Behind the Rigging & Animation of Fantasy Action RPG Characters Like Orcs and Dragons
Yeyo Studio breaks down its approach to rigging, skinning, locomotion, combat animation, creature movement, engine testing, and building reusable animation sets across characters with radically different anatomy and equipment.
Creating a compelling character animation is one challenge. Building dozens of animations that must work together as a cohesive gameplay system, across different weapons, armor configurations, character proportions, and engine conditions, is another entirely. That's the kind of production challenge tackled by Yeyo Studio, a 3D animation and art outsourcing team specializing in character and creature animation, rigging, skinning, and game-ready 3D assets.
One of the studio's recent collaborations with Blink Studios demonstrates just how quickly that complexity can scale. Yeyo worked across male and female human characters, three distinct orc archetypes, an ogre, skeletons, a necromancer, and 15 creatures, with roughly half of the creatures designed as larger boss-tier variants. Most characters required around 42 animations, spanning locomotion, combat, attacks, reactions, jumps, rolls, stuns, deaths, and other gameplay states. Another production batch involved skinning and testing around 100 armor sets against the existing animation library.
We spoke with Yeyo Studio about its production pipeline, building flexible rigs, animating humanoids and creatures, testing everything in-engine, and what developers should prepare before handing a character over to an external animation team.
For readers discovering this project and your studio for the first time, can you explain the main scope of Yeyo Studio’s typical work, and how you collaborated on Blink Studios’ Humans, Orcs and Creatures animation sets?
Oskars Dzenis, Founder and CEO at Yeyo Studio: Yeyo Studio is a 3D animation and art outsourcing studio focused primarily on game production. Our typical work includes character and creature animation, rigging and skinning, and 3D character, prop, and environment art.
We usually work as an external production partner, supporting development teams with specific parts of their pipeline—from preparing game-ready rigs and animation sets to integrating and testing assets in Unity or Unreal Engine. Depending on the project, we may handle a single production task or support a client continuously across multiple characters, assets, or development milestones.
Our collaboration with Blink Studios was divided into several batches, which we completed one by one. The project included human male and female characters with multiple variations in armor, weapons, and textures, as well as three types of orcs—the Warrior, Warlock, and Hunter. We also worked on an ogre, skeletons, a necromancer, and 15 creatures. Approximately half of the creatures were larger, boss-tier versions with additional details, such as a heavily scarred bear with arrows sticking out of its body.
Most characters required around 42 animations. Some motions could be shared between character variations, while others had to be tailored to their weapons, proportions, and combat styles.
Another batch included around 100 armor sets for the human male and female characters. Each set had to be skinned and tested with the existing animation library to ensure correct deformation across the full range of movement.
Your studio's expertise includes a variety of locomotion and character animation work. What kinds of animations did the team create for the humans, orcs, and creatures here, and how did you decide which ones best represented the project?
Oskars Dzenis: For humans, orcs, and creatures, the animation lists cover complete RPG animation sets that include: idles, combat idles, directional walks and runs, strafing, sprinting, rolling, jumping, falling, landing, buffs, hit reactions, stuns, deaths, and multiple attacks for different weapons. The Warrior, Warlock, and Hunter also needed actions that reflected their individual roles and fighting styles.
The creature animations followed similar gameplay requirements but were adapted to each animal’s anatomy, size, and behavior. These included locomotion, idles, attacks, reactions, special or boss actions, and deaths.
When putting the reel together, we did not want it to feel like a simple checklist of animations. Instead, we picked the most interesting and distinctive moments for each character and built small sequences around them. This helped us avoid repetition while giving the reel a stronger pace and a gradual build-up.
One of the interesting technical points is the orc rig, especially the ability to support different armor pieces on the same body. Can you walk us through how that rig was built and what challenges came with making it flexible enough for multiple armor variations?
Oskars Dzenis: The main challenge was finding joint positions and orientations that worked well across all armor variations. There was no perfect setup straight away—it was a process of testing different armor pieces across the animation set, making adjustments, and testing again. The goal was to make everything work naturally on the same body and rig while keeping intersections to a minimum.
What were the biggest skinning challenges on this project, especially for characters that needed to support armor, physical movement, combat motion, and repeated gameplay-style animations?
Oskars Dzenis: The trickiest areas were usually the shoulders and flexible parts like belts, clothing, and layered armor. They might work perfectly in normal poses, but once the character moves into a more extreme attack pose, parts can start intersecting. A lot of the work was about finding that sweet spot in the skinning weights - keeping enough freedom for the movement while avoiding visible clipping.
How did the team approach animation for both humanoid and creature characters? What differences did you have to account for when working on humanoids versus creatures like horses and dragons?
Oskars Dzenis: We kept the animation process collaborative. On smaller projects, one animator usually handles production while another team member reviews the work at regular stages. On larger projects, two or more animators divide the tasks based on their strengths and review each other’s work. This gives every animation a fresh pair of eyes and helps catch issues that the original animator may stop noticing after working on it for some time.
For humanoid characters such as orcs, we had to account for their body proportions, armor, and various weapons while ensuring movement remained readable and equipment did not restrict poses or cause excessive clipping.
