How to become faster? (3D game art)
polycounter lvl 3
Hey guys! I have a really huge problem in my workflow and i have no idea how to solve this. I'm working VERY slow. For example, I can spend like 5-6 hours without breaks (like to check FB or watch youtube), just to find out that I've done so little amount of work. I always find myself remodeling and fixing some little part that doesn't really affect overall looking, again and again. Sometimes I become obsessed with little stretching on subD model and again - spend a lot of time just to fix this. I can model the entire asset and then delete some parts because it doesn't have perfect proportions. And the situation becomes even worse when I'm not sure about some detail or part (for example if i cant find any reference from a needed angle of view). I just always have doubts if my assets look like they should be.
I don't think this is something software related, I spent a lot of time to customize shortcuts, learned scripting to automate some actions, and don't have any troubles to model anything, if I'm sure how it should look. I just can't understand how people able to provide high-quality work fast.
It mostly about high poly modeling, i don't have any issues working on LowPoly and texturing.
How can i overcome this?
Sorry for grammar.
I don't think this is something software related, I spent a lot of time to customize shortcuts, learned scripting to automate some actions, and don't have any troubles to model anything, if I'm sure how it should look. I just can't understand how people able to provide high-quality work fast.
It mostly about high poly modeling, i don't have any issues working on LowPoly and texturing.
How can i overcome this?
Sorry for grammar.
Replies
Even our best modelers here are deleting objects and building it new from the ground up.
If you failed the first time you learn that that was the wrong way. Next try.
But also it's important to develop the skill of being able to switch gears, between rough & fast vs. tight & detailed.
I need to be able to pull back periodically to see the bigger picture, making course adjustments as needed.
It helps to develop the skill to prevent yourself from getting stuck down deep in the weeds.
Work from large to small. Refine and complete your large proportions before going in and doing all those cool small detail work. At a certain point you can't keep second guessing yourself on these large forms. Something that helps with that is sufficient and consistent references. As much as we all would like to just pull models out of our mind for a large part it just doesn't happen like that.
Use your reference to set your foundation and build upon that one step at a time. This will stop you from putting hours into one small portion only to zoom out and have it look odd. On top of that it just comes with time. You've taken the first step in speeding up workflow with hotkeys and scripting so that's good. Just keep doing it and doing it, you'll start to notice the mistakes you make along the way and learning from them.
If it helps try recording your work sessions that also include an onscreen key/mouse click so that way you can go back and analyze things if you feel like you really messed up and wasted time during that session.
Kitbash
After watching a handful of gumroad tutorials/breakdowns from some great artist, I think this is obvious and true. Nobody's model looks awesome right off the bat. There is long stages where the best artists' work looks like kaka. But they have a vision in mind, and they keep at it until they get there. It's a matter of trying new things, redoing work as many times as it takes, and being systematic so you aren't making the same mistakes over and over again.
If you haven't already, find some artist whose work you admire and see if they have breakdowns/tutorials you can buy. The buttons they are pressing doesn't matter -- it's the logic and process. And something else that may happen is that you notice little things in the pro's model at different stages and, if it were your own, you know it would be bugging you. And you'd waste time trying to fix this "flaw." And then you realize a lot of what slows you down is just simple insecurity.
1. Make sure you spend enough time on the block out to get the proportions right. Figure out ways how to model non distructive in case of bigger changes being needed - for example using latice deformer in Maya to scale only certain parts of the mesh or stay as long as possible in as simple forms as possible..
2. For you it seems decision makeing is an issue. Bacially the lack of experience (two years are far from experienced) puts you in a situation where you still see things that you can improve and not enough sense of when it is good enough. Here is a simple trick for that: Deadlines. Make sure you have personal deadlines that you force yourself to make. I am doing it all the time. The first 80% of the deadline I am willing to experiment, make bigger changes and all that stuff, but when I reach the final 20% of the time at my disposal I make certain that I know what needs to be done to get the most out of it and I make sure I finish on time. Afterwards its up to others to give me feedback or say if something needs to be changed and then the cycle begins a new, but I never go into a task at work without having some clear idea how much time I have at my dispoal and when I'll reach the point where it simply has to be good enough.
3. Its still mostly experience. Most of my speedgain is decision making and seeing things that earlier I didn't see or was still 'researching' while doing. The more you make the more you know where to look to get the most out of your model/texture. Also things become automatic as you find best practices for yourself - not only modellingwise but also how to approach a project.
Not sure if i understood, do you mean basically rotate the whole model in a viewport to see if I dove too much into detailing?
I mostly trying to use non-destructive workflows, like using quadchamfer and layers with edit poly. Also, I don't think I bound to specific software, I can use Fusion or Zbrush, if i feel it will save my time. But as i said i don't think it's about modeling itself.
Well, I had that habit back in the days, and after realizing that I waste a lot of time to refine proportions after model almost done, I started to make blockouts. The funny thing, I'm still refining whole model on a later stage sometimes, AND spend a lot of time tweaking my blockouts before i proceed to actual modeling.
Now i have a library of various little parts like scews and bolts, and bought a script so i can easily get any of them in my scene. It saves time for sure, tho i believe i have a different kind of problem.
It looks like i take too much time tbh
A was a hobbyist for maybe a year, was sculpting in zbrush decorative elements such as baroque, rococo and other stucco stuff for milling (CAD soft included) for almost 3 years, and game ready modelling for 2 years. I mean there is a plenty of time I spent in 3d I believe. And not just sitting on a chair and receiving a salary but always seeking the ways to improve. And now I'm just stuck.
About deadlines - on my previous full-time job i was working far more than 9 hours each day to keep up with deadlines (even had to stay for a night to finish some projects a couple of times) and even became a lead artist (for a month, huh) but I had to quit since i felt burnt out after a year of returning home at 1 am. And the funny thing that i don't feel that i improved much during this.
Yes, for me decision making is an issue, and now i just don't see how i can solve this. My personal works are bound to be unfinished, since i always refining them, and sometimes i just envy people who post bad/unpolished artworks.
Just kidding, I'm slow as hell too.
Force yourself to stick to the deadline without going into overtime. Stop working whenever it you feel it needs something improved and simply try to finish on time. It's great that someone shows motivation to finish the best way possible to get the best result. On the other hand having your employee burning out or amassing overtime when not required by a specific situation is nothing an employer will really appreciate as it causes issues itself. I am certain that those additional hours you put in don't get you a 20-30% improvement on your model, but rather minor polishing that more often than not only you noticed at work.
Your level is probably still rising all the time, just the steps aren't as big as before. You can take your time and improve on private projects chasing perfection, but I would expect your quality level at work is fine enough with a regular 9-5 job without staying in for another couple of hours where you work tired and burned out. It's part of the job to know when something is good enough for the situation and where you need to put more work in.
I feel that i can relate to it
On the side note, does anyone do hold your breath/ a very slow inhale-exhale pace when you wan't to work faster? I do realize this some time but not sure whether it's a good or bad behavior, or even strange behavior