Hey there Polycount,
I was wondering what your workflow was when it comes to creating models. Mine has always been to start with a low poly base mesh unwrap it and build up from there. But, I've heard from different artists it's better to start from a high poly sculpt, and then work your way down.
I was wondering about your experiences with this, and if you feel one work flow is better than the other. Or if there are pros and cons to both. If you guys could shed some light on the subject it would be appreciated! Thanks in advance! Looking forward to hearing from all the different artists on here!
Replies
Regardless - while jumping straight to the sculpt can be tempting, I personally find it to be a waste of time and energy. My personal take for a standard, current-gen character would be as follows :
- Blockout model to get the proportions just right. This can involve regular modeling as well as rough sculpting. I personally consider this to be the most important step.
- High res modeling, involving sculpting as well as subdivision modeling.
- And from there, whatever is needed to get to the game model
I think this kind of discussion would be at its most productive if we talked about specific examples ! Is there anything in particular you have in mind ?
this.
I almost always start with a sculpt now days.
So I start with a rough sculpt to set the proportions. Then make the highpoly sculpt with all the details and finally I build the lowpoly on top of the decimated sculpt in 3dsmax or 3D-Coat.
However, when I am asked to add some small props to a character at the last minute I usually don't bother to create a highpoly. I just create directly the lowpoly in 3dsmax, unwrap it, texture it in Photoshop and create the normal with a handmade heightmap. That's not super clean but super fast, hehe.
So in conclusion, it really depends on the situation.
So it seems like for most cases if someone is starting a character they’ll build the model in a highpoly sculpt and subtract down. What would you say about current-gen props and environment set pieces? Would you say that they would get the same treatment?
Then I model the high poly.
For the low poly, I start by copying and pasting the high poly, removing the turbosmooth modifier, and applying a push modifier to shrink it back down to size of the smoothed high poly. I remove all of the control loops using the ctrl+backspace shortcut and tweak the mesh to get it as close to the shape of the high poly as I can within budget.
Finally, UV and bake.
With GoZ it's pretty easy to jump back and fourth between sculpting and poly modelling.
Is it a bad workflow to start in 3ds max and then import into zbrush to add on details and export out maps?
for cars I do low, then subD to high then back to low so its nice and smooth
for characters (not my thing) some start with zspheres in zbrush and some use a base from maya/max and bring it into zbrush
try different things and go with what you like best
you can see the whole modeling workflow from start to finish here :
[ame]https://www.youtube.com/watch?v=nLxBoZWDwrw[/ame]
I know a lot of college is what you make of it, but the longer I start to really dig hard into game art I realize there's a lot I missed out on.
Good ole' for profit schools! But, thanks again everyone!
Pretty pumped to check this out! Thanks for the share!
This is the best approach. If the client or your AD knows exactly what they want and gives you a fully rendered concept drawing I often skip the blockout and just start in on the highpoly working from large forms down to small forms.