Dynamesh: The Best Tool for Hard Surface Design

Cane has shared some of his thoughts on ZBrush for Hard Surface Design and is giving a 25% discount for his latest tutorial to the first 50 people.

My name is Cane, and I'm a 21-year-old 3D Concept Artist. I make weapons and other hard surface props for games and entertainment.

All too often I see 3D concept designers using tools that greatly hinder their ability to design in 3D. Whether it's dealing with annoying topology or the rigid constraints of CAD modeling, It's far less fluid when compared to the traditional methods like drawing or painting.

Over the course of my time as a 3D designer, I have been trying to find a software that can bridge the gap between the rigidity of 3D and the fluidity of 2D. 

“Sculpting software enters the chat”. 

It may seem counterproductive to assume that a software, used to make organic forms, could be so great for hard surface design, but it is. And in this article, I intend to show you why.

I designed this radio inside ZBrush using the DynaMesh feature. As you can see, it's totally possible to make complex hard surface models in DynaMesh. Now the question is why you would want to do so over the alternatives. 

Previously, I had been using polygon modeling to design in 3D, and it was great for modeling what already existed, but when it came to designing in 3D, it was extremely rigid. No matter how proficient you may be in your favorite modeling package, it will inevitably dictate the decisions you make. 

For example, in CAD software, it's easier to do booleans, but harder to manage organics. The results you see from CAD designers follow these constraints. often making models that appear stiff and lacking nice forms or sweeping bevels. In poly, it is identically inverted. Booleans are hard, but, organics are easier to manage, the result is fewer compound forms and a design that is generally more simple. With the methods I use inside ZBrush, I can marry the best of both worlds, getting amazing control of both hard surface booleans and the organic control ZBrush is known for.

But it actually goes much further than just “good booleans” or “fluid organics”, these are just the operations everyone can immediately relate to from each side of the aisle. But there's a whole new world of creativity that could never be implemented into poly or CAD softwares through improvements or updates. And that's the freedom of sculpting, I know, it seems like sculpting is the opposite of hard surface to many people, but it's actually the perfect tool. 

In any 3D software you use, there are constraints for everything…for CAD it might be – not being able to get the fillet you want because the software gives you an error, or trying to slightly adjust a form only to realize it’s not possible. For poly, it could be topology or booleans. Not having constraints is only something 2D artists have really been able to experience. But with ZBrush, it’s the closest I've gotten to that freedom in 3D, and it's amazing. 

To showcase the power of sculpting, I'd like to talk about a stage every 3D artist knows well, the blockout. In this stage, we slowly build up the primary forms of the model with low poly geometry. This is so that our proportions are established. In this stage, we don’t add details or bevels to anything because it hinders the editability of the model later down the line. For designers, these constraints are annoying. And in ZBrush, the blockout stage is the complete opposite of restrictive, it's the stage where you can be the most creative and free with your designs.

On the left side of the image above, you can see that the modeling is loose, but the design is defined. It took me less than 10 minutes to sculpt the blockout and you can see that the final design is extremely similar. That level of speed and freedom is just not possible in poly or CAD. As a 3D Designer, this is all the information I need to take the model to the next level and polish it to completion. 

In total, it took me 12 hours to finish this radio. 

If you are a 2D artist, you could make a loose base mesh like this, then draw over it in Photoshop for a more accurate final design. The freedom in this stage alone makes it worth scrapping the traditional way of doing block outs. Even if you decide to finish it in another program, your designs will greatly improve, due to the initial freedom.

As you can see in the image above, I’ve made 5 iterations from the base nozzle design to the top left. I spent less than 3 minutes on each of the iterations. Again, these are meant to be rough, we are finding the shape, flow, and feel of the nozzle. The number of creative solutions you can come up with when you’re not spending all that time detached from the design is mind-blowing when you first start to design in ZBrush. Imagine how long this would take in poly or CAD. My brain hurts just thinking about moving vertices or splines around. As my friend Himesh Anand says – “is like juggling 5 balls and drinking water at the same time”. A sentiment I think many using these programs feel... Especially on geometry that is semi-organic, there's really no way to design on surfaces like this without severe constraints. In poly, its constant reconstruction, moving topology around and suffering. In Fusion, it's so constrained I'm not even sure if you’re able to iterate without having a timeline back to the start of the mesh to rearrange the base splines.

I asked my friend Greg, a 3D Concept Artist using Fusion 360, how he would go about iterating like this in CAD – this is what he had to say ...

Clearly very excited to get started on the nozzle

OK, OK, that was a low blow, of course, ZBrush will be easier for semi-organic shapes. So, let's say we decided to make something traditionally regarded as a ‘hard surface prop’. Like a weapon sight.

