Getting Training For Characters - Reliable Options?
interpolator
So, this year, I had been hoping to get some good training. I want to make character art, so I want to learn more about that. Obviously, that's anatomy, proportions, etc that I'm still weak at, and more technical things too, like, hmm, specifics about UVing, or how many texture sheets to use, where when and why, and various other things I can't think of on the spot off the top of my head.
So I've been looking around at options, but now I'm just more stressed out trying to work out what's best. There seem to be various places offering courses online, such as Vertex School or CG Spectrum, and then there's other options like mentorships, like this one or this one, and various others aside.
On top of being unsure about that, there's also that character art in general seems very difficult to break into, and the industry's current condition being what it is, so even if I become sufficiently skilled this year, I'm doubting I'll be able to make any headway for a while.
That makes me think that rather than trying to do something that says it'll teach everything, maybe it would be better to go for something that just triples down on a few things, like a mentorship for the sake of accelerating my learning of anatomy to get to a point that's a little more solid rock and a little less plaster.
Basically, I just don't know which way I should be going, so I'd like some advice. Which things are reliable? Which things are full of crap? What don't I know that I don't know when considering these things? I've thought about it a bit, but I'm just not sure which is the best way to turn.
I'm talking about various options for training in general; the ones I linked there are just examples, rather than an exclusive list of things I'd consider. I want to do something that, if possible, is time, cost and skill efficient. I'd be grateful for any advice, because right now I'm just not sure what my move should be. Plus, I don't know which ones are reliable and which ones are full of crap; I know for example that I've seen Vertex School be shilled for by bot accounts.
This discussion has been closed.
Replies
Some of those schools might give you a big leg up on networking. But to (potentially) go in to debt for a slim chance at a shaky job seems pretty risky to me.
Just checked your sketchbook and now I remember some post you made before. It looks like you have been doing zbrush anatomy sketches for a year. And a year ago they looked fine. It seems like you have the artist curse where you perpetually think your work sucks. Except it doesn't. Perhaps you need somebody who has some sort of esteem in the industry to tell you that. Hopefully somebody will.
I'd say stop with the zbrush sketches and make complete game art characters and work on getting over your perfectionism. As an artist you could be putting content into games today that gamers would be drooling over. But not if you won't allow yourself to do that because you are fussing over perfectionism. It may help to set time deadlines and sacrifice your quality to meet the time deadlines. This will help teach you how to make judgement calls and stay productive and get out of a rut.
Remember that art is for the artist to enjoy. It doesn't have to be self flagellation. It can be like eating cake.
I'd consider that pretty quick pace but with your current skillset I would expect that you could output a satisfactory AA quality character on such a deadline. If you can't, I would not think it is a skill issue but rather a focus issue.
maybe start with genre; sci-fi, fantasy, realistic, etc. Art is communication; what do you want to convey? What is your character's motives or goals? What timeframe/setting? Personality? Rich or poor or it don't matter? Answering these questions can help you narrow down your character's visual appearance
How has it been working for you?
Anatomy is difficult to get right.
My philosophy is while it is certainly useful to get skilled at creating anatomy, maybe its best just to get used to sculpting to a point where you can see and compare differences between your work over a period of time, to train your eye so to speak.
Also have you asked this question on Scott Eatons forums or to Scott Eaton himself?
Scott Eaton conducts courses for studios where he routinely criticises anatomy in their published games, so I can't really say if anyone hired is actually perfect in anatomy, or their perfection and knowledge of anatomy got them work (unless they were perfect and got worse when they were hired)
He must have covered Gallery Abominate, that's seen as a good way of learning and also realising that no-one is near perfect at capturing anatomy correctly.
I expect you will be using Marvelous designer for clothing and shaders would start with substance painter for textures.
While character creator (the software I recommend) would cover a lot of ground, you could also look into the metahuman workflow.
Also realistic characters are a challenge to get to a suitable level of believability, and since you're likely to start in the industry as an associate, maybe its better to focus on clothing and associated props for characters while you practice anatomy.
While I certainly am continually improving on my character art, I did give myself plenty of leeway initially when it came to sculpting different bodytypes.
I think Scott Eatons course does touch on this aspect. though its more defined on anatomy practice.
