I don’t understand obsession with Industry Standard™ ( just like whishes that all software die and only Blender remain )…
Blender is decent software, and this Must be enough for every user. Will Blender become better if become 'Industry Standard? Will sun rise on west? Will my life improve in any way if Blender become Industry Standard??? I don’t see that people which use current IS ( Maya ) are higher, stronger, healthier, faster than people which use other software.
Instead better to work on personal improvements to become artist which can do Industry Standard quality work.
4 Likes
Becoming the industry standard in whatever it is, certainly matters.
It means more attention, and more attention means more money (in the case of Blender/Blender Foundation it means more donations).
More attention, money etc. = more means to develop Blender, to get it even better, faster and more efficient workflow.
1 Like
No… don’t even try to rationalize your wet dreams… cause is much deeper than just $$$$$$.
whatever I’m out.
When? If I had to guess…ballpark?
April 12th, 2034, at 8:57pm
Maybe give or take a few days…
Standards should be useful, like buying nuts for bolts or installing pipe fittings to your toilet that don’t leak…
Nah, February 30th!
I had a quick look at his channel, it seems to be full of tutorial videos but nothing specific to that topic. Do you have an example please?
While I do agree that it is not the main reason, I think it is safe to say it part of them, and not a small one.
I don’t like to sound defeatist nor ungrateful towards the dev’s work, but to me these looks like tiny ant steps towards what other softwares already do.
If that is truly the case then this thread should be locked, why should we discuss this at all if it will never happen anyway?
I do believe Blender can get somewhere if the devs. end up focusing on high performance solutions for tomorrow’s content creation technology (that is, imagining a day where the workflow is like this…).
- Sculpt the model with dyntopo
- (optional), if the dyntopo sculpt does not have a huge amount of polygons, add more detail with editable multires.
- Rig and paint the model without retopology or UVmaps (with the help of layers obviously).
- Use geometry nodes to add, subtract, or modify details
- Use more geometry nodes to automate aspects of rigging or to simulate certain details
- Return to any previous step in any sequence if you change your mind (without losing your work).
Or.
- create the entire asset procedurally with nodes, with any modeling being limited to small components.
- use attributes and nodes to automate texturing, shading, UVmapping, even animation.
Complete freedom from rigid task sequences and the elimination of the need to finalize every step before moving on.
That said, they will still need devs. to focus on a DCC app’s bread and butter (as many assets will still have low enough polygon counts to be handled easily with editmode tools).
I’m pretty sure it will be in the year of Linux on the desktop.
1 Like
By that time Desktop will likely be largely a legacy form-factor for casual client use, which perhaps arguably it already is becoming.
Oh sweet, that’s next year. Just as it was next year in 2020, 2019, 2018…oh.
1 Like
I didn’t know Vray is available for blender but knew about renderman
Linux at least (depending on the distro) has managed to evolve beyond ‘that OS for computer nerds’. General use no longer requires learning how to use the console, the main disadvantages remaining being that not all of the apps. you can get are packaged as neatly as Blender (that and the fact that you still have the gap in app. availability).
So it is not like progress was not made at all, but like how we can’t deny that Blender 3.0 will make even 2.8 look like a toy (if not only because a lot of areas will be so much faster).
I guess that would depend on how one defined performance. If we are talking about code level performance (edit a mesh, use a brush in sculpt mode, render an image, etc) then those are being worked on.
However, if we consider performance from an overall production point of view, so how fast one can actually get things done, then Bob sure has touched on a number of those. A few come to mind, like the wonky implementation of UDIM’s in Blender, compared to most other software. Or editing UV’s in Maya because it is just far quicker and easier then trying to do it in Blender.
There was the time when Bob really wanted to do a water splash/sim in Blender and spent a lot of time trying to get the result he wanted. Then turned to a Houdini user who got the result he wanted, on first try and in a much shorter time.
Now in the end this isn’t to say it all couldn’t have been done in Blender, but in the Industry, where time is money and that time is usually really short, then the path of least resistance or fastest production performance generally gets used.
No. That’s when Blender will gain sentience, and start making terminators out of default cubes to kill off the human race.
Maya will still be the industry standard during that time.
4 Likes
I very much disagree with those asking for a definition of “industry standard”.
It’s not so hard or serious question. A person asked for your opinion, why would you asking for a definition? If somebody at a bar asked you “what good beer do you know”, you’ll not answer with “define good beer” right? (I hope so). That’s silly. No need beating around the bush, you just have to answer what you personaly think.
If you think your “good beer” in need of clarification as to avoid confusion, you just clarify it yourself.
Now to answer the question.
I estimated around 2.81-2.82 that at 2025 it’ll become industry standard. Reason’s for that was:
- Development at that time was very active.
