Meanwhile, in other niches, Microsoft Office still beats open source office suites like LibreOffice; Photoshop isn't about to give up its crown to GIMP; Lightroom isn't losing to Darktable; and FreeCAD isn't even in the rear view mirror of Solidworks.
I wonder what will be the next category of open source to pull ahead? Godot is rapidly gaining users/mindshare while Unity seems to be collapsing, but Unreal is still the king of game engines for now. Krita is a viable alternative for digital painting.
KiCad, for PCB design. They have been making massive improvements over the last few years, and with proprietary solutions shutting down (Eagle) or being unaffordable (Altium) Kicad is now by far the best option for both hobbyists and small companies.
With the release of KiCad 5 in 2018 it went from being "a pain to use to, but technically sufficient" to being a genuine option for less-demanding professionals. Since then they've been absolutely killing it, with major releases happening once a year and bringing enough quality-of-life improvements that it is actually hard to keep track of all of them.
From the type of new features it is very obvious that a lot of professional users are now showing interest in the application, and as we've seen with Blender a trickle of professional adoption can quickly turn into a flood which takes over the entire market.
KiCad still has a long way to go when it comes to complex high-speed boards (nobody in their right mind would use it to design an EPYC motherboard, for example), but it is absolutely going to steamroll the competition when it comes to the cookie-cutter 2/4/6 layer PCBs in all the everyday consumer products.
I can think of a few more: Git obsoleted an entire category of commercial software seemingly almost overnight, VSCode has become by far the most widely used IDE (not entirely open source, though), TeX still dominates mathematical typesetting AFAIK (as it has for as long as computers have been used for that), (lib)ffmpeg is used everywhere for video/audio transcoding and between them nginx and apache still account for the majority of webservers. Most popular programming language compilers/interpreters/runtimes are open source too, of course.
So various projects have come up ever since to try to patch the UX.
I suspect in general "big companies" and the ecosystems they operate in are the main reason adoption (and subsequent investment and development) hasn't happened. The Microsoft Office suite is a good example, since many companies likely run the full Office + Teams + Outlook stack. It all "seamlessly" (not really lol) works together, and it's attractive to sell corporate solutions like that.
Super small nit (or info tidbit), but it doesn't take away from your overall message regarding production and scene scale.
Pixar does not and has not used Maya as the primary studio application, it's really only used for asset modeling and some minor shading tasks like UV generation and some Ptex painting. The actual studio app is Presto, which is an in-house tool Pixar has developed over the years since its earliest productions. All other DCCs are team/task specific.
Dreamworks is similar with their tool, Presto, IIRC. Walt Disney Animation Studio (WDAS) does use Maya as the core app last I saw, but I don't know if they've made any headway with evaluating Presto since 2019...
So while Maya is currently the standard, I don't believe that it's growing. It'll probably be around still in 20 years, with lots of studios having built their pipelines and tooling around it, with lots of people being trained in it, and because it's at the moment still better than Blender in some aspects like rigging and animation (afaik).
And also, how can you say Blender is not battle-proven? I mean, the big studios use Maya like fortune 500 companies use Microsoft Windows - doesn't mean Linux isn't battle proven.
I haven't used blender much. It's too focused on animation. I mainly make more engineering style things for 3D printing. Even though it can technically do that, the interface just rubs me the wrong way somehow. And i can use Fusion 360 for free.
I have used Blender for 3D printing, though. Architectural design, as well.
The interface is a rub at first, and then again after you get used to it and have to use something else. :-)
Those are different niches. Not even Apple has managed to budge Window out of corporate environments, though it is a lot more present than 20 years ago.
I imagine it's similar with Blender and Maya: do they fill the exact same space? Or is Blender adopted by different types of companies (probably smaller)?
https://www.blender.org/user-stories/japanese-anime-studio-k...
tl;dw: probably.
https://www.youtube.com/watch?v=ZgZccxuj2RY
https://www.blender.org/user-stories/making-flow-an-intervie...
