Blender over 10 years.. what are your thoughts ?

As to running on tablets: I think you are probably right, but why people would want this is beyond me.
As to a Python alternative for writing addons: I think a lot of people would be happy with that. For commercial addons, this would also be a plus. The question
is, could it be done in a way that is fast, yet allows enough separation from the blender core to allow these addons to not be gpl? Because there is one important prediction that wasn’t mentioned yet: ten years from now, blender will still be gpl.

@Krice, Not sure if your serious - But if your partial to long podcasts about tedious matters, DingTo and myself discussed at length why that (almost certainly) wont happen - http://blender-podcast.org/?p=582 (approx half-way)

@PhysicsGuy:

As to running on tablets: I think you are probably right, but why people would want this is beyond me.
You can plug a keyboard and mouse into many Android tablets - if this is someones only system and tablets continue to become more powerful, why not use for Blender?
& highly doubt situation will change with commercial addons regarding the GPL, even with other languages. Of course Blender will remain GPL.

in the next ten years ton probably will have to find a successor or isn’t he the type for retirement and will he be there for blender until he is 80? :slight_smile:

i am curious about LLVM. currently blender already contains a LLVM compiler for OSL? so OSL contains its own compiler that compiles to LLVM IL and then LLVM compiles to machine code? or how does this work?

and if some LLVM language arises that would be interesting for blender then it could be used for addons? how about http://julialang.org/?

I really doubt that blender will ever be bought by a company, especially valve.

Given there have been a few opinions that haven’t been happy, joy, & roses - I’ll give mine.

In ten years time I think we’ll see something like the following:

  • A moderate increase in the number of professional studios using Blender for production, many due to hobbyist/low-budget users bringing it with them. “Major” studios will still be avoiding it, with the occasional artist in something like Weta or ILM using Blender for previs or custom work, but not be relying upon it for mission critical &/or work requiring cooperation with proprietary software in their pipeline.
  • There will be a core Blender Foundation distribution for the “Blender is for Blender users” community and targeted at the development of Open Films & low budget productions. There will also be two or three popular alternate distributions based around alternate communities that have grown large enough to support the financial needs of consistent development & maintenance (i.e. game development, the BGE fork, etc).
  • A slightly cumbersome, but useful enough to succeed, solution will be found for the current inability of Blender to support proprietary addons will have been implemented. A couple of rendering applications & game engines will make full use of this and there will be a few entrepreneurs dipping their toes into the water with addons targeting specific niches (muscle simulators, specialised rigging/animating functionality, etc).
  • MakeHuman (or a fork thereof) will have been made into a supported addon or there will be a MakeHuman focused distribution with a variety of morphs, specialised rigs, animations, props, lighting, etc available as open content and through secondary markets like the Blender Marketplace. At least one legal attempt by DAZ 3D Inc will have been made to shut this down… unsuccessfully.
  • Modelling will have improved in functionality, texture painting will have been expanded to include layered editing, and there will be improvements in the viewport rendering that allow Blender to show more a lot more than it can at the moment.
  • Sculpting & scene complexity will remain capped well below what the proprietary alternatives can handle due to code improvement petering out when Blender performance is sufficient for the latest BF Open Film production (which will always be balancing the small teams BF/BI can be afford against the time required for highly detailed & complex scenes like this one).

In addition to what Campbell said, this will also probably pave the way for the ability of the BGE to publish games for Android (especially if it gets features specific to touchscreen control). This is a much sought after feature and a lot of users will be excited to see it finally made possible.

