recent high end graphics card to run it smoothly, but by the time we can expect a stable 2.8 release the hardware requirements would not be more scary then those of Crysis 3 today.

    Maybe the Cylces Engine, which is an almost real time path tracer too, if I am not mistaken, can be used as the base for a new, path tracing based Blender Game Engine, to replace the old OpenGL Engine?

  101. i love BGE. We will use different game engines until you complete the BGE. But when we see BGE is enough for us, we will come back to home.

  102. I think that GPL is great, I’m surprise reading comments abut “would love that GE have a license that permits proprietary code in”, because we belive in Freedom and Blender was liberated time ago for that. The “permissive” licenses, unlike GPL, let people take advantage to Free code to build proprietary products, and that’s bad for the user (Freedom in free software is for USERS), and very frustrating for contributors (your code is used against you, since many contributors are users themselves). Do you want proprietary solutions? Go out and find some. Do you want to use Free(dom) software as a free (no cost) starting point for your proprietari code? I’m very happy GPL prevents it.
    Do you want to earn money with Free (GPL) code? Would love it to happen, so study a fair and Freedom-keeper business model, and/or help educate people about the importance of Free software and the necessity of contribute back

  103. This is slightly off topic but I use blender for video editing. I like it better than anything adobe has but I think it would be really use full to integrate the compositor into the vse and also more audio editing tool so one does not have to do audio editing out side of blender. I am talking about fairly simple audio editing such as separating stereo into two mono tracks and a real pitch control and speed that does not change the pitch.

    If those things could be added I would be very happy and would make editing easier. Until then I will continue to use blender.

  104. I missed the 2.5 development. I will not miss this one. best chance to show my love to blender.

  105. Hi,
    as I am working in a studio people talk more and more about Blender. Still I cannot tell them when it will be possible to export and import with Alembic. Does anyone know when 2.8 is to be expected ?
    To me this is the only point holding Blender backward. The only thing TD want to know is “can i put that in my pipeline ?”.

    And what kind of flexibility particle nodes and the depsgraph refactoring will give to the artist ? Will it affect the control we have on any data in the scene like we would have in Houdini (yes i like it, still less than i like Blender) even if we cannot use nodes but still use python maybe?
    At the end what tool Blender is going to be? I understand the philosophy of making tools whose purpose is not to offer a delivery platform but to give artists tools to play with acording to their creativity. Still it would be good to know the shape, the personality – the power ? – of Futur Blender.

  106. I think Blender needs a full rewrite, and code reorganization. Perhaps to be funded on something like Kickstarter or Indiegogo. Allowing more developers to be hired on and work on this rewrite, while some developers remain on the 2.6x and 2.7x codebase to fix issues, and continue support.

  107. Hi !

    Like Emile and CONSTANTIN, I think Alembic is the most important feature for blender in the futur !

  108. When Blender support AMD cards?

  109. I wish they made blender gui working on android, because tablets and phones are now very capable of running it, (now I don’t Have a PC, just a Tablet) i tested last year “incomplete” port and it worked very well, it even run quite complex games with lights and normal maps.

    Nearly everybody could make 3D interactive apps just using a tablet/phone, no more big PC :). I think the blender GE has a brilliant future.

  110. Ton, et al,

    Ton: Thanks so much for your work on Blender. These are very forward-thinking ideas, and strong management direction: it is GOOD to get rid of duplicate/crufty code.

    I like to use Blender as a hobbyist, as a “simulation setup tool” or “walkthrough tool” so that I can model and light a perfect shot, then render a single beautiful still image.

    When you say “Interactive Mode” you are speaking to MY use of Blender Game Engine: as a pre-visualisation tool with some “logic” in it so I can setup and then RUN my scene
    while I WALKTHROUGH it.

    Perhaps “Blender World Authoring” is a better name for the new gaming roadmap; it implies the that there may be some different “World Player” out there in the universe (e.g. Unity, Torque, BGE, XYZEngine) to which Blender can send 3D scene and logic data via a plugin.

    This leaves room for “distraught” indies/volunteers who wish to contribute a clean-room implementation of a “Blender World Player” i.e. a new open-source project with a commercialization-friendly license e.g. MIT License.

    Long live the BWP!

  111. Yeah, don’t wait, I wanna see OpenGL 3.0 as soon as possible.

    ¡Good Job!

  112. I think if BGE gets integrated into interactive mode, Blender’s whole “idea” is done for and it’s going to take renderes with it. I stopped using Lightwave specifically because BGE 1) had a linux option adn 2) all of it was in one package and 3) it was open source so the community can say “hey we need this”, instead of the stock holders. If you remove the hope that BGE can be better than “eh”, why should I stick with Blender as a rendering tool? I’ll just go back to something that is more common (Maya, Lightwave, SGI, etc). I can’t be the only one out there that feels this way. Plus if GPL is your concern on the BGE, shouldn’t it be the same concern on the simulation render outputs as well?

    There are plenty of people getting past the GPL problems by wrapping the blend file so it can’t be “GPL’d”. Heck there is even a post on how to do it at (which seems to THE go to spot for BGE).

    I have learned Python 3 because of BGE. I have stopped using my paid for Lightwave 3D for Blender’s “goal” because people using a product is how things get changed because demand goes up. I’ve recommended it to others even. There are a lot of people putting a lot of hard work into the BGE. If it end’s up being “legacy”, you pretty much spit in their face. Either branch it off so the commmunity that wants it can use it or just admit that it’ll take some hard work to get it where you’d like it to be.

    If you branch it there should be a join API for both platforms called “Blender connect” or something…

  113. I stand behind this 100%, Ton is really peeking far into the future with this one.
    Specialized tools, workarounds and hacks are slowly fading away and being replaced by pure simulation as computing power increases.
    The whole industry is shifting towards this direction, not only the movie production industry but also the gaming industry.

