Hey everyone! I’m doing research to understand where artists are actually losing time in their workflow not where tools think we are.
Whether it’s tedious UV work, setting up material nodes from scratch every project, re-doing the same lighting setups, or something more niche I’d love to hear what you’d hand off to a tool if you could.
Drop your answer below even a one-liner helps. What’s the one thing that makes you sigh every time you have to do it?
UV work…always hated it…still hate it. It does not matter what program I am in.
Always debugging the problems, trial and error. It’s the only thing I seem to have problems with, not sure why.
1 Like
Hello !
I don’t think there is a lot of ground that hasn’t already been covered unless you came up with a completely new technique…
The most well known boring tasks are generally : UV, Retopo, Skinning …
And there is already a lot of techniques to avoid or automate them in some cases.
In the meantime, if you really want something precise/optimized you need to go manual.
Generally there are more possible automation in a specific workflow. For instance in animation you might need to deal with 200 shots, even more than 5000 when doing a series or a feature film. There is also a lot of assets.
Then there is a clear need for automation so every files is at the right place, you don’t have to set a render path, or you generally never need to navigate the project’s files because everything is automated. But these tools doesn’t make a lot of sense in another workflow, like archviz.
In this case, it’s more about having a database of assets for furnitures, materials, so setting up these recurring elements is super fast. Which makes less sense in a animation workflow. Or both uses asset management to some extent but they are not exactly organized the same way…
I think these things are best answered by yourself, if in your way of working you find something that can be automated or improved.
Good addons are made of that, hardops is a good example. But Gaffer for lighting is pretty interesting too, Auto Rig Pro is also an artist’s tool at first that became an addon eventually… I think it’s the way to go rather than trying someone else’s issues because it’s always tricky unless you really work hand in hand…
It’s also worth considering all the tools that are already out there, since it’s a business too, people always look for the next best things but when this is disconnected from an actual practice it became a bit artificial and it doesn’t really solve an actual artist’s issue.
5 Likes
I’m not a professional, but for me the vast majority of the process is soothing. Even the tedious stuff. Especially the tedious stuff, actually. It helps me to unwind when I’m stressed out about something (which is usually).
I suppose my biggest gripe is the organization system. I never fully warmed up to the ‘collections’ idea. I don’t think that’s an automation puzzle, though…
I feel like an arbitrary tree of folders (like a conventional file system) would be a whole lot cleaner and more flexible, with ‘collections’ serving very specifically as macro objects. And bring back the simplicity of indexed layers. That was nice. 
6 Likes
This is a very good point about tool/software/anything design! You need to experience the issue yourself to really understand it and come up with a good solution.
1 Like
- Getting full and accurate information needed to do the job from the client.
- Clarifying vague or conflicting information provided. Waiting for responses.
- Organizing disorganized information provided by the client
- Searching for some information the client provided in one of the 12 different ways they choose to communicate with me.
- Helping the client decide what they want
- Helping the client to do their job
- Providing psychological and moral support for the client
- Explaining why my work cannot be free or 10 times cheaper to the client
- Explaining why it cannot be magically done with AI with no effort or cost at all
- Gathering information about objects that I need to model. Dimensions, materials, gathering reference information about how stuff looks or may look especially when no good reference actually exists.
I am good with everything else. Especially with the 3d part - I actually like it 
15 Likes
For the UV’s… it a mathematical not generally solvable problem… and sometimes i think some artist simply do not know that… of course there are some “typical” UV usecase where some more fitting algorithms are available… but one has to know which to use when…
For materials and lighting setups: there are asset libraries…
Also i think the some (general) question like : “What part do you want to be fully automated in your workflow?”, is too usecase specific… or would “replace” the worker/artist…
Surely for any artist prepareing the workpiece be it the canvas for whatever technique or the wood to carve the big parts of or wrapping the aluminium folie arround some wirework or warming the clay before sculpting… is somekind of tedious. But it also prepares your mind for the piece of art because you have to think about how to handle it.
So the part which never could be automated for me for any software is the preparation to be in the right mood even for writing a text (likw this here too)… so if i can not concentrate and have not the time to do something then the only tool i need is some “mental readiness” and for this sometimes i need a little bit of “planning” when to do what; sometimes also only “in my head” and sometimes by looking at the calendar (i’m still not using my smartphone for this…) and maybe some piece of paper and a pencil or my pen…
Call me old school. But no app (nor some AI) ever will take away the fun and satisfaction i have to do things by myself… 
( I sometimes see some frustration on people super organizing everything with apps and internet services… some can not do anything when not online… 
)
Whenever i good one minute or even a day off for “waiting” for something i never got bored… because i’m soo much interested in soo mcuh things…
( Of course sometimes i need to pay a profesional for things i can not or not allowed to do 
)
…
Also: you can not loose or save time… see Momo …
4 Likes
@Martinz: “Hear, hear!” 
The only real problem is always the clients!
