A shocking revelation, but weapons are important in a shooter. But how can a small studio like ours produce enough of them, and each of high quality?
Usually, the development cycle for a weapon goes like this:
A designer designs a weapon: its functionality and the general look and feel. In the real world, that step often happens later, when you already have a model. “I have a hand cannon here, how do I make it unique?”
A concept artist produces multiple sketches. The designer and the art lead consult on them, there’s a few iterations. The concept artist produces the final art, showing the weapon from multiple angles.
A 3D artist takes over, models and textures the weapon.
Then you have multiple other steps, like the animator inevitably going back to the 3D artist, complaining that a certain part needs to be a separate model so they can animate it, but as for the core basic model, that’s it.
As you can see, many people are involved in something as simple as a gun. Designers, concept artists, 3D artists, animators, vfx artists, sound engineers – not to mention project managers and others like leads.
In our case, the process is a bit different. Thanks to the talent of Marek Kaplita, the 3D artist behind most Witchfire weapons, we’re able to skip step 2. He’s basically a concept artist and a 3D modeller in one.
More often than not, it’s literally two e-mails and we get a weapon we love. But for the purpose of this post – to show the general design thought behind Witchfire weapons, and how the game development really works – we have chosen a weapon that took us the longest to get it right. So here’s how it all went down with the Crossbow.
It started with a pitch.
“We could use a magazine-fed repeating crossbow, something retro like they did in Van Helsing.”, we wrote to Marek and added a few reference images from Google and Pinterest. Marek’s first 3D sketch:
The weapon had multiple issues, with these two being the most important ones:
Too small and thus suggesting weak damage output
Most of the top part was too openwork and way too modern
The next version looked like this:
Here’s a direct translation of our feedback:
Some good elements, the ropes are an interesting idea. Nice mag, nice screws. But it’s still too small, it’s like it fires matchsticks. We need a powerful crossbow that fires solid thick bolts. It’s not a regular weapon but a reward so it needs to feel like one.
The shape is still too modern and high-tech, you can see that in FPP especially. Nothing interesting in the limb, it’s just a flat metal. We’re not sure how you actually fire it or hold it, or maybe we’re not seeing something.
Basically, looks like you wanted to make a simple small crossbow, and from than angle it works, but we need a powerful bad ass reward weapon.
Marek asked if this was a better direction:
Better but still not there, plus the visibility of the enemy targets would suffer. You know what, fuck it, let’s make a single shot crossbow. It’s going to be easier to make the bolt feel powerful, almost like from a stakegun.
We have attached an example of such crossbow found on Artstation, and Marek came back with:
Somewhat better, the front is more interesting now. The back is weird and wrong: with the weapon pressed against the shooter’s body, it would rip their skin off when firing. The metal strip is too small and too tight for the rope. And it’s still too small. Look at this highly professional sketch ;)
“Taller”, vertically stretched weapons look better. Longer back, and please make it as retro as the front plus some ornaments.
We got this the next day:
It just doesn’t work. Still too small and not retro enough. The back is banal. The wheel with the rope makes the weapon look more like a tool than an exciting gun.
It also does not make sense as a crossbow. The power with which the bolt is fired comes from the limb and not from the string. The string is used to stretch the limb, then you release the locking mechanism, the limb returns to the original position, so does the string, pushing the bolt forward. The gun is not a sling-shot, the string is not stretchable. But look at your crossbow, the limb is rigid, even fortified, and so after releasing the lock, the string would just loll, and the bolt wouldn’t even fall off the gun.
The weapon is still not interesting. You took our sketch literally, it was just a general shape direction.
But here’s where it does get interesting. In order to show Marek some interesting shapes and designs, we sent him these examples:
Amazing, right? The funny things is, these are Marek’s own designs, weapons he just made for his Artstation page.
“Yeah, agreed, I guess I got too attached to the old version”, replied Marek and after a short reset break he sent us this…
…and we were finally on the right track.
The first texture pass:
The feedback was simple. Good direction but it was too plain. For the weapon to feel special, we needed ornaments and extras.
Cool, but we needed some dirt so the gun does not look plastic:
Still not quite there yet. We sent Marek a couple of examples of real life dirty metal sculptures to show how the age affects them. And what we got back, was finally the weapon we all loved:
So there you go, this is what it takes. To be honest, usually the iteration loop is much shorter, as Marek often gives us magical “Yup, this is it!” designs with his first try, but we figured that showing a more difficult work would better explain the entire back and forth process.
On our side, we learned to better explain the ideas we have in our heads in order to avoid wasting Marek’s time. A certain lesson from the old times when some of us were in People Can Fly comes to mind:
Eric (studio head) during a meeting: “Everyone, imagine a shoe!”
Eric: “See? I ask for one thing, and everyone has a different idea of what it means. This is the core issue with communication.”
Which is also the main issue with outsourcing. We got lucky with Marek, not just because we can both design and execute but also because he can feel the style we’re after. However, we still needed to get better at describing the shoe we’re after. Fortunately for us, we did, which we guess is why we never had such a long back and forth ever again.
We hope this showed the amount of work needed to make roughly 10% of a gun. 10%, as there’s still animations, sounds, visual effect, not to mention the actual design of the weapon’s features and the implementation of these features, including enemy reactions (exploding from the inside is a reaction) and the weapon’s unique perks…
Yeah, there’s still a lot of work ahead of us.
Question of the Week
The off-center crosshair is an old idea, first introduced by Bungie in Halo 2. Then we’ve seen it in other games, including Destiny, Prey, and Far Cry 5.
The reason for it is that it helps to bias players towards looking up. Since the player unconsciously – there are many people who finished all Halos and never noticed the crosshair was lowered – assumes the crosshair is the center, and so they compensate by angling their view slightly higher.
But why do game developers want people to look up? For two reasons.
One, people tend to look at the ground. Watch a lot of top Destiny 2’s PVP players and even they – people who understand the idea of pre-aiming very well – often keep the crosshair low. Game designers want people to see the world, not the ground.
Two, and connected to the above, the off-center crosshair extends your peripheral vision. This is a fantastic GDC presentation explaining the idea.
Having said all that, in Witchfire you’ll be able to choose between the 100% centered crosshair and the lowered one. We’re gamers too, and we love to be able to customize the experience.
The reason why we had the off-center crosshair in the latest video is a simple story, really. Most of the studio plays the game on KBM, but we recorded the clips using the controller, as it offers a smoother camera movement than the twitchy mouse. The person recording had the lowered crosshair as his preferred setting. That’s it.