I want a second opinion on a job I rejected.
polycounter lvl 11
Hi guys, I wanted to ask you what you think about the following situation I've recently been in:
I got a job offer from a new studio to be their character artist, it seemed very promising and they wanted me to make a test first, which is quite common.
A week overdue I finally receive the test, it was a very rough black and white sketch of a Dwarf wearing an armor composed of chainmail, leather and some other metallic armor pieces, plus weapons.
I was told to make it colorful and that they were aiming for a Skyrim like quality, and that the final model should be between 30k to 35k polygons.
This was all of the technical info I received, no info on what type or size of textures I was supposed to make or which files to deliver at the end of the test.
The test had to be completed in no more than a week.
At first I was a little annoyed but after thinking about it for a while I came to the conclusion that making a Skyrim-like type of work, with a target of 35k (Skyrim models are about 16k tri) in a week is next to impossible, specially if you are also hoping to get a game asset with textures and weapons.
The email I was sent also had this little thing attached, which Im guessing I wasn't supposed to see:
They are not paying me, I don't have any kind of contract yet and to be honest that sounds super shady to me.
At the end I decided to explain all of the reasons why I believed this task was unattainable in the most polite manner I could come up with and rejected the job as Im not sure I want to work with people who have this sort of expectations.
So, what do you guys think? Was my reaction too much? Do you think a task like this is doable in a week?
I got a job offer from a new studio to be their character artist, it seemed very promising and they wanted me to make a test first, which is quite common.
A week overdue I finally receive the test, it was a very rough black and white sketch of a Dwarf wearing an armor composed of chainmail, leather and some other metallic armor pieces, plus weapons.
I was told to make it colorful and that they were aiming for a Skyrim like quality, and that the final model should be between 30k to 35k polygons.
This was all of the technical info I received, no info on what type or size of textures I was supposed to make or which files to deliver at the end of the test.
The test had to be completed in no more than a week.
At first I was a little annoyed but after thinking about it for a while I came to the conclusion that making a Skyrim-like type of work, with a target of 35k (Skyrim models are about 16k tri) in a week is next to impossible, specially if you are also hoping to get a game asset with textures and weapons.
The email I was sent also had this little thing attached, which Im guessing I wasn't supposed to see:
Also are we paying for this art test in case the deliverables are useable?
They are not paying me, I don't have any kind of contract yet and to be honest that sounds super shady to me.
At the end I decided to explain all of the reasons why I believed this task was unattainable in the most polite manner I could come up with and rejected the job as Im not sure I want to work with people who have this sort of expectations.
So, what do you guys think? Was my reaction too much? Do you think a task like this is doable in a week?
Replies
just throwing that out there.
a complete game ready model in a week is do-able, but it won't be the best looking thing in the world.
basically, you should show them this:
Think about if all the headache with working with people that dont really know what they're doing is worth it, especially with no or a small contract.
edit: In terms of technical direction, studios will sometimes leave things out intentionally. When you come back and ask for additional information, it shows that you know what you actually need to get the job done and are able to open lines of communication to do so.
I would recommend anyone who ventures into any start up. Do NOT do a few things.
A.) Relocate.
B.) Take it serious as an established studio.
C.) Listen to anyone who speaks from a hot headed ego.
Lastly, for anyone doing work outside of personal interest, or Mods... Always play this in your head when you feel iffy.
[ame="
Technically I've already rejected the offer, but I'll see if I get some kind of positive reply on my feedback and manage to negotiate something.
really ?
in my experience retopology of a highpoly to create a 35K lowpoly model takes considerably more time than retopology to create a 1k model. the different could be anywhere between 4-8 hours to a day or two.
also, texturing something at 4k takes lot more time than something at 1k given that you maintain the quality level at 1:1 pixel ration.
It's not a myth as I did a test just like this but even more impossible. Granted it was only one day of my time wasted, not an entire week. Maybe it was a dick move, but they seemed nice, and I enjoyed the challenge.
On whole I turn down more tests than I take because I can't justify the time when I have paying work to do.
I think the one week deadline was a bit tight to create a character, but that's very much dependent on experience and skill level. If they were looking for a senior artist who has maybe 3+ years studio experience, then this could be considered a reasonable deadline. I know seasoned artists who have created high quality game ready characters for AAA titles in half that time, but that seems to be more common with outsourced freelancers.
At the end of the day though, if this didn't feel like the right job opportunity for you and something seemed shady, then you probably made the right decision skipping this one. I hope you find something more suitable next time
EDIT: just to note, the half week character creation time I mentioned is highly irregular and studios should not expect you to do this without extra incentive and reward on top of your basic fee. Even with a week to create a high quality character will most likely require a lot of crunch. It would have been more reasonable if they gave you 2weeks to create that character imo.