    Real-time path-tracing game engines like Brigade are without a doubt the future for the years to come (ref. John Carmacks recent speech at QuakeCon), and will lead to DirectX inevitable demise.

    “Offline” rendering/production and realtime will converge and everything still stuck in “Offline” will become obsolute.

    With Blender becoming more and more interactive, It’s inevitable that Cycles’s future will share alot with Brigade

  114. I think I like the ideas about GE.

    Maybe it can be the most easy to use GE to use for publishing on smaller devices and html5? So it shines on OpenGL ES?

  115. It’s a terrible idea to drop BGE runtime and fold it into an interactive mode. The integrated animation/modeling/game-engine workflow is the strongest pull of Blender to game developers. Otherwise why not use Max or Maya that are more standard ?

    The so proposed “interactive mode” is good for the film/animation projects and it’s already there. But it is DIFFERENT from BGE. The file/animation is ultimately offline rendering art form. While games stress on real-time and lean deployment. So making BGE multi-threaded, higher real-time performance and cross platform deployable is the right direction to go. Abandoning the current BGE runtime as an easy delivery option is a bad news to game developers.

    The only choice left for those who like BGE then is to branch or help alternative game engine GameKit to mature. Maybe GameKit should be made the official Blender Game Engine if the current BGE is abandoned.

  116. Another approach with BGE could be the same way as of cycles: make blender open with there node system for other game engines, like octane render embedded his engine now in blender, but keep developing GE. Then GE is only one of other game engines.

    – you keep the blender design flow to create your game content and game logic
    – offer an additional node system for GE: for logic parts etc. what can use and extend other game engines as well
    – fully integration of other 3rd partie game engines. This would be the system what now exists for renderers in blender (octane is now embedded!).
    – game engines are fully integrated in “interactive modus” what only needs to transfer the perpared stuff and logic to the selected GE. The data format should be a standard if something already exists world wide.

    First step would be to transfer current GE to an open node system. Therefore other game engines can create there own nodes to be flexible but fully integrated in the game creation process. This would be a dream. So the interactive approach could be developed for this. Then you have one face to many games engines!!!

    I could imagine that then other game engines would be integrated in blender. Next advantage is that GE developer can focus there effort on game engine performance aspect because they get a nice tool to integrate it. If i would be a GE developer i would be happy about it.

    What do you think about this suggestion?

    (btw. pls sorry, my english is terrible…;-)

  117. I did not read a single post but my guess is most game devs, like myself are pleased with making the game engine more integrated and furious that it will not be developed as a game engine…it is better to make a game engine that can do arch-vis walk-throughs, than to make an interactive system that can “attempt” to make a 3d game….one works both ways one doesn’t…I am very displeased with this aspect…

  118. You may consider having a chat with LEADWERKS devs. Their engine is not expensive and is being ported to linux. It already runs in windows and mac

    • “Blender integration: We want to integrate Leadwerks with the free 3D modeling package Blender. We’ll start with a Blender exporter that saves a model and all materials ready-to-use in Leadwerks, and look for other Blender features we can put to work in our engine.”

  119. I would like to thank everybody who works in Blender and makes it possible. It’s a great tool.

    Recently migrated from other 3D software to Blender, and I’m really happy with that decission. As a newbie game developer and also newbie in Blender I feel that Blender need to focus more in 3D production, game asset production, etc., but not to try to compete with commercial engines like Unity or Udk (I see that Blender is a great tool to work with these engines and many others, with better compatibility than many other commercial, and expensive, 3D software).

    If I have to choose I hope that the developers will focus their efforts in making Blender a better 3D assets production tool than improving the internal Game Engine.

  120. better playback for animators in the 3d viewport! Maya is kicking your ass on that front.
    Also f-curves die too much.

  121. I think that a very useful thing, while you guys reorganize the data storage, is to add a standalone C/C++ library for parsing data from blend files (even partially). And also to keep this library synced with each release. Perhaps it can also be used internally by blender.

    This will be most useful to integrate blender in asset pipelines of all kinds.

    Obviously one can write an exporter for a custom format or use existing formats. But then it has to be maintained constantly and, lets face it, Python is not the ideal tool to preprocess even the tiniest amount of data, thus requiring another intermediate format.

    With a library to parse .blend files, instead, one just parse them at the beginning of the asset pipeline without intermediate formats and everyone is happier.

  122. I would appriciate the nodification of more parts of blender. Maybe also animation and modelling, while there could be both systems at one time.

  123. I think that integrating the BGE more within Blender is a great focus.

    It would benefit BGE as an end solution, and benefit blender itself, integrating Interactive and Real Time use cases with the power of a dedicated 3d tool.

    Think architecture or movie previz… a dedicated interactive machinima player/TOOL

    game prototype development from within the 3d authoring package, with deep export ot unity / unreal / source / whatever

    What i like to add to the discussion however is next to RealTime or Interactive is also the upcoming paradigma of Immersive…

    That is VR is coming back quickly, and this time it works… ! oh the memmory of the 90’s….

    ie, what will be Blender’s position when VR modeling is around the corner?

    I’m not saying that this will replace traditional 2d/mouse digital content creation, but havind a RT Interactive mode of Blender Integrated… would allow developers/users to explore this.

    ie, you are modeling for a move, do some previz of a shot or environment… the get your rift on and explore … no experience much more than can be otherwise….

    SO thumbs up for an Immersive Interactive RT Blender !

  124. I want to see “OpenGL rendering” come fully into its own as, shall we say, “a legitimate source of rendering.” We should have a set of “OpenGL-driven nodes” in our kit, so that if we want to use GPU hardware for the purpose it was originally intended for, but apply it to a traditional frame-by-frame animation (or a single frame), it should be not only possible but easy to do that.

    For instance: “in my node-based pipeline, I want to start with an OpenGL-rendered base image, because I can get it fast, and then I want to (in a separate render-layer) also generate highlights using traditional BI (thus spending time-and-money only on those highlights), and merge them together such that the final output is a hybrid of both techniques being applied at once.”