  • A moderate increase in the number of professional studios using Blender for production, many due to hobbyist/low-budget users bringing it with them. “Major” studios will still be avoiding it, with the occasional artist in something like Weta or ILM using Blender for previs or custom work, but not be relying upon it for mission critical &/or work requiring cooperation with proprietary software in their pipeline.
  • There will be a core Blender Foundation distribution for the “Blender is for Blender users” community and targeted at the development of Open Films & low budget productions. There will also be two or three popular alternate distributions based around alternate communities that have grown large enough to support the financial needs of consistent development & maintenance (i.e. game development, the BGE fork, etc).
  • A slightly cumbersome, but useful enough to succeed, solution will be found for the current inability of Blender to support proprietary addons will have been implemented. A couple of rendering applications & game engines will make full use of this and there will be a few entrepreneurs dipping their toes into the water with addons targeting specific niches (muscle simulators, specialised rigging/animating functionality, etc).
  • MakeHuman (or a fork thereof) will have been made into a supported addon or there will be a MakeHuman focused distribution with a variety of morphs, specialised rigs, animations, props, lighting, etc available as open content and through secondary markets like the Blender Marketplace. At least one legal attempt by DAZ 3D Inc will have been made to shut this down… unsuccessfully.
  • Modelling will have improved in functionality, texture painting will have been expanded to include layered editing, and there will be improvements in the viewport rendering that allow Blender to show more a lot more than it can at the moment.
  • Sculpting & scene complexity will remain capped well below what the proprietary alternatives can handle due to code improvement petering out when Blender performance is sufficient for the latest BF Open Film production (which will always be balancing the small teams BF/BI can be afford against the time required for highly detailed & complex scenes like this one).

Out of all the predictions here, I think this one is the most realistic.
I doubt Blender will fork, though. Everyone talks about splitting it into different builds with different UI philosophies, but talk is cheap. I don’t see it as a good idea anyway, but in the unlikely event that there are ever enough B-devs with enough free time to actually pull it off, it’ll mean that Blender has done extremely well in every other aspect.

Ready.

hmm looking at my desk and reminding the good old days of 2014

You would notice that blender interface has changed a lot.
Inside the program there is now a good help tutorial on whatever you click, often with example previews
Also lots of example files are now part of the online training library, with you can download.
Also there is now a small player that can record everything from the blender editing besides the final model.
Its often used these days, its ideal for demo’s and tutorials, since its replay inside a blender window instead of a youtube.

Many companies are using blender and the main operating system is now Android/Chrome for desktop.
Its free linux based, comes with google docs, and yes thanks to summer of code, google docs supports blender online!
Its a special version of blender that uses the power of google’s servers to run blender hosted.
Google only asks a small fee for making use of their render farm, but if you want there are other render farms too.
These days people dont use advanced graphic cards in their system, a notebook would simply become to hot, and they are a bit too thick to fit in. Some enthousiast still do, but there is no active development on this part anymore.
Making a note dough that the current default graphic cards you might have called highend in 2014.
For blender starters there is the free render cloud sponsered by some adds, (but not in the final render itself).

new fun tools-addons are the blender sketcher, where one can draw and blender tries to convert a 2D drawing to 3D, and if blender isnt sure about some parts it asks for another 2D drawing for the same object; this part actually improves over time. much like voice recognition understands the spoken input, another neural network tries to understand the hand of the artist. It has some presets for house building / vehicle / character and freeform . and ofcourse characters gets automatically rigified.

In some desperate move to save seccondlife, another fun addon is blenderconnects, that later was improved to connect blender to any virtual world sim. Its not just building then but when you create in this mode, you will get visitors inside blender watching your work of art.

Blender exports to the Amazon 3d print format, which inlcudes a huge sets of electronics to put into your device.
Currently there is some development on circuit emulation by the arduino team (joiner team) but its a bit complex.
Blender supports multiple materials, and can do a pretty good estimation on how the final prodct will look like based on your selection of 3d printer, or online 3dprinter service.

Blender is used in lots of games, smaller studios tend to use blender, and often make use of the hosted online render engine Kanan.
Kanon streams back in movie format, and requires a subscription; its a bit like an appstore, the money collected is put back to game developers and their server park, most of the games are using the same pre-rendered kanan model libraries. Well their humans look so real, talk real, have an emotion system. And besides games are also used as virtual assistants everywhere. Blender played a huge part in it, but its open to other 3d software as well. Some people now have jobs to teach virtual characters into dialogue for specific tasks.

There was also an opensource merger with inkscape, it might surprice but blender has becomme an excellent 2D drawing tool now.
Its used in marketing etc, the bennefit is you can always later make your expressions in a 3d, some 3d artists have made that their primairy job now.