For general sculpting practice I would recommend capturing the look of disney models and keeping it very true to any concepts. They have very clean shapes so this should help with getting more comfortable with sculpting.
The assignments you've posted in your sketchbook do look good!
Clothing simulation does require a proper base model, so that's where anatomy comes in but with focus on larger anatomical forms.
This is why the emphasis on smaller forms and muscles is variable depending on how much of the character is clothed and the overall look , not to mention what players will actually be more sensitive to, Somethimes this is lost on the artists who try to perfect regions that should be low priority, hence there are tech solutions to bring uniformity.
In my experience, there was a seperate team for clothing and the work was mostly limited to texturing since most of the clothing simulation was procedural and done in real time. This is on FIFA/FC.
For dragon age, mass effect, iron man, character artists handled clothing, though majority of the clothing and prop elements were outsourced, the character artists mainly put them together in the studio and apprised outsourcing of any changes.
It was always outsourcing first, then development in studio and tech art solutions were always encouraged over traditional workflows.
Tech art solutions are being developed to reduce dependency on artists regardless of whether they are in house or outsourced. This doesn't mean artists will become redundant, just more that they will become more QA and specialized.
Usually at the studio you do get freedom to decide your workflow, though that is secondary to established pipelines.
Also this approach is something I've seen more in corporates since their focus is profitablity with high turn over to reduce cost.
There's a lot of incentive to enterprise.
Might interest you to read this post I shared with articles by Dennis Detwiller on his experience in the game industry and what he learned from the business.
https://polycount.com/discussion/234911/dennis-detwiller-on-escaping-video-games-and-entrepreneurship/p1?new=1
For character art, once you have the skill of sculpting down you can either apply to the game industry, film and television, toy industry with your portfolio.
Working in AAA certainly builds credibility for future endeavors, but in its current state it clearly isn't as stable as it should be. (given its growth)
From a design standpoint, you could crowdfund your idea and that does require more research and development including an understanding of marketing and self promotion.
Creating art that appeals to an audience is really the first step, and by appeal I don't mean perfection, just something the audience can resonate with, and honestly it could be anything.
There is merit in building a different twist with what works than risking it all with something truly original. But you have to try to see what the outcome might be.
someone else might say it's nonsense and that every studio they worked at used base meshes
if the goal is to get a job i think that it makes sense to try the path of least resistance first and only upgrade your firepower if that doesnt work
in other words, use every possible shortcut to make the best looking characters that you can, put your portfolio together and apply for jobs. if you get one, great you did so in the most effective way.
If feedback on your portfolio is telling you over and over that your anatomy is fucked up then you know that you missed an important step and can fix it.
Very rarely did a character artist in studio have to work from scratch and if they were assigned such a task there was considerable time alloted for it.
Lot of the work was outsourced, so its more a matter of how to finish a outsourced model than build it entirely by yourself.
The reason studios do this is because it is cheaper and there's a lot more work to be done on an outsourced model with regards to material assignment and integration.
For portfolio its there is more emphasis on showcasing a knowledge of fundamentals, and its okay to use a base mesh and while you don't need to necessarily mention in the portfolio that one was used, during an interview you can go into the details of your workflow.
I understand your fear about not having a basemesh to work from would make things problematic, but on the job for sculpting work where you start from scratch the team would evaluate your ability and see if they can budget for the learning curve you would go through to do such a task.
For example, suppose they want you to sculpt a small creature from a concept with no base mesh available so you're starting from a sphere, this would factor into the time you would need for the task amd your skill level to do the task.
I would also recommend having your portfolio evaluated by artists working at studios you want to join. Its a good way to build connections to.
This is what I've concluded as well while I'm working on my own character portfolio. No one would know or even care whether or not you use a base mesh since most people in the industry do and ultimately everything you sculpt is going to come from some kind of base object anyways. I personally say find the best base mesh you can and go with it. In your spare time just practice anatomy sculpts. Its what I do.
https://polycount.com/discussion/235382/reach-out-anytime-database/p1?new=1
That makes sense, practically. Perhaps it is just a complex.