    GPUs are powerful and are always becoming more-so. It’s exciting to see how Cycles is using them for “non-traditional purposes,” but I want to use them for the purpose for which they were intended: generating high-quality images very fast. I want to do this without being constrained to do the whole finished-image that way.

  125. CAD functionalities for precision modelling and Architectural Visualization, snapping (edge, centre), rulers along axis, polylines…etc. I assume these functionalities are minor.

    There is a huge community out there who do and wants to do Architectural Visualization, but are not using Blender because of the lack of these basic functionalities.

  126. As an amateur artist and game developer, I can tell you that Blender is much, much nicer to use than most game engines, but it lacks a lot of their functionality. As I’m looking at the issue, it seems that a primary cause of concern about making games in Blender is the GPL licensing. While I agree that this seems to “cripple” the Blender Game Engine, I’m also not sure that BGE would take off by itself without major development even were that not an issue. In my experience, BGE feels like a less-focused-on part of Blender, presumably because most of Blender’s users are more interested in other pipelines such as animations for films, or use external game engines. Rather than trying to “prop-up” the current game engine, I think it might well be wiser to scrap it and implement an alternative solution – just let me explain before you judge.

    I know a ton of work has gone into the current game engine (and I’m not proposing that we scrap that work, if you’ll read on), but the fact remains that it simply isn’t up to par with any major 3D game engine on the market. While it’s true that those engines aren’t free, they’re still the industry standard, and the current form of BGE is never going to get there until some issues are cleared up with GPL licensing. Even so, it would need a major project dedicated to it in order to get up to speed before a new game engine release outstripped it. Due to the nature of the industry, it is likely that BGE would be playing almost continuous catch-up for an indefinate period that could last years or even as much as a decade – all the while never gaining traction. Clinging to the current solution and trying to make it fit as a full-blown “game engine” seems like it would be a shot in the dark.

    As an alternative, I propose that we attempt to integrate the good parts of the game engine into Blender, like many others have proposed, and then drop support for it, allowing it to fall into the dustbin of first drafts. Like the BI rendering engine, BGE should be gradually phased out, and more and more of its “good” parts incorporated into a new, improved, and unified “Physics Engine” which would become the focus of a major project. This “Physics Engine” would incorporate the current physics of Blender, such as fabric simulation, smoke simulation, particle systems, collisions, and the like, and would also include optimizations originally intended for the game engine, as well as any new code that might be written to handle other types of physics or improve those that exist. Similarly to Cycles, the improved physics engine would focus on realistic physics for simulation, and would include certain features such as the ability to give solids a physical material with properties like breaking strength, density, and the like, incorporating a node-based system for physics that would allow behavior to be simulated with less “faking” – as an example, breakable objects could be shattered in response to collisions rather than having to be fractured by hand. Furthermore, the integration of an advanced physics system into Blender seems to me as though it would perfectly complement the integration of the Cycles rendering engine – physics-based rendering improved realism immensely, and so will physics-based materials and simulations.

    By transforming the “game engine” into a “physics engine” and improving physical simulations in Blender to take advantage of material properties and high-level real-world physics, much like Cycles realistically simulates the properties of light in order to produce a superior result, Blender would be transformed in a way that would make it a more unified and useful software. Instead of starting from the “end goal” of a game engine, Blender would be brought to a “solid foundation” featuring the internal physics that both simulation software and real-time game engines need. By my estimate, based upon the way in which Blender is currently progressing, it would not be overly optimistic to say that such a state of unified cohesion could be reached in roughly six months. Such an improvement would serve as a launching-pad for future improvements, which I can see including things such as PhysX optimization and game-engine like features such as building and baking parts of scenes to speed up render times and iteration.

    As to the GPL issue, I believe the simplest solution is for the Blender Foundation to release the Physics Engine under an alternative license. I have thoroughly examined the GNU GPL v3.0 and the license agreement for Blender 2.69, a process which took me more than four hours, and I have come to the conclusion that the Blender Foundation has every legal right to do so. The following text is taken from the license agreement.

    This License Agreement for the Blender software (“License
    Agreement”) is an agreement between Stichting Blender Foundation,
    Amsterdam, the Netherlands (“BF”) and you (either an individual or a
    legal entity) (“You”) with respect to the software product which
    this License Agreement accompanies (the “Software”).

    Based on the above excerpt of the agreement, it is very clear to me that it is the Stichting Blender Foundation that is granting me rights to use the software, and not the original contributors. This seems sensible because, just as Microsoft Word belongs to no one author, but to Microsoft itself, Blender should likewise belong to no one individual, but to the Blender Foundation which commissioned the project to begin with. This is as it should be: no one individual should exert the control of ownership over the software or any part of it. It is my understanding that the project that is Blender did originally belong, does still belong, and never stopped belonging for any period, indisputably to the Blender Foundation, which licenses it under the GNU GPL to the Blender Community. The code belonged originally to the Blender foundation; this is certain. Subsequently, it was revised, updated, improved upon, and added to by members of the community to whom it was licensed by the Blender Foundation under the GNU GPL; however, while the Blender Foundation licensed Blender to the Community, the process was never completed in reverse. At no point was it specified by the individual contributors that the Blender Foundation was the recipient of a similar license granting it the right to use any version of the software belonging to any of said contributors. Rather, all such contributions to the Blender project appear to have been freely given, with no expectation of retaining the copyright on said contributions. Just as a gift freely given no longer belongs to the giver but to the recipient, so these contributions to the Blender source code belong no longer to their authors but to the Blender Foundation. Even were the original authors dissatisfied with the Blender Foundation’s usage of their code, they would have no legal right to demand its return, having freely relinquished their rights to the Foundation. Once property is transferred, it belongs exclusively to the new owner. Thus, it is entirely the decision of the Blender Foundation as to how it wishes to license Blender, and/or any component part thereof.

