How long should it take to complete an environment asset from start to finish?
polycounter lvl 7
Hey guys, so today has been a really, really bad day for me.
I am a little over a year into my game art education, and today I was told by numerous people in the game industry in Denmark, (my teacher among others) that I am much, much too slow at creating environment assets to ever even be considered for hire by a game developer company. It was soul crushing to say the least, so I am turning to you guys for help.
I got this feedback after a 3 week project, 15 5 hour work days, where I in total made 8 different assets for a small unity game. Stuff like Mushrooms, stalactites, geothermal vents, rocks and ground terrain, nothing majorly complicated, but all stuff I hadn't previously tried.
So my question for you professionals out there is this, how long does it take you to create an asset in AAA quality, model and texture, from start to finish?
Almost every day I see people who succeed in finishing beautiful assets with a sub-title like: "2 hours spent in total", and I just dont understand how it is possible. Obviously through practice, but is this an industry standard? I feel so upset and hopeless when I see this sort of stuff, I can't see a way for myself to become as good as this.
Should I start timing myself? Simply start a fucking clock and have it go off after 2 hours and stop there? I really dont know what to do.
For those wondering, my workflow is this:
Model Lowpoly
Sculpt Highpoly
Unwrap Lowpoly
Polypaint texture
Bake Normal/AO/Cavity maps either in xNormal or zbrush.
Import into unity and see result
Create specular map from diffuse, sometimes gloss map.
Is all of this really supposed to be possible within 2 hours time? Do I just suck?
Sorry for the venting, I hope atleast some of you can relate and give me some pointers as to what I should do next.
Regards,
Nikolai.
I am a little over a year into my game art education, and today I was told by numerous people in the game industry in Denmark, (my teacher among others) that I am much, much too slow at creating environment assets to ever even be considered for hire by a game developer company. It was soul crushing to say the least, so I am turning to you guys for help.
I got this feedback after a 3 week project, 15 5 hour work days, where I in total made 8 different assets for a small unity game. Stuff like Mushrooms, stalactites, geothermal vents, rocks and ground terrain, nothing majorly complicated, but all stuff I hadn't previously tried.
So my question for you professionals out there is this, how long does it take you to create an asset in AAA quality, model and texture, from start to finish?
Almost every day I see people who succeed in finishing beautiful assets with a sub-title like: "2 hours spent in total", and I just dont understand how it is possible. Obviously through practice, but is this an industry standard? I feel so upset and hopeless when I see this sort of stuff, I can't see a way for myself to become as good as this.
Should I start timing myself? Simply start a fucking clock and have it go off after 2 hours and stop there? I really dont know what to do.
For those wondering, my workflow is this:
Model Lowpoly
Sculpt Highpoly
Unwrap Lowpoly
Polypaint texture
Bake Normal/AO/Cavity maps either in xNormal or zbrush.
Import into unity and see result
Create specular map from diffuse, sometimes gloss map.
Is all of this really supposed to be possible within 2 hours time? Do I just suck?
Sorry for the venting, I hope atleast some of you can relate and give me some pointers as to what I should do next.
Regards,
Nikolai.
Replies
There is no generic asset completion time, it's always going to depend on the individual asset/task, within reason.
When your a beginner don't worry about speed, if your rushing that early on its just going to turn out crap. I don't expect my juniors to be as fast as the seniors on our team, no where does.
Speed comes with practice and experience. Focus on quality, but be realistic. Obviously if your wasting time on facebook or tumblr then turn off your internet and buckle down.
You could always show us the stuff you've made, peeps could give you an estimate on them then.
Your first goal is to get an asset done. Second, remove all bugs, because buggy asset is not finished.
I think that's why people may complain about your time. I just guess, of course. I really don't like the moments when an artist agonizes before me that (s)he will have to "do everything again!". This won't be a problem, if (s)he did it quickly or used an easier method.
Some become so attached to their beautiful wireframes or perfect strokes in their concept, that they find it hard to start again
That's why programs like Substance Designer become so popular recently. That's also why you make an levels from modules, not every polygon from scratch. It's for making changes, iteration, cheaper.
So don't add detail by hand if you can automate it. Don't overthink things. Just go. If something looks good, doesn't produce visual errors (ask others for final review), then it's done.
Of course, it's always a temptation to correct these quads, tweak a bit these colours... so obviously I understand you
How to make things faster?
"Model Lowpoly" - maybe model just a quick base? Then sculpt, if it's faster for you define shapes by sculpting. Only then retopo the sculpt.
