Prop Workflow

cbates540

Hey all! So recently I've been doing lots of research on workflows because I've been struggling with my own and realized I didn't actually know what I should be doing, on some level. I understand that most organic/character workflows start with a high poly, retopo for low, and go from there. I'm having a hard time understanding the workflow for props though. It seems too slow to be the standard with props to have to retopo everything. 

With that in mind, is this what the typical workflow looks like for props/hard surface, or am I on the wrong track?
Create a base
High poly
Edit the base to closer match the high poly
UV
Make a cage from the low poly (if necessary)
Bake, etc.

Alex_J

First, I don't think hi-to-lo baking is "standard" for characters. It is what most tutorials show, but there is plenty of working artist here who don't use that as their go-to workflow.

To me, the main benefit of working hi-to-lo is for concepting. If you aren't crystal clear on the final design, it's good to just keep very fluid with dynamesh, figure it out, and then make your lo-poly model. One way or another you have to build that clean lo-poly mesh -- no escaping it -- but if you aren't sure exactly what geometry you'll need its better to save it until you do.

Same with props. If you got a clear concept and you are just making that exactly as you see, it's faster to build your lo-poly model, subdivide and sculpt details if necessary, bake, and then adjust that lo-poly further for game optimization if necessary.

I'll tell you the same thing I tell everybody -- don't sit and think about it too much. Just get started and go with a mindset of experimentation. Try both workflow styles. Once you know the tools thoroughly you'll have the freedom to change your workflow on the fly rather than being stuck in a linear, step by step process.

poopipe

Blockout 
High res
Low res

In that order. 
This way minimises time lost to large scale design changes.

Your large forms are defined by the Blockout - that goes in game and the team can work with that until the final art is done.

Smaller scale design decisions are defined by your highres model - you can resolve connection issues, frequency of detail etc etc. 

Low res comes last because it's makeup is determined by the highres - you don't know what geometry is required for the low res model until you see the highres.

It's a very rare occasion that concept art survives the build process 100% intact

cbates540

Thank you both for your insight!

So, as a follow up question, in a production environment, do you make the high poly, and then make as many of the changes as possible to it based on feedback, and THEN make the low poly? And hope that by then, any remaining changes will be minor, otherwise if it's something drastic, it's possible a new or very edited low poly will need to be made? Or is it some other way?

@BIGTIMEMASTER, I know I do have a tendency to overthink things, but I'm tired of feeling like everything I do has to be final because going back to change things ruins every step after, until it's just a huge mess. Changing things is always a pain to some degree but I know I'm doing something wrong when my workflow is this inefficient.

poopipe

yes ,  anything that minimises the chance of you having to do something twice is a benefit

musashidan

One of the most important aspects of dutifully working to a concept is a proper blockout. This will save a lot of time. It allows you to workout any inevitable '2d to 3d' issues that arise. It allows for rapid ideation, see what works, correct scale, animated components are functional, appraise the forms, and most importantly, broad changes are simple. 

It's enough so that you and others involved in the pipeline can voice opinions/critique, and make changes without hurting any feelings or wasting precious production time.