You pick a (stable) version, and use that API. It doesn't change if you don't. If it truly is a _major_ project, then constantly "upgrading" to the latest release is a big no-no (or should be)!
And these "most people" who are scared of a Python API? Weak! It should have been a low level C API! ;-)
I wouldn't frame it as "scared". The issue is that at a certain scene scale Python becomes the performance bottleneck if that's all you can use.
> You pick a (stable) version, and use that API. It doesn't change if you don't. If it truly is a _major_ project, then constantly "upgrading" to the latest release is a big no-no (or should be)!
This is fine if you only ever have one show in production. Most non-boutique studios have multiple shows being worked on in tandem, be it internal productions or contract bids that require interfacing with other studios. These separate productions can have any given permutation of DCC and plugin versions, all of which the internal pipeline and production engineering teams have to support simultaneously. Apps that provide a stable C/C++ SDK and Python interface across versions are significantly more amenable to these kinds of environments as the core studio hub app, rather than being ancillary, task specific tools.
Nobody talks about how Linux dominates the server space anymore. Nobody talks about how “git is winning” or getting “battle tested”. These are mundane and banal facts.
I don’t believe the same has happened to Blender yet.
For vendors, the former is obviously a no-go. The latter has the issue of be throttled by Python, so you have to effectively create a shim that communicates with an external library or application that actually performs compute intensive tasks.
Most (if not all) industry DCCs provide a dedicated C++ SDK with Python bindings available if desired.
Davinci Resolve is probably competitive with Premiere, but while free it's not actually open source. But either a viable competitor catching up or Davinci publishing the code could change that fast
I don't have the ability to compare these things in intimate detail, but Lightworks has at least been used for "real" productions [2] so I think it's production-ready.
gcc? It’s hard to imagine any of the projects mentioned without a good compiler.
OBS is on line 2 ....
Tbf, everything starts somewhere and all the proprietary apps you listed were not instant market leaders.
I can and do use all those FOSS tools just fine both as a hobbyist and professionally, my needs are meet. Others may not find the same, but I suspect there's just a lot of stickyness preventing even trying new workflows.
Mine aren't: GIMP is okay, FreeCAD is a complete joke. It is painfully obvious that their development is done primarily by F/LOSS enthusiasts rather than by industry professionals and UX designers. They are closer to being a random collection of features than a professional workhorse. You might eventually get the job done, but compared to the proprietary competition it is woefully incomplete, overly complicated, and significantly buggier.
The poor quality of FreeCAD is the main reason my 3D printer is collecting dust. As a Linux-only user the proprietary alternatives mostly aren't available to me, and FreeCAD is bad enough that I'd rather not do CAD at all. The Ondsel fork was looking promising for a while, but sadly that died off.
I've only ever used openSCAD, Freecad, and on shape.
And for me, Freecad had increased my use of my 3d printer, originally built in 2016.
Cad has a specific workflow, which is true regardless of the tool. Sticking to the right order of operations goes a long way to having a positive XP as does some basic intro tutorials.
It's not a tool you can bumble around and figure out easily, even with XP in similar tools.
Edit: oh, it was even worse before? I hope they keep going in this direction then
If you want to limit standard Office productivity to ones that were written with the GUI in mind, MS Office was the leader on the Mac before it came to PCs and crushed WordPerfect and Lotus early on.
I wouldn't say that, there's bugs in Freecad that drive me bonkers but I would be dishonest if I said it hadn't gotten more stable and better supports my Nvidia card today compared to previous releases
There is just something about it that does not click with me. Just selecting a foreground object even when the background was almost white never worked for me. Just so fiddly.
I can get stuff done with it.
Unity and Unreal are dinosaurs that target the shrinking console market. Godot is being built in their image. My hope is that something more versatile like Bevy becomes common so that we have something that could potentially compete with the next generation of Roblox.
GIMP 2.x was IMO better. Also much easier to compile and get running. Fewer things to worry about.
I think you paint a rather selective picture here though. Quite a lot open source software is really really bad.