For creatures such as horses and dragons, the main challenge was coordinating all four limbs with the spine, head, balance, and weight distribution. Their movement also had to reflect the animal’s size and anatomy. In many cases, quadrupeds can be more challenging because even a small timing problem in one part of the body can affect the entire motion.
Can you walk us through the animation production pipeline, from rig preparation and blocking to polish, testing, and final delivery?
Oskars Dzenis: We usually start by receiving the character models from the client, together with any armor, weapons, or other related assets. From there, we build a custom rig for each character. To speed up this part of the process, we use several in-house automation scripts.
Skinning is done using a mix of native tools and third-party scripts. We also have our own library of test animations that move individual body parts in different directions. This helps us quickly check how the skinning behaves and catch deformation issues early.
Once the basic rig and skinning are ready, we create a rough blocking pass. We show this to the client early so we can agree on the overall feeling and style of the animation - whether it should be more realistic or more exaggerated and stylized. This usually involves some back-and-forth until we find the right direction.
At this point, we can already bring the character into the game engine and begin testing. We normally start with the idle, walk, and run animations because they provide a solid foundation for the rest of the set. We check whether the animations and skinning behave the same way as they did in the original scene and look for mesh intersections or import-related issues.
Foot sliding is one of the most common problems. Sometimes we also see small, strange jerks at the beginning or end of an animation. These can be caused by animation-baking problems or engine import settings that need adjustment.
Once the idle and locomotion animations are working properly, we move on to the rest of the animation list. We keep the finished animations in the same working scene as references, which helps us maintain consistent timing, movement, and style across the whole set.
Each animation is then exported, imported into the engine, and tested separately. When needed, we also connect the animations inside the locomotion or animation system to see how they transition and blend together.
For animations with more extreme poses, such as attacks, we sometimes need to go back and make additional rigging or skinning adjustments. This may include changing joint positions or orientations and polishing the skinning so the character deforms correctly. After the final engine tests and any last adjustments, we export and deliver the complete animation set in the required format.
Several moments seem designed to suggest actual gameplay contexts, such as the horse-and-rider ground-smash sequence and the bear attack from a first-person viewpoint. How did references from RPGs, FPS games, and action games influence the way you staged those animations?
Oskars Dzenis: Since the client’s original vision for the characters was RPG-focused, we used familiar gameplay situations as inspiration and presented the action from multiple viewpoints.
The horse-and-rider ground smash was inspired by the dismount attacks seen in Diablo IV and similar action RPGs. For the bear attack, we switched to a first-person view inspired by FPS games and first-person RPGs. Placing the viewer directly in front of the bear made the attack feel more immediate and showed how the animation could work in an actual gameplay situation.
For the reel presentation, you also added camera transitions, camera shake, selected viewpoints, music, sound effects, and some VFX polish. How did you decide how much presentation work to add while keeping the focus on Yeyo Studio’s core rigging, skinning, and animation services?
Oskars Dzenis: At first, we made a fairly simple version of the reel, with similar camera angles and each character showing its animations. After reviewing it, we realized that we already had this kind of presentation in our portfolio, so we decided to take the characters on their own little journey and give the reel more personality.
Since we have some experience with game-trailer previsualization and cinematic 3D projects, we went for a more dramatic approach. We chose a strong music track and kept rearranging the shots and animations until everything clicked together.
Because many of the animations were loops, we used camera movement, different viewpoints, transitions, sound, and some VFX to make the shots feel less static. However, the characters always remained the main focus.
We usually work on this kind of presentation during quieter periods, often after finishing a client project and when we need something new for marketing. Client work always comes first.
Within a project, we also try to give team members some freedom to switch between tasks. After working on one animation for some time, for example, someone might move to rigging or skinning for a while. It helps us keep a steady pace, avoid blockers, and make sure we deliver everything on time.
Finally, what advice do you have for game developers or asset creators in terms of what they should prepare when approaching an external animation team for game-ready character animation sets? Are there specific references, skeleton requirements, engine targets, or implementation details that make the process smoother?
Oskars Dzenis: The most important things are open communication and a fast feedback loop. Ideally, the client should provide the latest approved character models, a preliminary animation list, visual references, information about the target engine, and a short explanation of how the character and animations will be used in the game.
On the technical side, it helps to define whether the project uses a custom or shared skeleton, its hierarchy and naming conventions, and whether the animations should use root motion or remain in place. The required frame rate, scale, export format, weapon attachment points, and any IK requirements should also be clarified early.
If an animation controller, locomotion system, blend tree, or existing engine skeleton is already in use, we prefer to receive it as early as possible. This allows us to test the animations in the actual gameplay setup instead of relying only on the original animation scene.
Not every client comes to us with a complete production plan, and that is completely fine. We can help organize the animation list, define the technical approach, and prepare an estimate, which we usually do for free. For larger projects, we prefer to begin with a small test batch—usually one character with two or three animations. This allows both sides to confirm the visual style, technical setup, feedback process, and production pace before moving on to dozens of characters or hundreds of animations.
Oskars Dzenis, Founder and CEO at Yeyo Studio
For more details on Yeyo Studio, check out their website.