  127. I hope for a christmass easteregg, to turn a series of photos of an object into 3D blender object. So we can more easily adjust reality.

    If i can get you the math you’ll get it, but its still my holy grail as for now 🙂 (i’m into software too C++)

  128. Turkish language support is not only good to have blederde

  129. Actually i’m working at home with Cinema 4D R10 (software made in 2007 year!!!) and in my company with 3D MAx. All these software are very similar and it’s easy to move workflow from one software to other, because both softwares has the same options (extrude, smooth, inset, weld, chamfer ,bevel, bridge, fillet for curves, many parametric curves like rectangle with round corners). User in C4D and 3D Max can make model precise (for example fillet for curves – i can use 5 cm in spline corners).
    I think blender is great software but in 2013 year not all basic features (like in other 3d modelling packages) are available in blender:
    -) working with fillet in 2D curves (splines) doesn’t exist
    -) good working bevel (not modifier) option is not working with all geometry properly (i checked it many times)
    -) better and more easy to use simple materials. For example gradients.actually i’m not able to make circle light emitter (CYCLES) with circle gradient texture (circle gradient used to emulate power light on edges emitter) to simulate halo object,
    -) mirror modifier is broken for me – i have to remove parts of geometry not used by mirror (behind mirror edge) to get symmetry work properly,
    -) in outliner – user should see relations between objects (parent, child, group) like in Cinema 4D objects manager (if 2 objects are in group than is created new NULL object titled GROUP and than 2 objects are INSIDE this null).
    -) too many features i have to simulate “by nodes” or with scripts which are not working at all.
    And many, many others…We are in 2013 year, and so many basic features are not working good or are missing. Get this stuff working good in 2.70 and more people will try to use blender.
    If Blender want to be “ready to use” in first minute after install than user can’t spend 20 hours to looking missing features with forums and scripts…
    Summary – last 3 weeks i’m trying to move my workflow from C4D and 3D Max and it’s horrible…My industry GUI habits and standards are useless in blender, I have to learn almost everything from scratch – only vertex, edges and faces are still the same.
    But i promise myself to not give up.I want to win this battle (yes, it’s battle) with blender. There is no software ready to beat me up 🙂
    Kind regards

  130. Ouu and i forgot about something very important – volumetric lights…why ONLY spot light has option to render volumetric light and volumetric shadow? I want to use other lights (point, parallel spot, rectangle emitter or area light) to make some volumetric effect. In 2007 year it was so simple in Cinema 4D…it can’t be so “hard to code” if it was possible 6 years ago in other software. I don’t want to simulate everything in “compositing” mode (and make tens layers and passes, or like in cycles render – first render volumetric with blender internal and than render the same with cycles and than connect both renders). I believe that so many amazing people are still working to develop blender, and some group of people will find time in next half year or one year, to try put blender to the same level (basic features available in other 3d soft) like C4D was in 2007 year.
    Summary – i’m so excited that community of blender is working so hard for many,many years and still has power to make next, better and better versions of blender:) I want to say thank You. Thanks for all open movie projects which are milestones for blender. Thanks for all great features like fluids, smoke, compositing, cycles,uv mapping, sculpting, bmesh, great armature and animation system. And now – i want to a bit more industry standards in features which are stable, ready to use and no more scripts needed from internet for basic functionality.

  131. I think the game engine should get more attention, as 6dof joints controlling armatures the programmers will be able to input muscle strengths, etc, and use “force maps” or “Torque maps”

    Animations are nice, if you want the exact same thing to happen every time,

    In the real world, my spinning back kick can be countered with a shotgun blast to my face,

    Animating every possible contingency – not possible

    Animating using overlapping forces to do anything = totally possible.

    Bullet 3.0 will have GPU based 6dof solvers, as well as the eventual integration of the featherstone solver algorithm,

    Add in a ability to attempt to match a actions animations, using force and torque, and everything could be done in engine,

  132. Hello )

    BGE give great possibilities! It’s very bad that BGE falls!!!
    possibilities – science visualisation ! Python + numpy + scipy + serialport + pyside + BGE = Nice standalone progect.
    BGE + python is fast and convinient developping process.
    For example I write some useful medicine visualisation of new science metodic by Python + BGE and I can make good program )
    Other game engine can’t do THIS ! Python with science libraries + and 3d rendering/modelling tool work as Unity program! This is unique offer of Blender!
    About GPL. There is a good example QT library. If developer want write GPL free program he use GPL3 version of QT library. But if developer want write commercial program for sell, he buy commercial QT library licence ) One product – qt libary has two licence one GPL and one Commercial.
    Why BGE developpers can’t act in the same manner? BGE convinient insrument with unique offer – python with his science powerfull libraries and very good visualisation engine in one box )
    Honestly – when I have good result and work program what I can sell I buy Commercial licence for BGE for 200 – 500 $ )
    People can sell their science programs and Blender found can resive money for commersial license ) It’s very convinientdevelopers can try BGE and work their applications and them if they want sell their programs they will obtain commersial licence )
    I think that BGE will intresting tool for scientists Python and BGE very comfortable and powerfull tool ) For Python that very popular in science there is no aternative…
    Interactive Blender with python scripts is good but something like maya.. without Maya on the other computer You can’t do anything….
    It was uniqe offer from Blender standalone )
    I think that if BGE support Pyside or PyQt embeding and embeding popular python libaries it were great tool for scientist and tool for Fast implementation of science work in real life! witout aniting ))

  133. Thank You for your great job on Blender.

    It is not a good idea to drop BGE. Most people do not use blender to make high-end fotorealistic pictures, most people comming from their game-console and wonna create their own game charakter and super-duper-hero. And then they wonna see the new hero runnig. If you drop BGE and change in “Interactive mode” there is no hero. No joy, no fun. If you remove BGE, you remove fun from blender, things going bored. This is surley the wrong way.