I think FreeCAD might be on a distant hilltop in their rearview these days, check it out again.
The most important improvement is the toponaming heuristic solver spearheaded by Realthunder.
Since that was merged into mainline, it seems that the devs keep breaking the UX and shortcuts without rythme nor reason, while the fundamentals are broken beyond repair.
I would never recommend freecad to anybody, even though this this the only CAD I use, and I actually write python for it for some automation.
I cannot live without freecad. But damn it's a mess.
which has been discussed here in the past:
https://news.ycombinator.com/item?id=37979758
https://news.ycombinator.com/item?id=40228068
https://news.ycombinator.com/item?id=41975958
which if it just had parameters/scripting would have a lot more potential.
You don't see the same success with Gimp, Krita,...
Of course Blender is a marvelous piece of software, no doubt in that.
GCC and now Clang has beaten Intel compiler (and others). Nginx has beaten/replaced commercial web servers. Stockfish has beaten commercial chess engines and Lichess is much better than commercial chess sites.
Just 3 examples that came to my mind reading your comment.
May I ask your opinion, as industry insider, regarding what makes a good and bad OpenUSD support?
DaVinci Resolve is getting better but still not comparable to After Effects.
Blender already has a lot of pieces in place to tackle this.
Hadn’t heard that. How many AAA vfx studios have left Maya for Blender?
If anything, OnlyOffice is incredibly polished. I don't know how they did it but it seems to work better and have a much cleaner UI than LibreOffice even though it's much newer.
Is this even true? My understanding is that 3D Studio and Maya used to be ahead.
really? I haven't done 3D rendering in a long time, admittedly, but back then Maya and Lightwave were miles ahead of Blender. Rhino3D too. Even 3DSMax was better. Lightwave seems to have sadly fallen off (unfortunately, IMO it was the best at one point, had excellent ray tracing). I didn't really Blender had come such a long way -- that's great.
Since you mention niches: Adobe InDesign has no OSS competition at all, and Illustrator is still much better than Inkscape.
It must be difficult when so much management is short sighted and focused on delivering short term profits for shareholders. Even academia is run like a business now.
Unless a privately held rogue company like Valve got interested its probably going to have to wait for a government/ngo/scientific. Industry, particularly the tech industry, is notorious for leaching of free and open source software and in some cases building entire businesses on it and not giving back.
It would be interesting to see if they would license that out further for some amount of money.
https://github.com/sandialabs/sgm
Originally open-source, but since taken back in-house. As I understand, which should not be construed as an accurate accounting, Sandia wants to flesh out the basics further before (potentially) open-sourcing it again.
But of course built-in intro of Solidworks was a way better UX.
What I'd do is:
- Spreadsheet workbench --> Create spreadsheet (name it "measurements"). (This is optional)
- Switch to Part design workbench --> Create body (name it "layout") --> select XY plane --> Create sketch --> Create Polyline
- Zoom out, start drawing the rooms in your house, approximately to scale.
- Before going into too much detail, add a dimension (select line --> "Constrain Distance") to the first line you draw, so that you can do the rest of your drawing approximately to scale. Then the general shape won't get messed up when you add dimensions to everything else.
- (If you have a photo or picture, you can import that to sketch over).
- Add constraints to match your room measurements, mostly vertical or horizontal distance constraints. Be careful not to overconstrain the sketch. (You can put the measurements directly into the sketch constraints, or you can put them into the top-level spreadsheet, create an alias for each cells, and then set the dimensions to reference those cells).
- Once the rooms are drawn, close the sketch and create a new sketch on the xy plane called "furniture".
- Draw some rectangles for your sofas / tables / etc, delete any horizontal and vertical constraints that get automatically added (they look like little | and _ icons), and instead apply perpendicularity constraints. Dimension your rectangles using only the "constrain distance" tool. Now you can drag them around the room and rotate them freely.