- BF got surge of money that will definitely accelerate development speed.
- Regressions from 2.79 and various hurdles of Blender in general is known.
- I knew at the time that development will not be something like “well we got broken/unfinished things like OSD/multires/edit mode performance/etc. so we will do nothing but fixing those”. Developers have their own plans, their own priorities.
Even accounting that, all major plans for resolving point 3 are not something that’ll need more than 5 years of Blender development (take into account points 1 and 2). You can see it right now. It’s nearing end of 2021 and since then we got better object undo, edit mode performance is in much better shape, multires got some fixes, animation playback is better, OSD GPU acceleration is coming, Everything Nodes now is more than just design discussions, etc. - Blender will not become industry standard by big studios efforts (mainly). They got their own troubles with ongoing projects and need for training if they wish to switch to Blender. Instead what I imagined is that small and mid-size studios are much more likely to use Blender either from the get-go or switching from other software since they are more agile and/or have less requirements. Eventually some of them will become big or will help big one to switch to Blender, like studio Khara.
And with continued development suddenly Blender have most, if not all what industry standard competitors have but with added bonuses like “multiple pipeline stages in one package”, then it’ll become “the future is now, old man” moment around 2025.
I still stand for that estimation but with added points:
- I like BF management of Blender development so far. Development is healthy adapting to increase of money and attention. And I definitely does not fear of “big companies taking over Blender development to their own pace”. There is still problems (patch review for example) but it’s something on a radar.
- Non-monetary investment from industry is more than I imagined. AMD, NVidia, Facebook directly contributing code is great help, arguably more valuable than just paying for an extra developer salary.
- It was not apparent to me at the time potential of interactivity in Blender. With each new feature it’s value will be amplified by all others it’s tied to. It’ll be not “Blender got feature X” but “Blender got feature X which make features Y, Z and A and B and C more powerful”.
For now it is mostly seen as a gimmick to most users/industry. But with continued development in that direction it’s power and value will accumulate until at some point it’ll click in someones head how you can use it to greatly optimize and enhance your pipeline. At that time using Blender tools will become The way to do things.
There is great chance it will happen in next 3 years.
As for what I define as “industry standard”, it’s pretty simple. Nobody really arguing that the likes of 3DSMax/Maya/Zbrush/Houdini is “industry standard”. So we know multiple of such thing, then by looking at them we can use one of following definitions or it’s combinations:
- It’s common to find it as one of job requirement.
- It’s common to see it used at other companies.
- When you tell someone “I use Blender to do X” you’ll not get weird looks (at best).
- There is no talking if it’s industry standard or not.
Pick what you like. 
3 Likes
Thank you! I enjoyed reading your response, and I so agree that industry standard does not need to be defined. It’s clear what it means
1 Like
I imagine that the cool kids, with the money, will buy Maya or something of the sort, and… well… the rest of use will use Blender.
Do the folks at Pixar or DreamWorks use Maya? Blender? Maybe. Maybe not.
Oh, wait. I’m talking cinema. What “industry” are you talking about?
Happily Blending,
Cal
1 Like
Take a closer look at performance in blender : 20k objects, 500 collections are “too much” …
- Rig animation still is 20x (or more) slower than any “industry standard” app - crowd animation while possible is not realistic in blender.
- Support for things like usd, udim are in pretty bad shape, not even close of basic requirements.
- While improving every days, sculpting still is far under dedicated apps.
- Solvers for fluids, tissue, smoke…, soft / rigid bodys are pretty damn slow and not really bullet proof.
- While good, eevee lacks major features, cycles too - caustics … .
- Geometry nodes is a step forward, but lack 10 years of ingeneering.
So in the end, things are envolving in the right direction, but the gap still is pretty huge for any massive industry adoption.
Blender’s fundation lack of community empowerment culture also is a huge issue, switching from small team of core devs to huge team of external contributors is a real challenge, not only in terms of communication, but also to make it a reality.
When you compare to other successfulls open source projects, there is a lot to do in terms of documentation for coding, inboarding, predicatable deprication policy, recognition of contributions value, communication …
And last and probably the biggest issue is the fact that gpl license model prevent any other dev to raise funds for any serious devlopment.
6 Likes
That is not what I expressed, so why do you say that after quoting me?
If you want a bit more depth on how 3.0’s evolution isn’t enough IMO, stephen_leger 's message sums it well I think.
If I go to work and start telling everyone I do my animations on ZB, I will get weird looks. 
Are you referring to these ?
https://lists.blender.org/pipermail/bf-committers/2021-August/051116.html
https://lists.blender.org/pipermail/bf-committers/2021-September/051129.html
1 Like