" This is what I would recommend you and anybody who is trying to join the industry. Jumping into the industry as a character artist is hard, not impossible, but simply you are setting yourself up for one of the biggest challanges. Why make something that is already hard, harder? I would much rather recommend you center your portfolio around 2-3-4 GREAT props, like top level, and that is going to open way more doors than any good characterThink about it this way, if you are a big company, would you rather trust somebody with no experience to make a gun, or rather let them work on a character that has to animate and has to have way more technical knowledge? Also, another thing to consider, once you are inside the industry you can much easily switch positions. These jobs rely a lot on connections and recommendations, rather than open positions. So, in summary, work on great props, small pieces, and continue working on characters for yourself. "
also i saw your sktechbook and again i think you did the same mistake i did , i spent too much time on anatomy and guess what ! most of the characters you see in games are not naked ! like 90% of a character is cloths , gears , props etc .. i know it's frastuating ( well for me at least ) because i enjoy doing sculpts and organic stuff , but i think the safest and quickest road to get into the industry especially for somone in our age is doing hardsurface game ready props and be so good at it . and after u get a job u can learn doing character by yourself while also learning from the studio .
Maybe check out the past "Quarterly Character Art Challenge" threads (it was bi-monthly at some point iirc) and consider doing something in that format yourself. Since characters can be complex, it would be good to get familiar with all steps and go through the process several times. Given the 3-month timeframe, you would also have enough time to flesh out the specific anatomy. If you followed that template, you would create 4 characters a year, some better than others, or unfinished - still lessons will be learned and it's a push towards getting something deforming and in engine.
But a challenge from the past could be done today too :-B
Maybe you just need to do a character that looks really hideous, because then it can only go better
But seriously, the most I learned from doing those challenges is too not care/obsess as much and just getting it done within a certain timeframe. Even if the result is not great, it's then something to work with. Based on my experience, on a production iterations are normal (oh the design changed we're doing this now, oh design doesn't fit the lore anymore, ...).
Outside of current market realities the main barrier to character art positions on the employer side is that they don't have a need for as many character artists as they do for other departments, but it really comes down to what constitutes a character art team.
Majority of AAA studios have established pipelines to take care of anatomy and other repetitive tasks, so your fundamentals help in navigating this workflow and in being resourceful.
Really depends on the job and responsibilities which should be clear from the listing, and it does help to get in touch with artists currently working on those teams.
Personally I wouldn't get obsessed on perfecting anatomy since reproducing it is a skill that is life long learning and studios automate this part of the character creation pipline exactly for this reason.
A skill that I think would also add value is valuable is data entry and management since a lot of art work in studio is reviewing outsourced assets and repurposing from asset libraries.
I believe transferring a design into a believable thing in 3D has a significant craftmanship/technical component, and without going down that path over and over again, how does one find the challenges and solutions. Probably a crucial result of this repeated process is getting the confidence in ones problem solving, at least the knowledge "I did it once, I can do it again".
For the moment most portfolio advice is centered around first impressions and demonstrating fundamentals which doesn't really assess for resourcefulness and resilience that a demonstration of iterations and revisions provides.
I get the feeling it's a common problem. You can only show substance if you can get people paying enough attention to take the time to see it. So you need something really flashy and quality in the first place or nobody will give you that much time.
if you spend too long looking at any realistic style 3d art it always looks like shit because the comparison is real life and the tech just isn't there. So you have to accept it the way that it is once you reach end of deadline, and then on next model you can try out new ideas to do a little better in the time you have.
Though you do get to go more into this during an interview since it is seen as an asset in production.
Well ... where are the pictures ?
Alright, 3D character art, then, if you need the specific. Character models.
There's also more emphasis on the high poly since automation takes care of a lot of the process to make models game ready.
And a greater focus on hair which is challenging to get good at.
Did you look into the reach out anytime data base?
You can schedule free 30 min one on one's with senior artists in the industry who are volunteering their time.
https://polycount.com/discussion/235382/reach-out-anytime-database#latest
... That's pretty much the last thing someone interested in learning character art should consider.
The point is not to learn this or that ultra-specific workflow from this or that studio.
And @Rima yes of course, I meant images of what you've been working on
Using a base mesh is definitely sensible later. It's not efficient at all to start from scratch every time. I don't want to rely too much on it, though; I need to be able to do any given part of it myself. Even if there are automated processes available in work in some places, I can't let myself rely too much on those. If I can only perform adequately when a machine is picking up all the slack, it's not a convenience, it's a crutch.