- If you want to make 3D models for these too, create new Part Design bodies for each room and each piece of furniture, create a shape binder referencing the master sketches in the Layout body, and then extrude the sketches using the "Pad" operation.
That's about as much tutorial as it makes sense to pack into a HN comment. If you give it a try, I hope it works out for you!
FreeCad is rapidly evolving and quite a few tutorials are already using the v1.1 dev builds. Pay attention to the version used in tutorials as you can run into trouble following them if you are on an older release.
Inkscape is good for typing dimensions into rectangles tho
I’ll check out Inkscape as well. I’ve tried using some raster graphics in the past, but I couldn’t type dimensions and had to use the rules and guides with snapping. It mostly worked, but was a bit annoying.
I'd like to hear someone's perspective on how difficult it would be to unify OpenUSD and CAD file formats so that they are portable between programs?
take a look at https://Plasticity.xyz. It's not open-source, but it's got a small, highly dedicated team behind it. It's built on Solidworks' kernel, so it's quite robust.
Also take a look at solverspace, caligula, FreeCAD, ...
Hard nope.
Parasolid is powering practically every major CAD system. Its development started in 1986 and it's still actively developed. The amount of effort that goes into those things is immense (39 years of commercial development!) and I don't believe it can be done pro-bono in someone's spare time. What's worse, with this kind of software there is no "graceful degradation": while something like a MIP solver can be useful even if it's quite a bit slower than Gurobi, a kernel that can't model complex lofts and fillets is not particularly useful.
3D CAD is much harder than Blender and less amenable to open source development.
Can you help me understand why this problem is so hard?
Now, generally speaking, in a CAD model most surfaces will be “analytic” (plane, torus, conical, arc, line, etc). But whenever some complex surface that joins these surfaces is required, (NUR)B-splines are the principal technique for “covering” the gap.
Firstly, you probably have a variety of analytic shapes to represent — things like lines and circles in 2D or cubes and spheres in 3D. Even seemingly simple questions, like whether two such shapes intersect or not, can require a significant amount of logic to calculate the answer. That logic will often be specific to the exact combination of shapes you have, because the number of freedoms and nature of any symmetries in the shapes you’re working with can mean you would use completely different algorithms for superficially similar situations.
Secondly, while you’re probably going to implement a lot of analytic calculations, in realistic models you’re probably going to end up using numerical methods as well. That can be because you need to work with geometry like Bézier curves or NURBS surfaces that has many freedoms. It can be because even if you start with convenient analytic shapes, new geometry that you derive from those shapes, for example by offsetting a single shape or by combining details from multiple shapes as in constructive solid geometry, won’t in general have an analytic shape itself.
By the time you allow for the numerous different types of constraint that you might want to enforce between different types of geometry and the numerous different ways you can construct new geometry from geometry you already have, the scale of the problem explodes. And on top of that, almost everything you do is going to have numerical sensitivity issues, and all but the simplest algorithms are going to need detailed, careful analysis to make sure you really have covered all the possibilities. In this field, “edge case” and “corner case” are literal terms and not just figures of speech!
To give a practical example, without looking up how to do it, could you confidently calculate whether two arbitrary cuboids are completely separate or they touch or intersect somewhere? As another example, given an arbitrary parametric surface, a sphere in a position just resting on that surface, and the constraint that the surface of the sphere must remain tangent to the parametric surface without intersecting it anywhere, how would you calculate the path the centre of the sphere will follow if you introduce gravity to start the sphere rolling in a certain direction along the surface?
These are relatively simple problems in the field, but each already has some subtlety that leaves the “obvious” solutions incomplete. Solve a few thousand problems like that, each unique and with its own calculation strategy, and now you’re starting to get a practically useful geometric modelling system. (You’ve also probably had a team of dozens of mathematicians and developers working on it for decades.)
This is already complex and fiddly enough. Just having a stable 2D drawing environment that uses a constraint solver but also behaves predictably and doesn't run into numerical instability issues is already an achievement. You don't want a spline blowing up while the user is applying constraints one by one! And yet it's trivial compared to the rest of the problem.