    Nobody has said, that BGE has to be as good as Cryengine3, but it could be a little bit better. As example: If i export the standard-scene (cube) as exe, then i get a 74BM exe folder. This is a little bit much for 8 Vertices. So my suggestion for the next numbers ist: from 2.7 to 2.8 only improve Game Engine, and then with Blender you have a super tool on the market: It is a excellent fotorealistic renderer, a game-content editor, and with a single click it is also a game engine to have fun. There is no other Tool who can do the same.

    I know it is a lot of work to create a game engine, but why not integrate ogre or irrlicht in blender? irrlicht with its support for opengl, directx or the sensationell apfelbaum-software-renderer are standing alone and i have the feeling, this pice software is looking for a home :-).

    So, this are my 2 cent.

    Best regards.

  134. I don’t want blender to move to MIT/BSD etc. They offer no patent, copyright withdrawal and copyleft protection.
    Blender should become gaming engine and GPL is not hindering it. The only one questioning this are proprietary developers. If blender shifts even partially, we will see more proprietary projects who just use blender’s inside and less contributions. If we don’t shift, the amount of proprietary will not reduce. The majority of top commentors criticizing gpl hardly ever read the license. GPL is applied only to source code and specifically insists on free speech, not free beer aspect. Is it hard to write GPL code of a game and distribute it for money? I have a tons of games done via Unity on Android, all loaded with Ads and crap. I miss Linux apps on Android :(, the whole culture is CUT. Blender can easily reach cryengine levels, but if it refuses in the first place, he will not, never.

    I don’t understand why do some people contribute to gpl, while misunderstanding, hating it, posting comments how bad it is and how good closed source platforms are (that ban gpl, ban freedom of choice). They make me remember the snake from the bible, and the punishment is a sure way if one hears to them. Don’t like GPL – go write your Unity, closed, platform limited, feature cut, owned by NOTyou. 🙁

    • totally agree! GPL is to preserve user’s freedom is to put free software developers in advantage above proprietary ones (they can see the code, improve, use often without worry about what can be done and what is forbidden by the license, etc). In the long term, it preserves user’s freedom.
      User’s freedom that GPL defends is not “I do whatever I want”, but “I can do whatever does not hurt/limit the freedom of other users”, that’s crucial.
      I would not happily contribute to Blender if I know my code would be “stolen” and incorporated in proprietary products (so they have the advantage of my “free” work + their work). We have to defend our freedom.

  135. I’m a game developer and I agree with your vision for the future of Blender Game Engine. Now it’s to late to create a real and professional game engine. Indie devs works now with other tools.
    But Blender do need a interaction mode, event for simple demos and to test some behaviours.

  136. My greatest concern regarding the BGE is game logic(path finding, logic bricks, python scripts). Will game logic still exist in future blender releases? I have no problem with off-loading rendering to something like Ogre, indeed those projects probably do a better job. But BGE is more than just a render engine, its a render engine with logic, and that is what makes a game engine. What other opesource game engines are available right now? I only know of GameKit but its game logic portions are far less developed than BGE(I tried making a game in both). If there are no good opensource game engines out there that can be integrated with blender then by dropping BGE as a game engine something valuable would be missing from the opensource world. So my suggestion is to move game rendering out of blender and into something like Gamekit while keeping the logic.

  137. and about fix the pole target bug?
    impossible to finish my stretch legs and arms rig without pole target.
    with ik trick doesn’t work.

  138. I am really not sure what the changes of BGE means.
    Will it be possible to make a game with blender or will the standalone player and logic vanish or not be developed

    any further.

    For my part I liked BGE. Sure here and there could be some more features.
    But after all there can be made nice results in a super fast way.
    BGE was the reason I learned Python and at least one of the reasons I moved to blender.

    I always hoped that some day some main bugs get resolved and some features

    like spot light shadows and improved and memory management would be added.

    Now the BGE or at least the way to make a game becomes removed?
    Are you sure that this is not an idea of a developer for other commercial game engines?

    The only problem is that there are to few people knowing about the existence of the BGE, they don’t know what it

    can do and how fast and easy.
    Maybe they don’t know how to use it a way they can licence a game under their terms or just how to distribute it

    in the correct way. Clarify the problems and solutions, make this things better known and you’ll be fine.
    But please don’t let the GE fall apart.

    Maybe you are suffering a ‘marketing’ problem not an problem with the GE.

    However I like to see it developing.

  139. I suppose I’ll throw in my 2 cents as a developer and a Blender enthusiast. (this is regarding the direction of BGE)

    IMO, having BGE as a separate component doesn’t make much sense from a project structure point of view. When I first dove into the code I was surprised that there was a separate source folder just for BGE. There has got to be some dupe code there that is a pain to maintain. So, from that point of view I think its a great idea to integrate the interactivity into blender itself.

    Regarding the GPL, it is a problem if you want people to use BGE in a commercial product. If you don’t, then it isn’t an issue. However, I would hope that you consider that many developers that use Blender over Maya/3DS/etc are those who may also be into indie development. Having a Game Engine for them in Blender would naturally draw more users to Blender for modeling. I understand it would be difficult to contact all those required to change the license, but IMO it would be worth it.

    If that isn’t possible, I think it would be in our best interest to put some effort into an interchangeable format for the new interactivity so that a MIT blender interpreter (like gamekit) can have greater success.

    I’m one indie developer that looks for every reason to use BGE. I know that it will require more programming than Unity (for adding features and fixing bugs) and I know I should really focus more on the game than the engine BUT I can’t develop a game I plan to sell on a platform that, by license, restricts me from integrating and selling on the most lucrative platforms (Steam and iOS)

    If we are being humble about the ability of the engine, then we should also be humble about this GPL issue…

  140. NURBS!!!!
    We need NURBS python API at least

  141. we need a better GE, I would like if BGE were more like Unreal Engine and less like Blender. 🙁 sry blender, but u know im right.

