Independent game studio creating Frog Frag and Krivda. Small team, big ambitions.https://sisyphus-gamedev-studio.github.io/en-ushttps://sisyphus-gamedev-studio.github.io/en/#news-item-0/https://sisyphus-gamedev-studio.github.io/en/#news-item-0/Sisyphus Studio opens its doors. We are a small team ready to take on complex challenges. Our goal is to create deep, well‑crafted games that respect the player’s intelligence and leave a lasting impression.Tue, 20 Jan 2026 00:00:00 GMTSisyphus Studio opens its doors. We are a small team ready to take on complex challenges. Our goal is to create deep, well‑crafted games that respect the player’s intelligence and leave a lasting impression. Our story began in August 2025 with a team of three people and an ambitious goal — to create a large‑scale modification for a popular game. But we quickly ran into a fundamental problem: trying to implement our creative and technical ideas on top of someone else’s product was a path of compromises and “duct‑tape” solutions. Seeing that even the best mods on the market suffered from optimization issues, we made a key decision in November of that year: instead of patching up someone else’s foundation, we would build our own. We realized that creating games from scratch is an order of magnitude more difficult, but we were inspired by complete creative freedom, the ability to personally answer for the quality of the product, and boundless opportunities for professional growth. That choice became the philosophy of Sisyphus Studio: we create an environment that fuels ambition at every level. Experienced professionals get not just tasks, but creative freedom and the chance to add a landmark commercial project to their portfolio, expanding their expertise at the intersection of different disciplines. At the same time, ambitious beginners find fertile ground for growth here, learning from masters of the craft while working on real‑world tasks. And so our story began. The story of a team that consciously chose the difficult path for the sake of high ideals!https://sisyphus-gamedev-studio.github.io/en/#news-item-1/https://sisyphus-gamedev-studio.github.io/en/#news-item-1/Our first project is officially announced. Frog Frag is a chaotic deathmatch shooter with procedurally generated maps and weapon crossbreeding.Sun, 25 Jan 2026 00:00:00 GMTOur first project is officially announced. Frog Frag is a chaotic deathmatch shooter with procedurally generated maps and weapon crossbreeding. Frog Frag is a fast‑paced deathmatch shooter where every round is unique. Maps are procedurally generated, weapons can be crossbred, and the only rule is to be the last one standing. Right now we are laying the foundation for our new shooter. We chose a first‑person perspective because we believe it’s the best way to ensure maximum dynamics and absolute fairness in combat. And to make every new run feel unique, we are adding roguelike elements — a system of passive buffs that will take replayability to the next level. Of course, this is just a small part of the mechanics that await you. All in good time! A small insider peek from the development kitchen: initially we planned to make the game on the Godot engine. But when Unreal Engine 5 received a major optimization update, we realized — that was exactly the power and flexibility our project needed. The switch was worth it! Subscribe to our development diary to be the first to know technical details and not miss upcoming playtest announcements.https://sisyphus-gamedev-studio.github.io/en/#news-item-2/https://sisyphus-gamedev-studio.github.io/en/#news-item-2/Frog Frag Dev Diary #1: Basic weapons and the crossbreeding mechanic.Sun, 15 Feb 2026 00:00:00 GMTIn the first stage, we decided to focus on creating custom mechanics and a basic arsenal. The logic of crossbreeding and choosing the ideal weapon “bases” is a serious challenge. Here’s how we approached it! The weapon crossbreeding mechanic is the heart of our game. How well it’s implemented will directly determine how fun the game is to play. To simplify the 3D design pipeline and avoid spreading ourselves too thin, we decided to start small: we approved 3 basic weapon types. Work is now in full swing around them — we are doing both theoretical calculations and practical tests in the engine. Our goal is to make the basic guns radically different in terms of characteristics and functionality. We want their hybrids to feel unique, behave unpredictably, and even be absurd at times. The work is progressing in parallel: while the creative director and the code team are assembling the mechanical innards of this trio, our 3D designer is preparing blockouts of the “pure” versions and visualizing the results of their crossbreeding. As soon as the concepts get the green light, the pipeline shifts: the 3D department moves on to texturing and detailing the models, while programmers and game designers immediately start prototyping the next weapon and calculating its combinations with the already finished arsenal. Stay tuned for development news — more to come!https://sisyphus-gamedev-studio.github.io/en/#news-item-3/https://sisyphus-gamedev-studio.github.io/en/#news-item-3/Development Diary #2: Research on CAD-to-Blender pipelines and polygon mesh optimization.Thu, 16 Apr 2026 00:00:00 GMTIn indie development, optimization is not just a final polish before release — it is the foundation on which the entire content creation process is built. Recently, our 3D designer conducted a major study focused on a problem many studios face: the loss of mesh quality when transferring CAD models into Blender. With standard export (for example, to OBJ format), mathematically perfect CAD geometry often turns into chaos: extra triangles appear, edge density becomes uneven, and artifacts show up on fillets. That complicates further work with shading and UVs and hurts in-game performance. We decided not to accept this and tested several pipelines using a complex model with many chamfers — a vintage Coca-Cola Vendo refrigerator. Key findings from our research: Basic export (OBJ): The fastest but lowest-quality approach. The mesh ends up uneven; corners and chamfers get covered in excess polygons, requiring long manual cleanup. Plasticity Bridge: A great tool for fast iterations and real-time geometry checks. With tuned parameters, triangle count drops significantly (in our test — from 62K to 16K), but complex areas still retain unnecessary cuts. Processing through PIXYZ: The clear winner of our testing. The software preserves edge-flow logic and drastically reduces polygon count (down to 6.8K triangles in our case), producing a clean mesh that is almost ready for a game engine. The secret ingredient: We found that tessellation quality depends not only on software. Adding extra guide sections at the CAD modeling stage "hints" to the program how to interpret complex shapes, minimizing artifacts. What does this mean for our project? Integrating PIXYZ together with preparatory guide edges in CAD completely changes our hard-surface asset pipeline. First, we save hours of routine work. Modelers no longer need to manually clean generated meshes (retopology), freeing time for new mechanics and detail work. Second, we get highly optimized models with clean shading. Third, lower polygon counts directly improve performance — our game will run smoother even on weaker systems while keeping high environmental detail. We keep pushing our boulder toward the summit — and now we do it with a bit more tech under the hood.