How I Turned VR Views Into a Real Community
A lot of VR developers today are getting views. TikTok clips go viral. Funny gameplay spreads. People watch. But then…Nothing happens. No players. No community. No retention. I’ve been there too. So instead of guessing, I started building a system inside my own VR projects to turn views into an actual community. This is what actually worked. I reached ~500 members in the first 3 months as a solo developer—and it’s still growing. And in this blog post, I will show you how! The Hard Truth Most VR games don’t fail because they are bad. They fail because they disappear after the first impression. You might even go viral and still end up with zero players. Because: Views don’t stay. Systems do. The Core Idea Views don’t convert by themselves. You need progression for humans, not just gameplay. Think of your community like a game: Players should join → progress → earn → belong Step 1 — The OG Hook (Early Identity) When I started: I posted short clips (simple, raw, not perfect) I pushed one message: “Join now and become an OG.” And early joiners got: OG Discord role Early identity Insider feeling 👉 Why this worked: People don’t join communities. They join identity. Step 2 — Moderator Grind Next, I introduced something very important: 👉 A Mod Role to grind for Players could earn it through activity Contribution = status Now, Discord wasn’t just a chat. It became a place where you can level up socially. Step 3 — Exploit Platform Behavior (Live Streams) This was one of the biggest breakthroughs. Instead of only posting videos, I created a constant presence. Long YouTube live streams Even automated replays Always something “happening” Why this works: Platforms reward presence, not perfection. 👉 YouTube pushes live content heavily 👉 VR gameplay is perfect for passive watching Your game becomes something people run into, not search for. That changes everything. Step 4 — Content Creator Program Then I built this system: Content Creator → Post a video Elite Creator → 1,000 views Gold Creator → 5,000 views Rewards: Roles Cosmetics Recognition Players stopped being just players. They became: Creators, marketers, and community leaders Step 5 — Live Events = Real Conversion Finally, I focused on something most devs ignore: 👉 Live moments I ran: Playtests DJ events Movie nights Discord hangouts At one point, I noticed something interesting: When nothing was happening, nobody joined. The moment something live started, people showed up instantly. Because VR is not content. VR is presence. People don’t join because your game looks cool. They join because: Something is happening right now, and they don’t want to miss it. The Real System (Simplified) What I ended up building was this: Content → Identity → Progression → Creation → Events → Community Each step feeds the next. The Mistake Most VR Developers Make Most developers do this: Post update Share feature Drop the Discord link But players don’t care. Because: There is no identity No progression No reason to stay What Actually Works If you want to turn views into a real VR community: Give early identity (OG roles) Add community progression (mod/status grind) Use platform advantages (live streams) Turn players into creators via Content Creation Programs Create live moments both in Discord & Game (events) Final Thought Your VR game is not just a game. It’s a social system. And if you design it right, you don’t need to chase players anymore. Your community brings them to you.📰What’s New in Meta XR SDK v85 | Start Mentor Workshop
Hosted by Quentin Valembois (Valem) • March 2026 • Meta Horizon Start Overview (what v85 focuses on) Meta Horizon OS v85 includes updates across: Building Blocks (including Multiplayer + AI Building Blocks) Mixed Reality (new Spatial Test Framework and Scene-less MR support) Locomotion updates (input mapping + revised control scheme) What’s coming next (FrameSync) Watch this part: 00:00 Building Blocks (Unity): faster setup, less boilerplate Building Blocks are modular, drag‑and‑drop capabilities for Unity projects that can automatically configure project settings and required components (e.g., passthrough, hand tracking, MR setup, etc.). The intent is to let you focus on the code that’s unique to your experience. What to take away: Building Blocks = the “fast path” for adding platform features correctly. They can handle annoying setup details (settings, manifests, component wiring). Watch this part: 01:07 Multiplayer Building Blocks: Photon Fusion 2.1 support What’s new in v85 Multiplayer Building Blocks now support Photon Fusion 2.1, unlocking newer Fusion features for Start devs who choose Fusion as their networking provider. Fusion 2.1 highlights: Forecast Physics Object Send Priority Large Data RPCs (Fusion used to have a small payload limit—large data RPC expands what you can send) Custom Tick Rates Faster Host / Master Client Switching Configurable AOI (Area of Interest) Player Unique ID When you drop a Multiplayer Building Block (e.g., auto matchmaking), you can choose provider: Unity Netcode for GameObjects, or Photon Fusion (now v2.1 supported) Also, voice chat support in the Building Blocks flow is tied to Fusion in the demonstrated setup. Watch this part: 02:19 AI Building Blocks updates: more accurate boxes + image segmentation What’s new in v85 AI Building Blocks improvements: More accurate 3D bounding boxes for object detection overlays 2D Image Segmentation option appearing in the tooling Watch this part: 06:45 “Scene-less Mixed Reality”: why it matters The problem with relying only on Scene Model/room setup Scene Model (room scan) is powerful, but can be: Overkill for simple MR interactions (e.g., “place one object on a surface”) Not runtime-updated (doesn’t naturally account for small moving objects in the moment) The v85 direction: more seamless MR workflows The workshop emphasizes using depth-powered environment raycasting to collide against real geometry without requiring a full scene model workflow. Key concept: Environment Raycast uses depth sensing to “hit test” the real world, enabling placement and interaction without scene model dependency. Watch this part: 08:27 What’s new for Scene-less MR in v85: Scene-less World Lock World Lock (and what changed) World Lock keeps virtual content stable in physical space even when the user recenters. Previously, this was tied to scene model workflows. What’s new in v85 Scene-less World Lock—world-lock behavior that works without requiring a scene model. Why it’s a big deal: It moves MR closer to “drop in/runtime MR” without asking users to preconfigure a full room scan for basic anchoring behavior. Watch this part: 12:11 Debugging MR got a major upgrade: Spatial Test Framework Before: Immersive Debugger The Immersive Debugger