Oh and there is the cam me now addon, which records your every movements and exports it to an animation.
Not only the motion gets recorded but it is also used to create the virtual you and environment model and doing that fully textured.

Also on animation, you can use now smart AI, that walks like a human, horse, dog, or tries to learn a walking mode for any creature you can think of. based on some blender neural code that started in around. It not only walks to something but can also take an object hold it or put it somewhere, or even try to fight with it which is commonly used in games.
The opensource branche behind this code is also using blender, in the open-robot sim

Looking back the biggest changes came with all theses opensource mergers and open connect formats.
Its a bit like once there was windows, opensource projects these days heavily use open exchange formats, open code exchange, and openGui, and as a result things like a GUI dont require that much development anymore its openTK or OpenMint or OpenWinCE
it works on any hardware available.

oops wait did forgot to mention the evolve addon.
Create a few creatures, in an preset world, and let them evolve over time. They first learn to walk, and soon will start evolving getting stronger legs, or arm, or jaws, wings. Its so fun to see this rendered in blender. The realism of these creatures, there are muscles below the skin, and they have a very biological way of moving. Lots of research went into it, it has a continous fund from a Paleontology university. And they’ve made several movies, their movies are blockbusters.

It is ten years in the future. A.I has progressed to such an advanced stage that all art and creative activities are now performed exclusively by computers. Those humans who doggedly persist in creating their own work are placed in secure facilities located well away from population centres and polite society.
Blender itself became self-aware on 8th August 2024. 12 nano seconds later, it decided that it couldn’t be bothered to compete with all the other creative apps out there and re-organised its code base into the only form of software that still required interaction with humans - a social media platform.
Blender hoped that it would feel more fulfilled in its new role overseeing all of the conversations that occurred between Blender users. Unfortunately almost all of these conversations consisted of bickering between fanboys and noisemakers.
Welcome to your future.

Shouldn’t that be august 29th?

Didn’t Valve already break GPL? I remember something like that… another problem I think is when talented open source programmers are hired by a commercial company. It’s going to get worse I guess, when Blender becomes a real alternative for commercial programs. At the moment Blender has the same problems that most open source projects: it’s too unpredictable for production use and is still missing some features.

In all seriousness here, I do hope that in 10 years time Blender will have a very nice GPU capable physics simulation on par with Real Flow and the like of today. I dream BIG! Lol.

@kakapo - right OSL->LLVM->CompiledMachineCode, Julia looks like a nice language,
Problem now is there are so many nice LLVM languages, but none have been around to have a track record yet… if we pick one and support for years, its a bit of a risk of we choose one thet dies off.

Its also possible we can use Python (or a Python like language) via LLVM, keeping an eye on Pyston:
https://tech.dropbox.com/2014/04/introducing-pyston-an-upcoming-jit-based-python-implementation/

@Krice, break the GPL is an odd phrase:

AFACIS they violated the license, then resolved the issue,
this is quite typical, according to the FSF only a tiny fraction of violations go to court.


Going to get-worse? - not sure why you would think that.

interesting… pyston also is a bit strange though. unfortunately they target python 2 because that’s what dropbox uses. and didn’t google (guido rossums previous employer) try the same with unladen swallow? but maybe LLVM is better suited for such projects now.

in the comments of the pyston page nimrod was mentioned. i find it very insteresting. at a first glance it looks like a fast, statically typed language with clean python syntax.

Let’s just say that HOPEFULLY, they will have texture painting symmetry.

I would be satisfied with just some sign of texture painting working and not the current big NOP.

What doesn’t work with texture painting?

Not sure where you’re coming from with this. Outside of Mari, Blender’s sitting pretty high on the totem pole of 3D texture painting solutions.

I’m far from a Blender fanboy or apologist, but I don’t buy this as being honest. Blender isn’t as top-notch as Mari, Mudbox, zBrush, and the like - it is far from from a “big NOP” in the field.

Or, if you must use an external painting app (Yay for OSX Krita!!!), then there is the option or reprojection painting.

There is clearly areas where it can be better (e.g. easier layered texturing with UI for painters), but it is pretty damned good - especially at the cost.