  142. Is there any plan for vertex normal control, and transfer info like maya, this stuff is really important for high level game art.

  143. Split or share GPU memory with System memory in cycles? I think it would be the best implementation if possible, I very often overcome the Vram and then I have to go back to CPU rendering.

  144. I have been making maritime historical basic routes in the Blender Game Engine for a few years,described with descriptive text on my site. Mainly 12 mile terrain mesh but have tried 30 miles and 50 miles square Isle of Man and Isles of Skye.

    The problem is if altering vertices to create depths they ‘fly back’ to within 10,000 metres of xyz line..but if mouse used and not typed in they stay put.

    Anything that helps ‘further the cause of making basic coastal maritime routes appreciated’ So far made tides or sea swell as animated planes,driveable floating schooners and barges, etc, and animated solid boats moving on preset courses for up to an hour or two. Docks are up to 3 miles c1833 Liverpool and also working locks.


  145. I am really looking forward to the Sequencer rewrite. I already use it for Movie cutting. It would be great if one could use the compositing effects on the tracks directly, e.g. different green screen backgrounds for different movie tracks.

  146. Hi, Sculping please push it to new Limits.
    There is no Sculping feature on the list ..
    Hope 3 d viewport rewrite help to improve the vertex count ..blender can handle only 1/10 of the z brush can .
    Sculpings needs curve brush to pull out or in Sharp geometric Polygons faces in both modes , multi res. and dynamo topo.
    Its so important, blender can not creat a out pulled Polygon face only with a texture ,lol
    That bring’s no good workflow and bolean with primitiv Even the same horror.
    The model Elite of model sculp hard surface programs show really what can happen with good Tools.
    3D coat and z brush show whats possible .
    Voxel sculping is the best think I ever used step back to blender no change but on time. Blender will have a good sculping and modelling that is streamlined for good Workflow .
    I pray for that..

  147. An open-source alternative to MicroSoft’s ‘Project Spark’ would be very welcome.

  148. I’ve lost all faith in Blender as they have not supported Alembic Import properly yet (despite the obvious need and screams from the users).

    My personal need is to use Cycles from Houdini. Once I have a good GPU render pipeline for Houdini I will no longer care.

  149. I think people are forgetting if you don’t like a new version YOU CAN STILL RUN THE OLD VERSION. If people use the old version for BGE great. If BIM (Blender Interactive Mode) gives more tools, with more power and more pretties, then you’ll BGE users migrate to the new version. I care about scripting in Python and the ability to use shaders w/ newer technologies (OpenGL 3.x+ as 2.1 is part of their “Legacy” branch now). Rip it out and do it right the first time. (well 2nd but who’s counting). I see the desire to update it but a lot of these posts are fear of losing the ability to do X. Plus game engines are generally out of the realm of super indie developers budgets and not everyone likes Windows to develop on. I understand the focus isn’t on publishing and if there is a desire to publish content on Android (there is already A LOT) then someone else can step up and tackle that issue. All this hoopla about “well what about the old way of doing things” is non-sense becasue old versions are availble to download all the time. If you like the old way of doing things, go run the old version. If you want some feature that only a new version has… guess what…

  150. Blender has been in my life since I was twelve years old; quite honestly the best nine years of my life spent making games and animations, connecting with this huge network of open source users. Not only is this cool, but it’s free to the public, “for us to express our inner artist,” along with math, algebra, physics, logic, science, and engineering.
    Forget all doubts of this program; blender 3d is the best 3d program for all of the above.

  151. In the near future, I’d like to see the developers devote a cycle or two to cleaning up some of the annoying features or lack of features in the existing release.

    Most of these I think are like very minor but would make life easier, especially for beginners.

    For example, why are there so many steps needed just to get textures activated in Texture Paint and Sculpt mode? Instead, why not add and delete brush textures directly from their respective panels.

    I’d like the ability to create seamless brushes in Texture Paint mode. Automatically offsetting a texture image, cloning the seams, desaturating, adjusting the curve, and perhaps a high pass filter would make life easier pretty much eliminating the need to constantly go to Gimp or Photoshop to create brushes. And while you are at it, give me the capability to invert the colors on brushes. I often find it necessary to convert black and white brushes by inverting the colors. This would avoid having to go to Gimp or Photoshop just to invert b/w images.

    I’d like to be able to delete materials and textures from the cache without having to close and restart blender. At one time there was an add on that did this but as Blender has been updated, it no longer works.

    I’d like to be able to see the full file names of images in the Display Mode instead of having to click on the image and look up at the image name in the File Name input.

    I’d like to be able to tweak texture brushes on the fly for contrast, hue, color, transparency, specularity, etc.

    In the User Preferences I’d like to be able to set paths for more control. For example, I’d like to designate a separate folder for add ons not bundled with each new Blender release so Blender would by default look to the scripts folder for included add ons and also to my separate designated folder for other add ons.

    I’d like to set a path for brushes I use in Blender.

    I like the tabs but I have no control over what functions they contain. It would be a significant boost to my workflow if I could move functions to custom tabs in a way that makes sense and only for those functions I need. As it is, developers are scattering their functions in new tabs, functions that I’d like grouped are scatter in other tabs, and functions appear in tabs that make no sense. Some functions appear in almost every tab. I know that the way Python scripts are organized for Blender this would not be easy, but surely someone knows how to tame the tabs and put them in control of the users the way they want to organize them.

    I’d like to be able to save the organization of panels while I’m working instead of having to make all the changes in the default scene in order to save them how I want. I would like to save the panel structures the same way I can save my user preferences while I’m working.

    I could go on, but hopefully you get the idea. There are literally dozens of these relatively small annoyances that if fixed would be a significant improvement in managing workflows and saving time.