Having 3D features analytically (not numerically!) interacting with each other means someone needs to write code that handles the interactions. When I click on a corner and apply a G2 fillet to it, it means that there's now a new 3D surface where every section is a spline with at least 4 control points. When I then intersect that corner with a sphere, the geometric kernel must be able to analytically represent the resulting surface (intersecting that spline-profiled surface with a sphere). If I project that surface into a sketch, the kernel needs to represent its outline from an arbitrary angle — again, analytically. Naturally, there is an explosion of special cases: that sphere might either intersect the fillet, just touch it (with a single contact point), or not touch it at all, maybe after I made some edits to the earlier features.
Blender at its core is comparatively trivial. Polygons are just clumps of points, they can be operated on numerically. CAD is hell.
Fornjot has been attempting this: https://www.fornjot.app
It's going to be years or decades before it's competitive though. Also, it looks like they switched to keeping progress updates private except to sponsors, which means I don't actually have any easily-accessible information about it anymore which is sad.
The tricky bit is having a G2 (or even G3) fillet that intersects a complex shape built from surface patches and thickened, with both projected into a new sketch, and keeping the workflow sane if I go and adjust the original fillet. I hope one day we'll see a free (as in speech) kernel that can enable that, until then it's just Parasolid, sadly.
I read a couple of your CAD comments, what an interesting space. It never occurred to me how complex computing 3d geometries is.
I have used it to make quite a few functional prints, with the help of making sure my scene units are correct and a CAD plugin.
If I put some holes in something that are 1mm from the edge, but then I print it and see it doesn't line up and needs to be 1.5mm, in Fusion I can just change one number and it all updates. Doing the same thing in blender would likely be very difficult.
Many people complain about it being a mesh editor but it works for me. The sheer variety of tooling and flexibility in Blender is insane, and that's before you get to the world of add-ons.
I want to learn Geometry nodes and object generation as I think they will address a lot of the "parametric" crowd concerns. This v5 is meant to be a big step in ease of use of this.
Also, I'm not sure if the different tooling lets me see all the flaws of online "parametric" models, or whether I'm being pedantic. They get frustrating. I have Gordon-Ramsay-screamed "How can you fuck up a circle!".
>Mesh formats like stl cannot represent a circle by its position and radius, while a parametric format like step can.
This is where I think the Geometry nodes can help. A node (function) can be used to represent the circle with inputs and outputs set or changed as required.[0]
I have not fully explored this space though and so my "hopes and dreams" may well be as useful as thoughts and prayers...
[0] https://docs.blender.org/manual/en/latest/modeling/geometry_...
Maybe it is the export or something. I run the 3D toolbox and often models are not manifold.
I see things like two circles in slightly different positions but both are connected in different ways to the surrounding "single" instance model. Things like this mean you end up with "infinitely small volumes". There is no fully enclosed "volume" and so mathematically there is "nothing to 3D print".
As a model this makes no sense to do, and so it irks me.
But clearly the slicer software doesn't care or autocorrects and people make their 3D print happen just fine.
KiCAD was also a meh ECAD FOSS alternative 7-8 years ago, now it is by far the tool of choice for regular ECAD designs. I can see FreeCad getting there by 2030.
It seems like it has lots of capability but still "punch your monitor" levels of difficulty just trying to do the most basic stuff.
MangoJelly has done an amazing job in churning out high quality tutorials for FreeCAD: https://www.youtube.com/watch?v=t_yh_S31R9g&list=PLWuyJLVUNt...
(this is just one playlist, there's a lot more on his channel).
Deltahedra is a great YouTube channel for getting the basics.
And it presents nonsensical problems, like offering to create a sketch on the face of an object and then complaining that the sketch doesn't belong to any object. So you have to manually drag it under the object in the treeview. So gallingly DUMB.
Despite all that, I will wrestle with its ineptitude before giving Autodesk a penny. I get stuff done with it and respect those who give their time to develop it.