As you can see from this scope design, it's capable of producing models that look like they were made in Fusion. The quality of the final result is up to how much time you spend fixing the imperfections. For this scope, I didn't spend a lot of time refining it. I made it to be rebuilt as a game asset. Even with minimal refinement, it’s quite clean. Although I have not yet attempted to bake down one of my concepts, I believe that you could potentially refine your mesh to the point where this could be possible.

So why not use Fusion to make something like this? Again, it comes down to the methods of construction. If we were just recreating this sight, I'd use Fusion 100% of the time. But we are trying to make design decisions. We don't know if our choices will end up working or not, we might spend a bunch of time trying to implement something that in the end, does not work. We need tools and methods that are fast, intuitive, and enable exploration. 

I don't want to waste time constructing a spline from a 2D plane for cutlines, booleans or slicing my mesh. I want to be able to draw it onto my design, like a 2D artist would, without constantly taking myself out of the process. 

Imagine if 2D artists had to constantly switch into different menus every time they made a brushstroke. Or if they made their brush too big, they got an error telling them they "can't do that" for some unknown reason. It would drive them mad.

The time it would take me to do all of these operations in Fusion 360, compared to ZBrush, is unbearable when you start getting the hang of it. 

Let's start with this clip curve operation. In ZBrush, I hold down CTRL + SHIFT and draw out the curve. It automatically projects from my camera orientation and allows me to quickly make cuts into my mesh.

In Fusion, if you want to do that same operation, you must first create a work plane, then orient it to the angle you would like to cut, click on the plane to make a sketch, select the spline tool and draw out the cut. After that, you must go out of the sketch, select the spline you’ve made, then choose the slice operation and select the object you’d like to slice, then, delete the unwanted part you just sliced.

Yea. Now imagine doing this every time you want to make a cut that isn't perfectly aligned to a surface… the result is that you just won't do them as often, and your designs will suffer because of it.

What about cut lines? In ZBrush, I can simply draw them directly onto the mesh using the DamStandard brush. If I want to make it go from one axis to another, I just draw it from one to the other. Quick and easy.

In Fusion, if I want to make cutlines that travel across differing axes, I must construct multiple different sketches from each of the different axes, extrude them all into one big boolean, split body, then select the edges of the split bodies and give them a chamfer.

The slowness for this kind of operation is borderline unusable for concepting when it gets complex enough. What if I want to change the line I've made slightly, or do multiple iterations to see what works best? Am I just supposed to take 5 minutes every time I want to see how a cutline would look? I'd rather just do a line over at that point.

The last example I have is meant to illustrate how the“sketch” feature from Fusion would look in ZBrush. While it's called a sketch in Fusion, it's more like a surgical procedure. A sketch in ZBrush is done with the mask tools. it's loose, and can be refined. It's not limited to just one surface at a time and can be used to get the same boolean operations Fusion sketches are known for. It also lets me do Draw Overs directly on my mesh, allowing me to simulate design decisions I'm thinking of making by seeing it done in a mask first.

Our tools should not be a nail in the coffin that stops us from exploring. While Fusion offers a lot of freedom compared to the traditional poly modeling packages, from the perspective of doing hard surface in ZBrush, it’s rigid and slow. Fusion is an engineering tool, meant for executing on pre-planned schematics. it's not designed to explore and iterate on aesthetic design decisions.

To show you the process I use to create Hard Surface props in DynaMesh, I've released a piece of content from my tutorial for free. This is a time-lapse of me making the stream light TLR-6.

As you can see, the methods I'm using are quite unique when compared to the usual methods used to create props like this. 

I don't recommend using ZBrush for pre-existing props, as the normal methods in CAD would be faster for recreation. ZBrush really shines when you start exploring a design.

If you’d like to learn the methods I use in ZBrush to design props like the radio above, then check out my tutorial – Hard Surface In ZBrush

USE CODE [DISCOUNT80LVL] to get a 25% discount. 25 copies will be on ArtStation and 25 on Gumroad. THAT'S 50 COPIES. HURRY!

In this 3 part tutorial, I will give you a complete walk-through of all the tools I use, the logic behind them, as well as show you examples of how I use them in my process.

In part 2, I will break down my previous BB-L Radio design piece-by-piece, and show you in real-time exactly how I went about constructing the model in DynaMesh.

And in part 3, you can watch it all come together in a 3-hour time-lapse where I make the BB-S Radio from start to finish. interspersed with talks about the software, my design process, and keeping your mesh clean for renders.

You’ll only need a very basic understanding of ZBrush, as the methods I use are not technical and can be understood quite easily.

If you’d like to follow Cane on ArtStation, you can do that here. Also, don't forget to join our new Reddit pageour new Telegram channel, our Discord, follow us on Instagram and Twitter, where we are sharing breakdowns, the latest news, awesome artworks, and more.

Join discussion

Comments 0

    You might also like

    We need your consent

    We use cookies on this website to make your browsing experience better. By using the site you agree to our use of cookies.Learn more