One thing I've learned over the years is if something doesn't feel right with a job/contract, it usually means it's not worth pursuing. Every time I've gone against an initial uneasy feeling I had, it ended up being a waste of my time.
I did, they say that if I had a symmetrical layout I could use a single 1024 texture map, for a character with a 35k tri count, full armor and weapons...
Im starting to think they don't really know what they are asking for at this point.
Im sure they have the best of intentions, but I've been in a few projects that never were and Im getting pretty tired of that, if Im putting my time into something I want it to be released.
Also, I wouldn't want to be the guy in charge of rigging a mesh with such a high polygon density, not to mention the CPU overhead of all those needless vertex deformations.
Can you find out whether any of the HODs have any shipped titles? Are the directors game veterans?
Sometimes being knowledgeable in an inexperienced start-up can be a real meal ticket, if the management are receptive - and you can find yourself quickly climbing the career ladder.
If they're stubborn enough to think they know what they're doing whatever you advise, your job will be hell on earth. I've just gone through exactly the same thing this summer.
I'd find out as much as I could about the company and it's employers. If that info is not out there, maybe there's a reason for that.
texturing aside, if my target mesh is 35k, then i would also assume there's a high poly to be made too... if you work smart, you'll be working in a "backward compatible" subdivision style that will allow for extremely fast retopo, that keeps all of the form of the model anyway.
It's all down to planning.
* The specs of a model do make a difference for time sake.
* I've never heard of a studio giving an impossible task to see how you cope. That doesn't tell a hiring manager anything about your skill, and I'd know, I've been hiring people for over 12 years and the test you describe to see how they cope is pretty meaningless.
* To the original poster, this test was a sham, and the studio seems like one as well if they don't know the very basics of art creation and they sent you a lame concept that someone there cooked up. If they had sent you a concept that was of high quality, as though a concept artist had done it, and offered to pay incase they used it, that would be better; but some napkin drawing just shows a lack of experience and commitment.
i would be interested to see your workflow.
this "backward compatible" subdivision style doesnt really work for efficient topology for ingame model.
the way a highpoly base mesh is created for sculpting should really have very little to do with how the realtime ingame mesh topology is needed. highpoly basemesh is usually all in separate chunks while the final lowpoly is more like an envelope cage covering the silhouette. this envelope lowpoly mesh cannot be created by just using a lower subd level, it has to be created with retopology for the majority of it. at least this is what has been done in most of realtime game character out there till now.
this "backward compatible" subdivision style modeling you are talking about is more appropriate for vfx and film where every mesh element is separated in chunks so that once the highpoly is done you can just take a lower subd level of your sculpt, optimize it a little and call it a day.
this is rarely the case in game characters. may be you will see it in cinematic character, but very rarely in game character. and if it is done it is not an efficient way of making lowpoly for current gen.
even if you do use the lower subd of the highpoly, it still take more time to create a 35k tri lowpoly from a highpoly sculpt as opposed to just doing a retopo to 1k triangle. retopo to 1k triangle usually would not take more than 1 hour while creating and optimizing that 35k tri model for efficiency would take at least 2-3 days.
from personal experience, all i can say is that it takes lot of time to properly create a realtime model with polycount such as 35k vs a 1k model. of course there can be exceptions, but generally it takes LOT more time.
*EDIT* almost forgot to reply to the original poster, sorry
so i have actually denied 2-3 tests that are close to your situation, with the exception of them providing a texture specification.
regarding the expectations from a test, it should always be ok to expect high from a test HOWEVER, the test should never be a full asset with that high spec so that it can be used commercially and specially if there is no payment. i think you made the right decision to deny the test.
that being said, you should have asked them about payment option in case they just overlooked it or it got lost in communication between HR and other managers.
If anything I'm overly specific when setting briefs for tests, mostly to see how well they can listen. If there's something they query it's a good thing, good communication is a two way street.
Other skills aside I could argue general anatomical proportions against a bust, but it being a dwarf character...
Well, when I got the test to enter Gameloft they did require the files, but it was clearly a model they weren't going to use anywhere and the test itself was definitely doable on the time provided (a 6500 tri model in 10 days, no big deal).
it's common, yes. done it before.
I think you did the best thing you could in that situation, sure you're always going to have to crunch now and then but if you show that you are willing to do 16 hour days during your art test they're going to expect it all the time.
It makes me feel all tingly inside =]
As others have stated, you made the right move Linfo and you've learned a bit from this experience as well. Good luck!