  152. i like the idea of the BGE becoming part of blender 3d, this means that features only have to implemented once and it works instead of implementing things twice, instead i have no idea why it was made separate in the first place, but anyways just because it is being merged with blender 3d doesn’t mean that it can stop being a game engine, it doesn’t mean it has to loose the focus of the current BGE and become something different, it can stay the same, all that would change is the core design, meaning there would only be one instead of two separate cores (one for blender 3d and one for BGE), in fact the user experience would not change on the surface because although the BGE is currently separate, the fact that it is separate is kept from the user on the surface and you can’t tell from an artists point of view, you can only tell if you look at the overall architecture of Blender 3d and the BGE as a whole, so merging the BGE would have no effect on the user aspect of using the BGE if it was merged because we would be none the wiser unless we checked the architecture, and an artist is really only interested in one thing, actually being able to use the programme, most artists aren’t actually interested on how it works just as long as it does work.

    what amazes me is that they said that they were not able to make a game engine with the capability of something like cry engine or unity 3d, but yet they have hardly touched the game engine with the way plus with very few people using it because it is GPL, but here is the thing, there exists many game engines out there no where near the capability of cry engine with lots of users game engines like RPG Maker and so on, esssentially regardless if the engine is not the best in the world, that doesn’t mean that no one is going to use it, people still use rig make and engines like that even thjpugh they are technically shit in comparison to something like cry engine.
    what they should do is upgrade the foundations of the game engine, upgrading the foundations of the game engine will make a strongger game engine overall, it may not be pretty and have good graphics but everyone would prefer i am sure (a wooden building built on solid ground over a brick building built on marsh land lol).

    if i had any say in the matter of which i don’t but i do think this would help, i would integrate the blender game engine with blender 3d like they are doing but i would keep it focused as a game engine, i would then go and build a solid 2d pipeline, like what unity has just done, but i would start of with that, starting of with a solid 2d pipeline creates a solid foundation of which to build upon, it could even be integrated with the 2d compositor, and the UV/image editor could be improved as what you use to create the 2d vector graphics and so on, it could also be the same system you use to create 2D GUI’s for the games, this would create a solid foundation from which people would be able to author amazing 2D games.

    i would also have a new blender play authored for the BGE released under the MIT license, no not the BGE or the blender 3d, just the standalone player, but instead a new standalone player made released under the MIT licence so that standalone games produced would not inherit the GPL licence, and also since the new licence would GPL, this would also allow people it encrypt their standalone games allowing them more privacy and control and ownership over the games that they produce.

    but anyways as usual, i don’t know what i am talking about, i get that, so i will now keep ideas to myself.

  153. I’m all for the game engine. I have used Cryengine and Unity and the creative freedom you get with blender – in terms of game design – and how easy it is to create a prototype in just a few days/weeks, just can not be compared. I didn’t know ANYTHING about making games about 3 months ago. Now I’m coding in python like a beast, and how powerful it is amazes me every day.

    I genuinely LOVE the Game Engine. You can do basically ANYTHING in it if you put your mind to it, and aren’t lazy do do it all yourself. But at least if you DO do it all yourself, your game won’t be screaming “CRYENGINE” or “UNREAL” straight when someone looks at it.

    The ONLY downside to the ge (for me at least) is the performance/rendering. Seriously, that’s the only thing. I don’t really care for web support and all.

    Python is more than capable of creating complex gameplay. You have a full on animation system in blender with the ability to model and texture, without having to worry about importing your models/textures/animations into another engine.

    I would actually go as far as to say that if the developers worked on the GE’s performance a little, to allow a faster, nicer rendering, I’d be willing to pay royalties after a game’s release to blender. Because I think the engine, and how easy it is to use is unbelievable.

    So many people have created so much amazing things with the engine, just to prove how good it can be if you know what you’re doing. I’d be the happiest person to see improvement, but the only thing that really needs working on (I think) is the performance. A quicker, smoother gameplay with shorter loading times, and less lag, so that you can create nice looking games but not only for pc’s worth thousands…

    I’ve learnt so much already about game logic it’s unreal, and I’m looking to learn more and more in the future, and without the BGE, that’s wouldn’t happen…


  154. All current Macs that don’t use Intel GPUs have AMD/ATI ones. The current Mac Pro has some seriously meaty GPUs on board, and all that computing power is inaccessible to Blender because it doesn’t have working OpenCL support. If anyone who knows this stuff can look at the OpenCL situation again (or if someone with influence at AMD/ATI can poke their driver team) I’d be really grateful.

  155. WE are currently working on the compositor so has anyone thought about adding audio support? This would be a great project a could lead to audio composting

  156. We make Images, Animations, Simulations and Interactivity, all using the generously free Blender. The internet is the method of delivering our creations to all corners of the earth. The browser is the universal gallery window, regardless of device or OS.

    If Blender could output Interactive 3D for the browser, then we would have a complete set of tools, reaching from the point of creation right through to the delivery destination.

    “Real artists ship” Steve Jobs

  157. Later that day, we visit the Vape Lab , a lounge that launched this spring in Shoreditch, east London.

  158. The main brake BGE is licensed.

    If that were resolved, many other vital problems could be solved with a roadmap. For example, export for touch devices, update logic bricks, etc.

    As for being at the height of the great engines of video games, I think that should not worry. Not all games are “Crysis” not all want to do “Crysis”. Even more now that live in the physical and virtual shelves games like “Call of Duty” or “GTA5” and “World of go” or “Limbo”.

    I agree that BGE has survived because there is nothing like now. It is a rarity that makes me very happy :), and why I think we must fight.

    I think you could try to make one last effort to try to resolve the issue of the license. Just as when he fought for the liberation of Blender.

    I’m not saying the problem of the license is simple. But to understand all the magnitude of that problem, it could translate into a budget. Nobody better than Blender Foundation to do so.

    With a budget would be put on the table the amount of work needed to solve the problem of BGE license. And like Blender’s license at the time it was released, it could provide an opportunity to support this project. I speak eg kicktarter.

    I think it is nonsense, I myself would work generously for that to happen. I am very interested in using the BGE for commercial purposes because my passion is video games and what Blender is, beyond being a 3D program.

