yet another retopo or not thread

dzibarik

So I have heard numerous times here that hard-surface retopo is a big no-no in production. Why? Because if people want to change something afterwards it's hard to change high poly in any sculpting app for example. But then I see how Redstorm[ame="http://www.youtube.com/watch?v=gd3-6t59uU0&index=40&list=WL"] uses this workflow for Division [/ame] and Max Aristov (the senior artist on DOTA2) use Zbrush for modelling hard-surface sci-fi guns.

Digital Tutors also posted a training course where they retopo a high-poly revolver instead of removing control loops and adjusting geo to match high poly better.

I also had bad experience with tranforming a high poly model into a low poly mesh by removing loops and adjusting geo because I tried to use as much less geo as possible when I modelled high poly and when I started making low poly from that I actually had to add more geo to maintain smooth shapes of the car I modelled and remove control loops at the same time. It took me pretty long time and I think I'd have retopoed it in the same timeframe, if not faster.

And while people say it's easier to adjust a traditional model my experience says otherwise. After you added control loops it's not that easy to change a shape radically, you have to remove control loops, change the geo, then add them again. Sometimes it can be not that fast.

Soooo.... if I retopo my mesh afterwards anyway why should I stick to traditional box modelling for all hard surface models? I mean why shouldn't I use Meshfusion/sculting app/CAD app/MOI/whatever and then export high poly for retopo?

some MOI examples:

oglu

is that MoI example in the last image really a retopo or a boolean..?

dzibarik

oglu wrote: »

is that MoI example in the last image really a retopo or a boolean..?

sorry, I posted it accidentally. It's just boolean.

FelixL

I usually take the highpoly and remove loops. You're right about having to add loops for round stuff, but I don't see it as that big of a deal. The pics you posted seem too dense to me, unless this was somehow a huge object.

dzibarik

FelixL wrote: »

I usually take the highpoly and remove loops. You're right about having to add loops for round stuff, but I don't see it as that big of a deal. The pics you posted seem too dense to me, unless this was somehow a huge object.

they are dense but the point was that it's easy to make a high poly where you want and then retopo it.

I'm just confused that there is no "traditional" way on which everybody agree. You hear folks claiming that sculpting hard-surface is not suitable for production and then you see other industry folks doing this. It's important to me because I have to know which skill I should focus on.

I think I'll stick to traditional subd for the time being. I'm just interested in different perspective on the subject.

passerby

you will quickly find there is no one "Right Way" it really comes down to a per artist thing, yes there is some truth in what artists are saying when taking the side of subd or sculpting, but neither system are perfect for all uses.

In my personal opinion which should be taken with a grain of salt you should learn subd first so you have a good grounding of what makes good topology, since even when sculpting you sometimes have to deal with bad shading or pinching due to the topology not being good or not being even enough.

FelixL

It's best to know traditional subD. Having to do even the most simplistic shapes in Zbrush and being forced to deal with the numerous downsides and heavy meshes isn't advisable.

Most people agree that Zbrush is great for doing hardsurface if you haven't nailed down the design yet. If you have a detailed concept, and the model is big (car or tank, etc) it's better to use traditional subD.

Of course, that could change with the next zbrush update that has some traditional poly modeling tools built into it. You might not have to dynamesh everything and the workflow might not be as destructive as it is now.

dzibarik

I'm actually alright with subd. There is a LOT of room for improvement and I'm still not as fast as I'd like but since the industry is rapidly changing I may find skills that I work so hard to improve to be useless in a couple years.

But that's the way it is, I think, and I should deal with that.