I know the answer depends on how complex the model is and whether you have a good model sheet or not and skill/experience, but what is the general time frame studios general give you for. Ill list some images and you guys can give a approximation on each one. I want to get jest on how fast I'm expected to work. A bonus would be to also give an time frame for each phase of the model modelling, UV unnwrapping, Texturing.
I'm finding hard surface to be quite a tedious process. I'd be flabbergasted if you were expected to complete models like these from scratch in less than a week.
Replies
lol yeah likely none of these have been produced within a week. i mean under ideal circumstances, nobody giving feedback and you just do your thing, you can do things quicker. but in production with, lots of constraints, dependencies and oppinions, you can expect any of these to take weeks to months on the robot or the character. the scene is "just" a highpoly model so it will not take as long, the gun itself, maybe 2-3 weeks from blockout to textured lowpoly. in general, the bigger the production the longer you can expect things to take.
just my experience as a soloist, but there is hardly ever a point where "the model is done."
i think when the game is finished, then maybe you could say so. I dont imagine it is any different on a team, small or large.
The character for instance - maybe to get an initial pass you could have it ready in a week or maybe a couple long work days. But a million things outside the geometry and textures of the character will change, and many times that will mean you have to update the character as well.
Probably on a bigger team, individual parts of the character are made by separate people too.
So I think if you can achieve the same level of quality, time wont be a concern. To get high quality you will be doing many redos, and every time you do that you'll pick up speed naturally just because you'll learn how to automate/simplify anything that you can.
The more times you do a similar thing, the faster you get at it. Hence, a specialty is developed.
The numbers in this image arnt to far off. Money = Time
what the..... You telling me I could sell a big studio a model of a traffic cone for $4000?
That is not how it works but effectively yeah, could happen
Traffic cones are $3600 according to the image. I can do them much cheaper if anyone is interested - for only 1600. Pay one get three for free.
"The numbers in this image arnt to far off. Money = Time"
Well makes sense,
Recall, when GTA5 was released total production costs were initially speculated to be around $250 million...
We modeled a car where the cg one cost more than the real one. It was a small cheap car but still.
Assets arnt "final" only if the geometry is ready.
yes the traffic cone is not a standalone item. It is likely one tiny piece of a larger puzzle. So the cost of the traffic cone is not only the guy who moved the verts around, but the art director who designed the entire pipeline, and the tech artist who figured out best way to optimize it all, and so on. One thing has influence of many people on it so pretty hard to say exactly what its real cost is.
I remember we charged over 5k for a render of a beer glass to go on a billboard once and it was apparently a lot cheaper than taking a photo
cg isnt cheap, but it is cheaper
No, if they're buying assets from a library they're not gonna be willing to pay that much. But if they're making the asset in-house, so that they have full control of art direction, the total costs associated with it could easily hit four thousand. That's not just the artist making the initial assets, that's also stuff like the leads' time reviewing it and giving feedback, tech art making sure it works with the physics system, time spent on reworks due to changes in narrative and art direction, and so on.
we just had an asset that hit a anniversary just right before it was considered done
stuff like that can happen