    So regarding licensing issues BGE, I have only one question: how much does it cost to solve this problem?

  159. As a casual user i might not hold any weight but the thought of game engine abandonment scares me. My perception is if licencing is not an issue (you want the privilage of making your own game but sales is not strictly important) it should be possible to make one as sophisticated as possible so long as you can supliment with a bit of python knowlege to make something that can rivel anything commercial in terms of quality and capability (with accecptible hit in speed for heavy games).

    If this feature is lost its bound to dived users who will simply have to hold on to the older versions to make games; but ultimatly will loose out as the versions become depritiated to hold up. (and piss people off).
    Up till now it seemed blender was (it IS now more or less) geared to being the ultimate ‘do everything’ tool for film/modeling/rendering/simulating/games. Feature wise it trumps all comercial 3d apps because of this and given the announcement of reactor particles return (yay), it will really suck to say goodbye to a feature because it has licence issues.
    If content created by the user is considered there own i don’t see the problem- some people might still sell games and could be revinue or credited to blender. But, if the game engine is reduced to a simulation tool ONLY, with no export then there will be a huge loss in what this softWare has achieved over the course of its evolution. Blender should grow as it has- by addressing: “wouldn’t it be awsome if you could do ‘this’, in blender! “, issues. NOT “if you can’t sell it, dump it”. That would simply suck monkey balls.
    I’m just getting into the bge and hope for a long safe road ahead.
    Imagine what kids could create for themselves given the rise in education for coding and python plateforms. BLENDER IS THE NEW LEGO.

  160. Renderman ,the render engine of PIXAR is now free for a non commercial use. In the future will it be possible that blender support this renderer ? ( sorry for my English 😉 )

  161. I think it’s all going in the right direction. It would be głód to also think about handling mocap animations (recording inside blender, transfering between rigs, removing feet slide and jitter and so on). There ar addons, but none od them works well.

  162. Tho I hate to see the BGE break, trunk integration *could be OK. Problems mentioned by Ton >2 years ago are considerable (tho prolly other than what BGE users are concerned about?) Fact is, BGE does not need to *compete with other dedicated GEs.

    The all-in-one, “BlenderIsTheNewLego” (grin) approach is what makes me beam about B3D. The ceiling of distribution-size is, however, significant.

    AlsoRE Andrew Brooks: For Blender -=> interactive WebGL 3D, see– a beautiful use of B3D, sans BGE. Also a transparent FOSS business model: pay for play on sensitive uses.

  163. Need volumetric in AMD GPU rendering.
    BTW Blender is awesome!

  164. Why not instead of calling “Blender Game Engine” or “Interaction Mode” just call it “Blender Engine”, if you look Unreal is call “Unreal Engine” and there’s “CryEngine” also the other one is “Ogre Engine”, so why not call/rename the “GE” to “Blender Engine” and by the way if you abbreviate it or shorten it it will read “BE” ;-), that’s what I think about rename the game engine project. Blender itself already competes with other 3D software like Maya and ZBrush for example, why not the game engine or in this case the way I call it “BE”, other than that Blender is great and more user friendly than Maya and ZBrush (and I have both and preferred Blender)

  165. Blender could be easily expanded into a VJ tool, only if some basic option like ui midi mapping and live microphone input in viewport was available. Personally I don’t think it’s that hard to implement.

  166. I cannot commit to using Blender seriously until they have a good implementation of Alembic IO.

  167. in my opinion also new descriptions of geometry with higher precision should be inside blender 2.8, 2.9 or 3.0:

    see phd Raph Levien in year 2008

    “From Spiral to Spline: Optimal Techniques in Interactive Curve Design”

    Normal Bezier Splines have weakness in precision (see Side 141).

    He shows new spiro splines and optimized bezier splines.

    With this technique it is possible to have less problems in geometric operations with splines.

    Dr. Levien made first code for font design. In open source program INKSCAPE this new spiro spline is available since versin 0.47.


    I hope we see some spline thinking of Dr. Levien in Blender soon.

    Example Code is available on site of Dr. Levien.

  168. new live path effects for spiro spline and bspline in development Version of Inkscape 0.92

    also in Blender in next time?

  169. Have you thought about all “not yet implemented” messages ?


  170. I wonder where we are with major and important issues like replacing the layer system with a proper system that doesn’t limit the number of layers, and with removing the crazy 16 material limit on a mesh. These are major issues with people wanting to use Blender professionally, as both make importing and managing complex modelling projects extremely difficult. There is MONEY in supporting architectural users, and without a proper layer system and a sane limit on the number of materials (like available memory rather than some arbitrary small integer) Blender isn’t as anywhere near as useful for architects as it could be. The Blender institute could make money supporting architectural users, money that could be spent on more Blender paid developers.

    Yes, I’ve read about the formidable problems with making a decent layer system, and removing the limit on the number of materials, but both of those seem far more important than whatever happens to the BGE. Many people who would love to adopt Blender for professional use can’t do so because of these two issues. I model professionally and often have to struggle to adapt imported models or make models in silly numbers of parts because Blender won’t let me add a decent number of materials to a mesh. When, please?

  171. I wan’t you guys to dealt with the issues like glass shader and refraction shader. and sharp shadow option in emission shader

  172. Drag and Drop Materials Library PLEASE! If somebody can suggest this:

    Prototyping drag and drop materials are now standard in nearly all modern 3D packages.

    1. Being able to build a shareable and transferable materials library folder in the Blender directory.

    2. In a visible materials library tab, to be able to simply drag and drop the materials live into the viewer over the 3D object parts, where it automatically assigns the material to the part.

    3. Keyshot, Modo (the one that ripped off Blender’s modeler), etc…

    I know you guys are magicians over there, but maybe somebody can work out this wizardry in Phython?

  1. Leave a Reply

    Your email address will not be published. Required fields are marked *


share this story on