I'm sure I could grind harder and learn more and make FreeCAD work, but I'm not sure why I'd bother.
The algorithms it enables are fundamentally more capable and robust than traditional kernels based on linear algebra (vectors and matrices). You can do really fancy things like interpolating in space and time robustly, find extrema in high-dimensional phase spaces, etc...
This could potentially allow straightforward and robust solvers for kinematics, optimal shape finding, etc...
Every few decades there's a "step change" where some new algorithm or programming paradigm sweeps away the old approach because suddenly a hobbyist can do the same thing solo that took dozens of developers a decade in the past. I suspect (but cannot prove) that PGA is one of those things.
Previously discussed here:
https://news.ycombinator.com/item?id=30597061
My thinking is to approach the problem from a fundamentally different angle. There's already constructive solid geometry (CSG) kernels, triangle mesh kernels, and NURBS-based kernels. Their mathematical foundations are very different, which results in wildly different behaviour and capabilities.
I came across PGA while studying physics, saw some vaguely CAD-like CSG demos and I realised that it could be yet another mathematical foundation on top of which CAD applications could be built.
Notably, variants of GA and PGA are already used in robotics, inverse kinematics, etc... including 5-axis milling, so it's not unheard of in industry. However, it's always used as a "spot" solution to work around a problem such as gimbal lock, or interpolating transformations. Typically by converting back-and-forth between linear algebra representations and some variant of GA temporarily. I'm thinking of using PGA throughout as the foundational geometric elements.
You're on point that there's a tremendous amount of money captures by Autodesk for CAD software that could be better directed at the open source community instead.
Software like OpenSCAD and FreeCAD are obviously not suitable for much commercial work, and have very irritating limitations for hobbyist work, in my mind a big part of that is the UI and Blender has a good and established UI at this point so I'd love to see the open source CAD that provides an alternative to vendor lock in come from a Blender add-on instead of a separate program.
I am no expert but as I understand it the primary difficulty with developing good alternatives to commercial CAD software lie in the development of an effective geometric kernel.
It seems to me that if a developer of an opensource CAD program develops it as a Blender add-on they can effectively outsource the remainder of the development efforts to the Blender community while focus can be made on the CAD kernel itself.
Radial tiling my beloved, and a seemingly far more straightforward array modifier <3 Faster volume scattering for non-homogenous volumes.
For those wondering "where the AI is", the new Convolve Node might be it :) Convolutions are a pretty generic signal processing operation (Hadamard product) which are also used in neural networks which work with images. Realistically though, this will be mostly useful for wonky hand-crafted blurs.
The new sequencer looks fantastic, too. I always went to DaVinci Resolve but I might be able to go full blender. Compositor modifiers in the sequencer is also very welcome.
This is incredible for me.
It’s powerful and pleasant to use. Even the release marketing page is beautiful and well-made.
I like open source as much as the next guy, but outside of developer tools there is little that comes close to Blender in terms of utility and UX.
Is it funding? Specific individuals? Are there PMs and designers? Whatever it is, it’s working!
Relying on individual donations from users helps a lot with blender being aligned to the interest of its actual users. There is not one or a few corporate sponsors controlling everything.
Plus the GPL license which protects the freedom of its users.
The creator apparently was selling it as freemium software in 1998, and then the bubble burst and the corp shutdown in 2002. But the creator created a non-profit called the Blender Foundation, launched a Free Blender campaign [2] (the forum post is still up!) to raise money from its users and bought out the rights to the software from the investors.
[1] https://www.blender.org/about/history/
[2] https://blenderartists.org/t/free-blender-campaign-launched/...
Blender nodes have come a long way over the past decade and it's incredibly satisfying to see the care with which they have been developed. Blender's node editor is my personal favorite node editor I've ever used in any software, and I often find myself wishing other software adopted some of their UI and UX conventions.
Been a happy user since, oh, v2.75? And looking forward to being a user for many more releases to come.