But, seriously: I have always been impressed with the “plug-ins” system of Blender, and the way that said plug-ins can integrate themselves into the user interface as though they weren’t built in Python. There are an astonishing number of plug-ins that are shipped with the product, but not automatically turned on.
As a programmer, I will readily give praise to the clean engineering that was done to make all of this happen. It is … (highest praise here) … "a Cool Hack.™"
I wouldn’t pay for an LLM that just automates what I already do in Blender, but I’d pay anything for one that could handle every single item on your list, step by step. I’d happily donate all the emails with my clients over the years to help train a model like that. 
5 Likes
Maybe LLMs could handle very small part of the emails and electronic communication…
But to be honest if the task is quite hard even with natural intelligence… Maybe it’s not going to be trivial to artificial one… 
As for the rest of it… It’s not possible in principal. i mean let’s take collecting reference… LLM is not going to organize a trip and go take pictures for you.
It might sound like a good idea for communication with clients. I’ve tried asking LLMs for help with emails… It was really bad… LLMs don’t get the specifics of the job and on top of that they make it all sound so… as if I was 18 and was applying for my first job. I don’t want that talking to my clients. I want me there. My character. I want to develop personal relationships, because that is what matters if I want to keep the clients and at least attempt to deal with all the listed issues in communication with them. LLMs would absolutely ruin that. Bad idea. At least for my purpose. It could probably work at McDonald’s just fine.
4 Likes
Oops, I got carried away and missed that point. You’re totally right!
As for the rest, let’s dream. It’s free. Maybe in the future, hopefully not too far off, we’ll have real AIs (not LLMs or similar toys) capable of handling clients just like you would.
2 Likes
Well, I think the best way to delegate client relationship and focus on creativity is to work for a studio/company… It’s true that when doing freelancing you need to account for so many things beside 3D and 3D is then the easy part.
In the meantime freelancing is better paid also because of that. And that client relationship could be interesting too. This is where you work your social and diplomatic skills which are generally useful in life.
While clients are generally a PITA still in 2026 I think we lack a lot of social interactions so it would be sad to delegate those opportunities to AIs.
3 Likes
If it was a dream it could be a nightmare. It will not be AI handling clients, it will be AI handling other AI the clients use. So… that’s just another useless layer obstructing actual communication then Not a good idea even on paper.
3 Likes
Haha, when you put it that way, I’ve got no choice but to agree with you completely! 
I also have to say, out of all the hypothetical scenarios I know about what would happen to our civilization if a true AI emerged, my favorite is definitely the fusion between them and us. From that perspective, it wouldn’t be a bad idea for the part of your “self” taken over by AI to handle the administrative tasks, while the organic side focuses on the creative stuff.
There is just no winning with this, right? Sorry for being negative But yeah, automation… It’s hard. Everything about computers is already automation. Everything in Blender is already a kind of automation. So I think it just comes to developing it further. But I sense this question might not really be exactly about that…
3 Likes
Y’know, I was belittling “clients,” but really only in jest.
One of the most memorable moments I ever had was immediately on conclusion of a messy, “save the system in a hurry” web-site rescue/redeployment project. (Involving about 100 sites at once, all of which had been perilously close to “going dark.” Several million dollars of on-line sales every month.)
The morning after a flawless “go-live,” which remained flawless, I coincidentally happened to encounter the owner of the company(!!) at a favorite coffee shop. He was “all smiles,” and so was I. He reached out, shook my hand warmly, and just said: “Thank you.” Those two words, and I was in orbit!
But, yeah, it does matter when you do quality work for a client, working your a&s off to fulfill every promise that you made to them, and they … thank you. Because you know what that means to them, and therefore to you. “Corporate employees” don’t get that experience. I guess I chose “the road less traveled by,” and that has made all the difference.
1 Like
No need to apologize at all! Honestly, from my perspective, you’re doing exactly what a discussion forum should be about: bringing some critical thinking to the table. I took it as a joke, and while a little humor never hurts, the truth is I 100% agree with everything you’re saying.
2 Likes
I think the UV and the baking part from high poly to poly. Kinda very technical and any technical side can somewhat be automated, but sadly not fully if the flexibility is important.
Well I’ve researched andi built an addon that can fully automate your UV work and I can sell it to you for 50% off
https://superhivemarket.com/products/smart-uv-tool
I would say baking is one of the least supported features out there, not just in Blender but in other software as well, one mistake and you practically have to redo the whole process of selecting all your objects and rebaking all the map types. Not to mention how skewed projections are badly handled everywhere, relying on Marmoset or Substance Painter for baking is also problematic. If you export both your high and low poly and one of them turns out to have some kind of mistake once baking is done, you have to go back to Blender, fix it, and re-export everything. A bridging tool might help with that, but I still think that baking is something that should be 100% done only in Blender before forwarding the images to texturing software of choice.
In the end, a modular and non-destructive approach to baking would be really something, at least in my opinion.
2 Likes