"Sculpt Highpoly" - avoid some of it by generating fragments of normal map from bitmaps. NDO, DDO, Knald. It's easier to re-generate a fabric pattern than to re-sculpt everything. Even in ZBrush, sometimes a noise from a pattern is enough.
"Unwrap Lowpoly" - try some plugins like TexTools? can't avoid hard work here really
"Polypaint texture" - learn Substance Painter or Mari, so you will be able to paint with layers of tiled textures, make corrections easily (not by re-painting)!
Don't always think of a time per piece! Because it's often more important to prepare a good library of tiled textures, zbrush alphas. To find a method. First rock will take you 6h, but then you will be able to do 30 textured rocks in 1h each.
Perhaps I should start doing more small projects, setting a strict timeline for myself and uploading the end result for critiques from people on this site.
"should I start timing myself?" yes...use a stopwatch. Sop it when you cease working on something not related to the asset- this includes taking a moment to browse the internet, stepping away..etc. This will allow you to see how much time you are actually spending on the asset.
When I am pressed for time I find it helpful to do 20 minute sprints with 10/10/15 breaks. So, work 20 minutes, 10 minute break, work 20 minutes 10 minute break, work 20 minutes 15 minute break. When you take your ten minutes, step away from your computer- have a snack, do some push ups..whatever
Clicks add up. Are you digging for tools in your 3d app? Make sure you know the hotkeys for your base toolset, or any tool that is going to be used extensively during that work session.
You see someone post..finished this hyper accurate Ferrari during my lunch break today, ask them to breakdown their workflow in terms of time and test yourself against those numbers.
"basemesh"
Sculpt
lowpoly
-don't restrain the silhouette of your sculpt to your lowpoly...build your low-poly around your sculpt...not your sculpt around your low poly
good luck!
There is no ''timeline'' for each asset, it depend on a lot of variable, I won't repeat what others said, but I agree with most of it.
Piece of advice:
Don't work 15.5hours a day. It took me a long time to understand that. Your not gonna be productive. And as you see it hurt your self esteem.
AND: When people post on forum awesome work they made in 2minutes, is most likely not true.
You say you worked 15.5hours a day for 3 weeks: That's 40.68hours per asset, Indeed long time if is a static object with no behavior, again it depend on the quality your looking for, I have no clue.
Don't beat your self up, everyone goes through bad phases, especially while learning.
Things to remember when comparing yourself with what you see on the internet:
- Everybody lies
- People are nearly as bad at remembering how much time they spent on a task as they are at estimating it in advance
- They may not have started from scratch
- They may already have practiced a particular workflow for creating that asset many times
- Yes, *some* people are just really that good
- They could have been doing that type of things every day for many years
- Everybody lies
And you probably know the apocryphal Picasso story: when asked why he demanded so much money for a 5 min sketch, he replied that it did not take him 5 min. to make it but 30 years
Well, everything turned out fine for me anyway. (At least so far, it is only 3 years ago)
Just study, practice, work.
Each asset will require some learning if you haven't done it before, that is probably the biggest factor for how long it will take to complete. For example it took me a couple of months to understand and implement the workflow I used for making a weapon model, The next one took about a week to complete and looked much better because of techniques and fuckups I learnt from the first one. A lot of people focus on one discipline of game art and iterate on their projects so they can perfect their techniques and work very quickly.
Timing yourself might be a good idea if you've been told you're not working quick enough. You can build a library of how long it takes for you to complete a specific task and use that to factor an estimation of completion, finding where you need to pick up your pace. A friend told me you should multiply your estimation by 3 and then you have a realistic idea of how long the asset would take to complete. But every asset is different and estimations are usually off. just be proactive with what you have learnt and don't over compare your production time to others.
I second that
When you have a deadline or are working in a studio you should know when something looks good enough to submit. Don't be anal and don't overpolish. When you do portfolio work it should just look as good as possible imo.
It's good to look at another artist's work because it's a great way to find motivation. But don't start comparing yourself to him/her as you have no idea how experienced they are. And yeah, everybody lies, haha ^^
I'm going to try to stop comparing myself to people who no doubt have much more experience than I do, and just keep on trying and not let myself feel down.
For clarification, It was 15 days, 5 hours a day that I was working on the project, not 15.5 hours spent per day, so thats better I guess
Thank you all for the feedback, I much appreciate the support and insight.
Time to get cracking on a tileable brick wall as it is a must have in every environment artists repertoire, perhaps after that a box or even a barrel!