Donate to Blender! [0]
I'm very excited to see the addition of structs and closures/higher-order functions to blender nodes! (I've also glanced at the shader compiler they're using to lower it to GLSL; neat stuff!) Not only is this practically going to be helpful, the PL researcher in me is tickled by seeing these features get added to a graphical programming language.
If you haven't heard of Blender before, or if you think AI will replace all the work done in it, fair enough. But I'd still strongly suggest looking into what it is and how it works.
These days I wouldn't want to bother doing manual animations or mesh creation. The computer must do this for us.
Learning the shortcuts is really the way to go though. Initially painful but 100x more productivity in the long run, and really the way Blender is meant to be used. Honestly just doing a single basic tutorial will have you learning all of them in no time.
Always nice to see these updates though, Blender has really come a long long way.
Now I want to look into it more, but I'd imagine that "Blackbody" and sky generation nodes might still assume a linear sRGB working space.
Since people are always asking for “real world examples”, I have to point out this is a great place to use an agent like Claude Code or Codex. Clone the source, have your coding assistant run its /init routine to survey the codebase and get a lay of the land, then turn “thinking” to max and ask it “Do the Blackbody attribute for volumes and the sky generation nodes still expect to be working in linear sRGB? Or do they take advantage of the new ACES 2.0 support? Analyze the codebase, give examples and cite lines of code to support your conclusions.”
The best part: I’m probably wrong to assert that linear sRGB and ACES 2.0 are some sort of binary, but that’s exactly the kind of knowledge a good coding agent will have, and it will likely fold an explanation of the proper mental model into its response.
If you make a color space for a display, the intent is that you can (eventually) get a display which can display all those colors. However, given the shape of the human color gamut, you can't choose three color primaries which form a triangle which precisely contain the human color gamut. With a display color space, you want to pick primaries which live inside the gamut; else you'd be wasting your display on colors that people can't see. For a working space, you want to pick primaries which contain the entire human color gamut, including some colors people can't see (since it can be helpful when rendering to avoid clipping).
Beyond that, ACES isn't just one color space; it's several. ACEScg, for example, uses a linear transfer function, and is useful for rendering applications. A colorist would likely transform ACEScg colors into ACEScc (or something of that ilk) so that the response curves of their coloring tools are closer to what they're used it (i.e. they have a logarithmic response similar to old-fashioned analogue telecine machines).
or you are saying if there is some intermediate transform that makes color go beyond P3 it will get clipped? then I understand...
Exactly! The conversion between ACES (or any working color space) and the display color space benefits from manual tweaking to preserve artistic intent.
Seems like in 10 years AI will basically make it pointless to use a tool like this at least for people working on average projects.
What do folks in the industry think? What’s the long term outlook?
The fact that it “seems easy” is a great flag that it probably isn’t.
Really no one can predict the future.
What blender and other CGI software gets for free is continuity. The 3D model does not change without explicitly making it change.
Until we get AI which can regenerate the same model from one scene to the next, the use of AI in CGI will be severely limited.
Recent news on major AI scientists starting "world AI" companies confirm this trend.
So 3D soon will become a very important tech even compared to today.
It's like 2D art with more complexity and less training data. Non-AI 2D art and animation tools haven't been made irrelevant yet, and don't look like they will be soon.
I recently used WAN to generate a looping clip of clouds moving quickly, something that’s difficult to do in CGI and impossible to capture live action. It worked out because I didn’t have specific demands other than what I just said, and I wasn’t asking for anything too obscure.
At this point, I expect the quality of local video models (the only kind I’m willing to work with professionally) to go up, but prompt adherence seems like a tough nut to crack, which makes me think it may be a while before we have prosumer models that can replace what I do in Blender.
AI coding agents didn't make IDEs obsolete. They just added plugins to some existing IDEs and spawned a few new ones.
You are asking for industry predictions from industry professionals in an industry you know nothing about while assuming a lot about that industry.
Why do you think they should do all the heavy lifting for you?
You might as well ask ChatGPT what it thinks because it seems you already have an idea of what you want the answer to be.