How fast should I be making models?
polycounter lvl 6
Hello,
I have a question that is very important to me. I worked as a trainee programmer for a small game company. I thought I was good at it, but it turned out, that it doesn't really matter what I can do, but how fast I can do it. And I was really slow, way to slow. In the end I had unfinished daily tasks even from two weeks before and didn't get the job.
My question is, what is the speed for a starting 3D artist that gets the job? I mean for example, how long should it take to make a 15k poly character, or a 8k weapon model.
I would be really happy if someone posted at least one example of what they were making for their company and how much time it took them to finish. And I mean people with small experience, not veterans.
Do companies often split the work, so one graphic does models, other textures them etc. or one guy does a model from start to finish?
Sorry if it was answered here before.
I have a question that is very important to me. I worked as a trainee programmer for a small game company. I thought I was good at it, but it turned out, that it doesn't really matter what I can do, but how fast I can do it. And I was really slow, way to slow. In the end I had unfinished daily tasks even from two weeks before and didn't get the job.
My question is, what is the speed for a starting 3D artist that gets the job? I mean for example, how long should it take to make a 15k poly character, or a 8k weapon model.
I would be really happy if someone posted at least one example of what they were making for their company and how much time it took them to finish. And I mean people with small experience, not veterans.
Do companies often split the work, so one graphic does models, other textures them etc. or one guy does a model from start to finish?
Sorry if it was answered here before.
Replies
No one hires someone to produce crap really really quickly.
If you've managed to hit industry level quality then here are some things you might want to do to speed up.
While I agree taking your time while learning is great and all. When you start working on a tight development schedule, speed becomes a factor. The trick is just practice.
i think the other way around also works, not for getting the job as the results will never be presentable. Basically what i did with Hanno back in the days when we started. And our english wasn't very good (well its not very good yet either, but back then it was bad) we've been watching timelapsed modelling videos.
And really wondered why the artists are alway sooooo fast. So we sat down and trained ourselves, by speedmodelling.
We never reached the inital speed of the timelapsed videos, but we sure gained some speed and control over our work in a short timeframe. We started with doing a head in an hour, later we did stuff in 20 minutes.
The trick is to stop at the timeframe, no matter how it looks and go on to the next. Think about where you lacked something (knowledge, workflow etc.) and try to improve on the next piece. You can really tighten the way you work and you might see problems and will search for solutions find scripts to speed up the slowing down parts of your work.
+1 i thought guitar years ago, when i was in high school, and always got people who wanted to play stuff fast like metal, to do exercise to increase there accuracy, and things like economy of motion.
the speed comes onces it is all part of your muscle memory.
except for racer445 maybe
That is why I suggest sketching out ideas vs concept modeling, rapid iteration to find the best idea vs a long time between iterations. A lot of people will block in some basic shapes in 3D and then sketch out over the top of a screen grab to kick around ideas so its a mix of the two but idea is the same, find out what doesn't work as quickly as possible to find what does work.
I really like the Disney animator quote, I always heard that planning is important, but this showed me how important it is.
That is exactly what I mean. I began my work as programmer, I was trying to make everything perfect and suddenly project leader asks me "are you finished? Our client is waiting and he's pissed off". Then company lost some money because of contract. Client insisted to get unfinished product, so we gave it to him and it went live in that state. So when hired, time is important.
The problem with me is I always need to make everything perfect to every little detail.
I asked about time to make this and that done, because I want to have some reference. So I know that, when I can make this thing in that amount of time, I can send job application. So please, if you could give some examples, I need to know what to aim at.
Thanks for advices.
Like everyone said, practice and plan. The more practice you have, the better you are at knowing exactly how to tackle a model beforehand and make less mistakes along the way.
Perspective.
Keep in mind what the end customer will see. Step back, zoom out and realize that not everything needs to be perfect, striking a balance between good enough and perfect is often the difference between delivering the customer is happy with and not.
"You can't see the forest for the trees".
Realize that when you are building a forest, you are not building 4,000 individual trees all with critters. By all means, make it look good but remember its not the UV's on the knot hole that the customer will see when they look at the forest. If you spend 3hrs shuffling UV's around did it really make that big of an impact? Is the customer going to open it up and complain that you used 94.5% of the UV space instead of spending 3 extra hours improving that to 94.7%? They probably won't praise you for the added optimization but the will complain if its 3hrs late.
Plywood cows and Hollywood sets.
We create illusions and approximations of real objects. We don't factually recreate objects bolt for bolt, branch for branch, atom for atom. Only put enough detail that needs to be there to make it believable, in the way that it will be viewed. Don't obsess about eyelashes if the character is going to be a silhouette on a distant hill.
There is beauty in simplicity and optimization.
This kind of thinking pays out in huge ways in games and allows us to do so much more with so much less.
A smart guy (Perna) pointed out in a long gone thread that often people get their models up to 90% finished pretty quickly then spend the majority of their time tweaking and tuning things that really aren't making that much of an impact. At that point you can walk away and the customer is probably happy with it even if you aren't.
I'm sure other people could give you better examples but here are some examples from where I work.
A few weeks (3-4) for an important building.
But all that's env related, not sure you're aiming at that.
one important aspect, what do you enjoy?
The reason I ask, is that you first tried as programmer, and failed, then come ask questions about art position. It's good to be prepared for what expectations are for any job, but make sure you really are after it. Cause you could just as well with the experience made continue on programming and build up there.
If working on game art suits you more, that's cool, but don't be shaken just by that one obstacle that was too high yet. It's not a total binary lifetime decision, but it will need a lot of effort on your end, so you better feel well from within
You just have to make sure your own quality control is in line with the rest of the game assets.
Thank you for your tips Mark. This really changes my view on some of the things and will make my work easier.
That is exactly what I needed.
Usefull information, thanks.
I was always more focused on graphics and that is the thing I always enjoyed. The case was that there was no job for a trainee cg artist, so I took what was avaliable at that time. It was a mistake, I didn't feel good in there at all and that also had an impact on my performance. I lost this job, so I have a chance to do it right this time. And I want to be prepared.
Thank you for all the support. I read it very carefully and it really has a big impact on my way of thinking.