Yeah, I'm just going to have to nut up and do a project even if it turns out to be dogshit. I'll do that.
I do think that since a lot of time is allocated to learning pipelines at the studio, it may be advantageous to streamline the more repetitive aspects of a workflow provided there's a good grasp of the fundamentals.
Like with the workflow I follow, I managed to allocate more time to detailing, refining and finishing, reducing tasks like retopology, rigging and weight painting to 10% of what it used to be before. And it replicates a pipeline I used in studio.
For artists starting out I'm thinking that learning character art and getting hired are two seperate things, and you should atleast satisfy the minimum requirements in the job listing.
I've not seen the advice of having "portfolio artwork match studio quality" to be applied consistently, or atleast that isn't the only factor considered when it comes to hiring.
Does help to keep learning and refining your process to create artwork which hopefully is applied on the job at some level.
For example, with the 2 realistic characters I have, I've tried to be very thorough with breakdowns, but even something like this is considered acceptable for an application,
https://www.artstation.com/artwork/rJNkOa
and I assume the recruiting team would go more into details, creation and breakdown during an interview and utilise a candidate for their strengths in a team based environment.
Definitely helps to have a base mesh since it takes care of your topology and you can maintain consistency across a range of models.
At EA, base meshes were created by the character technical artists with input from character artists since they do have to hold up to rigging and animation considerations, so its a collaborative effort.
Also I wouldn't call automation a crutch since its a very budget/time conscious approach by a studio which I feel character artists looking to join should try to replicate.
Interview wise reviewers do put emphasis on a knowledge of fundamentals, even if that knowledge isn't applied directly 90% of the time, i.e your task isn't to build from scratch but to jump in at any stage which requires good knowledge of the whole process.
Hey , glad to see someone on the same boat as mine ( im also in my late 20s and i started learning 1.5 year ago ) , im afraid you are making the same mistake i did so im gonna make a conclusion of all the advices i took from other people who work in the industry
" This is what I would recommend you and anybody who is trying to join the industry. Jumping into the industry as a character artist is hard, not impossible, but simply you are setting yourself up for one of the biggest challanges. Why make something that is already hard, harder? I would much rather recommend you center your portfolio around 2-3-4 GREAT props, like top level, and that is going to open way more doors than any good characterThink about it this way, if you are a big company, would you rather trust somebody with no experience to make a gun, or rather let them work on a character that has to animate and has to have way more technical knowledge? Also, another thing to consider, once you are inside the industry you can much easily switch positions. These jobs rely a lot on connections and recommendations, rather than open positions. So, in summary, work on great props, small pieces, and continue working on characters for yourself. "
also i saw your sktechbook and again i think you did the same mistake i did , i spent too much time on anatomy and guess what ! most of the characters you see in games are not naked ! like 90% of a character is cloths , gears , props etc .. i know it's frastuating ( well for me at least ) because i enjoy doing sculpts and organic stuff , but i think the safest and quickest road to get into the industry especially for somone in our age is doing hardsurface game ready props and be so good at it . and after u get a job u can learn doing character by yourself while also learning from the studio .
I see what your saying, but disagree with most of it.
Your not taking into account something that matters a lot to employers: interest. They're going to take into account how much you like doing the things your going to be hired for and they often ask you outright during the interview what is it you like to do. I know this from experience because the 2D world is similar. I made a somewhat general portfolio several times and every single time the very first question I was asked was "so what is it you like to do?". When I told them character design, guess what they all responded with? I wanna see more of that, not all this other stuff your not interested in. With 3D its the same way and frankly this is true for all jobs.
Basically....they're not going to hire you if you hate doing the job before you even get it and if you do manage to get it anyways, at some point you'll be let go due to lack of engagement. Now your back to square one.
As far as props being "easier" to get in with....I disagree as well with that one. Whats easier to get hired in is what are you good at doing and what you like doing. What if OP is horrible at doing props yet great at doing characters? I think the easier route for him is going to be characters.
