I'm opening this thread where we can share art tests that 3D studios provide as a part of the job application process. Here are 2 tests for environment position, from Splash Damage:https://www.splashdamage.com/envarttest01https://www.splashdamage.com/content/environment-art-testI realized I probably wouldn't be able to pull off that first one, I don't have enough complicated hard-surface experience and now I can see that. Don't even know where to start nor how many of these details would a low-poly model have. The second one seems pretty simple to me, though. I'm a relatively new self-taught amateur so yeah, it's to be expected I guess.Anyway, I'll also post more when I find some.It would be interesting if we tried modeling some of it and talk about it, be it characters, assets, environments, texturing, lighting, doesn't matter.
One more (you are supposed to recreate it I guess).
>>597908I don't understand the second one, is it like, a sewer or what?
>>597909Is this for Splash damage also? Great idea by the way, I will post any tests here if I come across them.
>>598303Unless that's for concept art, that scene seems to be on a completely different level than the other environment they have listed in the first link.
>>598240>>598309This one? I think it might be a dock of some sorts.
>>597908Looking at the criteria, it seems people still primarily expect baked models for environment pieces like these, but I honestly would just go full FWN and eschew the high poly and bake altogether... I wonder how would they react?I mean, just look at the scale of the character next to that object... what kind of map resolution would be enough, even with aggressive reuse? I’m sure someone with experience can do this well even within the limitations, but nowadays there are more efficient methods, especially with displays hitting higher resolutions.As far as testing goes, they should preferably ask to provide a timeframe or even time-lapse for how long it took to make the piece, because more often than not, the only thing that matters in actual production is how fast you can churn out content, regardless of the specifics of the art style. Any dummy with a 3D program can eventually make the model shown, but if it took you forever to make it, then you’ve still got a problem. The best thing you can say in your letter is “it may not be perfect, but I made this in just [hours][minutes], that’s the thing that is worth anything.
>>598303>>598309No, that one is from this article:https://magazine.artstation.com/2017/03/game-environment-artist/That's some environment from Robert Hodri, Doom's environment artist. He said it was made for some art test, so he probably did it from the concept art.But yeah, that dock or sewer or whatever definitely is a lot simpler. I don't know how old are those tests though, I don't see the date anywhere.You can find plenty of threads about this on Polycount btw. E.g.http://polycount.com/discussion/192820/recent-art-testOkay, this was done for a junior position, as a first job in 3D:http://polycount.com/discussion/189511/my-failed-art-test-the-columnNow I'm scared, I'll have to try making something similar. I mean, guess the only "hard" part is that decoration on top. Those images in the OP are a refined version btw, scroll through the thread and you'll see the original column model they failed him on.
>>598342How would you go about making it?I think baking the highpoly is worth it for the general surface quality to be good without adding a ton of topology perhaps.And i was thinking of maybe doing some of the details as decals or floating geo, that way they will maintain decent quality when mipmapped.red is the lowpoly parts i would make and the green is parts i would try to make decals/floaters for.I think using a detail texture would also be possible.So it would be 1x2k texture for the base shape, 1x1k texture for the decals and 1x512 texture as a detailmap.I've never tried making something like this but does my method sound reasonable?
>>598342>As far as testing goes, they should preferably ask to provide a timeframe or even time-lapse for how long it took to make the piece, because more often than not, the only thing that matters in actual production is how fast you can churn out content, regardless of the specifics of the art style. Any dummy with a 3D program can eventually make the model shown, but if it took you forever to make it, then you’ve still got a problem. The best thing you can say in your letter is “it may not be perfect, but I made this in just [hours][minutes], that’s the thing that is worth anything.Pretty much what I was thinking. I could stumble through this in a day and probably make something up to game asset standard but I'd be in trouble if there was any sort of time restriction.
>>598348Can see why they failed his original model.Maybe it's just because I'm a lowpoly artist, but his redone model seems pretty good. Like way better than a junior slot.
>>598416I’ll have to think on it. The most time consuming piece I can see right away is the base part with the bulge sticking out, because you pretty much have to use subdivision surfaces to make it, and there’s no efficient way to optimize a smooth surface that’s curved in all three axis than just throwing in enough polys to keep it looking round. My guess is that this is the intended stumbling point they put in. If I had any say in the design I’d make them a little more angular and in line with the rest of the object. Maybe I’ll make my own spin on it for kicks.