There is nothing wrong with doing characters and plenty of people get hired in their first 3D job as a character artist. You simply need to get good at it and market your stuff to the right people, pretty much as you would for any kind of job.
a: there are definitely more jobs making props than there are jobs making characters. so... if your goal is to get a job ...
b: most characters are covered in clothes and props
on the other hand though ...
are we interested in getting any job or are we interested in getting a job as a character artist?
There's a big difference between modelling for self improvement/fun/etc and modelling for work.
Making lovely anatomy studies achieves two things
1: teaches you anatomy
2: shows potential employers that if they ask you to make something you'll be able to construct it properly because you have solid fundamental skills
it doesn't show anyone that you can generate a complete character - which is what the job is..
When you're being paid your job is to generate an end result using any and all resources you have available - the process is not the product.
are we interested in getting any job or are we interested in getting a job as a character artist?
This is the key. If one is after a character artist job, they should be doing character work. Character jobs are fewer sure, but then again so is the number of people capable of getting those jobs and also because its difficult to do, many people drop out before getting to the finish line so thats all the less people you'll have to compete with if you keep going. Props are generally perceived to be easy work and that leads to a lot of people to do them, so you'll inevitably have a larger pool of competition to compete with that can also do the job more effectively, just as you can.
Ubisoft Toronto has Model artists, but at other studios its more specific like Weapons Artist, Vehicle Artist, Hair Artist, Clothing Artist, and sometimes its just Character Artist with the description mentioning that the Artist should have a focus on clothing etc.
I think more than portfolio and skill its really down to networking when you are looking for a job, meaning certainly you need portfolio and skill but it absolutely helps to know people on the inside who are put your PPR in the spotlight.
This thread was interesting
https://www.linkedin.com/posts/richard-k-glass_most-applications-arent-even-looked-at-anymore-activity-7189026187900346368-cU0v?utm_source=share&utm_medium=member_desktop
lot of advice in the comments too.
As far as learning and demonstrating character art goes, there really isn't a ceiling, but a huge disparty in what passes for production work in studio and the standard an artist sets for themselves in personal work.
It can become necessary to pivot over to other opportunities to fully use your potential and many are compelled do it as the only means to increase in seniority alongside the salary bump even if the responsibilities of the new position don't necessarily correspond
Not every position that a studio relies on to operate is listed, therefore you can't infer how studios work just by looking at their job openings. Such openings are, by definition, for the roles that need to be filled because a studio doesn't have enough staff currently available for them - nothing more, nothing less.
"I think what I'll probably do is decide on a project and when I've blocked in all the character's basic anatomy and shapes that make them human enough, retopologise that and save it out as a base mesh for future use before I start adding all the things that make it individual. Assuming it's actually sufficient quality to bear reusing."
This sounds like a very convoluted thing to do when starting out, and a waste of time and efforts IMHO. All you need is a regular polygon modeling toolset + something to create textures with (any 2D painting software would do) + a few weekends or a week worth of free time. Just because Youtube video tutorials are all about highpoly sculpting and retopo and Substance Painter doesn't mean that it is a good way to learn. The lower the specs you use and the more limited the toolset, the faster will you be able to iterate (and learn).
Anyways - for now there is no way for this whole thread to be productive since you haven't provided anything that would help people tell where you are at in your learning path at this time. Depending on your current skillset (both technical and artistic) the suggestions could be very different.
I am not being ironic or sarcastic here. If you are judging what to learn and how to learn it based only on what *appears* to be the norm to you from the outside, chances are, you'll likely be wasting time. Because by definition, as a beginner, you can't guess things.
I can gurantee you that someone learning by doing great models the "oldschool" way and pumpin out, say, 30 of them in a year (also implementing a few of them in game) will be miles ahead of someone spending months on sculpting, retopo, and substancepaintering a "highend" model - both technically and artistically.
Just recently I've seen people pick up Blender from absolute scratch in order to make their own custom characters for Bomb Rush Cyberfunk - a game based on Dreamcast era specs. After about a month or so, they end up knowing how to model ingame meshes, how to use UVs both the regular way and the palette/tiling way, paint skin weights, import to Unity to apply the Humanoid rig, and learn a third-party toolset to implement their characters.
They may not know how to push millions of polies in Zbrush, but that part is by far the easiest to figure out anyways ...