can help you inspect/tweak MR-related settings in a build and visualize MRUK data (mesh, collisions, navmesh overlays, etc.). New in v85: Spatial Test Framework (MRUK tests) The Spatial Test Framework brings automated testing principles (Unity Test Framework) into MR workflows by letting you run tests across multiple room prefabs/scene configurations. What it enables: Automatically validate your MR logic across many room layouts (bedroom, office, living room, etc.) Reduce manual “try it in 20 different spaces” testing Write your own tests by extending the MRUK test base class (as demonstrated conceptually) Watch this part: 18:50 Locomotion: input mapping + revised control scheme v85 includes an update to locomotion documentation and recommended patterns around locomotion input mapping, focused on making multiple locomotion systems work more cleanly together (especially avoiding input conflicts like “UI ray + teleport ray at the same time”). What’s new: Revised Control Scheme Watch this part: 26:24 More to come: FrameSync on Meta Horizon OS FrameSync is as an upcoming/important performance-related development, aiming at: More consistent smoothness Fewer still frames Lower motion-to-photon latency Watch this part: 29:24 Quick reference Photon Fusion 2.1 “What’s new” (from slide): https://doc.photonengine.com/fusion/current/getting-started/preview-2-1/whats-new-2-1 Locomotion input mapping doc (from slide): https://developers.meta.com/horizon/design/locomotion-input-maps/ FrameSync blog (from slide): https://developers.meta.com/horizon/blog/framesync-meta-horizon-os Environment Raycast example video (from slide): https://youtu.be/r9gedHRY0rc
What VR Developers Can Learn From Animal Company’s Viral Growth
In the past, growing a game meant running ads, contacting influencers, and pushing constant updates across social media. Today, the growth dynamic is shifting. Some of the fastest-growing games are not succeeding because they post more marketing content. They are succeeding because their gameplay naturally creates content that players want to share. One of the most interesting recent examples of this approach is Animal Company, a VR title that achieved remarkable organic growth in a short period of time. Within its first six months, the game reportedly generated: 1B+ TikTok views ~500,000 peak daily active users 9× growth in paying users Instead of relying heavily on traditional marketing channels, the game’s design itself became the marketing engine. For VR developers, this case offers several valuable lessons. I will share them in this blog post. 1. Design for Watchability, Not Just Playability Game developers traditionally optimize for playability: mechanics, progression, difficulty, and retention. But in today’s attention economy, there is another critical design factor: Watchability. Watchability means designing moments that are entertaining even to someone who is not playing the game. Animal Company’s gameplay consistently produces moments that work well on platforms like TikTok, Reels, and Shorts. These moments often involve: chaotic multiplayer interactions unexpected physics outcomes exaggerated emotional reactions social improvisation between players When these moments happen in VR, they are naturally clipped and shared by players. The result is organic discovery. Players become marketers simply by sharing their gameplay experiences. 2. Social Chaos Creates Shareable Moments Many successful VR experiences lean heavily into social unpredictability. Animal Company embraces this philosophy by building systems that encourage chaotic, funny, or surprising interactions between players. These include elements like: physics-driven movement expressive avatars playful or comedic animation systems environments that encourage experimentation The goal is not to script funny moments. The goal is to create conditions where funny moments are likely to happen. When these unpredictable situations occur, players instinctively record and share them. 3. Emotional Reactions Translate Well to Video VR has a powerful advantage compared to traditional games: embodied reactions. When players are immersed in VR, their responses are often more physical and emotional. Animal Company leverages this by using: spatial audio cues environmental tension sudden encounters or surprises These mechanics create intense player reactions — screams, laughter, panic, excitement. For social platforms, these reactions are extremely compelling to watch. A viewer might not fully understand the game mechanics, but they instantly understand someone screaming in VR. That emotional clarity makes clips far more shareable. 4. Turn Players Into Performers Another key idea behind the game’s success is treating environments as stages rather than just levels. Players are given tools and systems that allow them to perform: comedic skits stunts chaotic multiplayer challenges improvised social moments In other words, the game encourages players to create content inside the game world. This transforms the typical player into something more powerful: a content creator. And when hundreds or thousands of players begin creating content, the game’s reach expands far beyond the original player base. 5. Build a Creator Ecosystem Games that grow through social media often develop strong relationships with their most active creators. Animal Company reportedly implemented an invite-only creator space within its community — a Discord group where top creators can communicate directly with the development team. This type of ecosystem has multiple benefits: Creators receive recognition and support Developers gain direct feedback from influential players The community feels more connected to the game’s evolution Over time, this creates a feedback loop where creators help drive discovery, while the developers support the creators’ ability to produce content. 6. Community Feedback as a Development Engine Another notable aspect of the game’s development approach is its responsiveness to the community. Frequent updates and a close relationship with the player base allow developers to quickly react to emerging ideas, memes, or trends within the community. When players feel that their ideas can influence the game, they become more engaged — and more invested in the ecosystem surrounding the game. In many cases, community culture becomes just as important as the gameplay itself. 7. Monetization After the Community Forms One of the most interesting strategic choices was delaying monetization early in the game’s lifecycle. Rather than introducing monetization immediately, the focus was first placed on strengthening the social gameplay loop and building a loyal community. Once players were emotionally invested in the game and its culture, customization options and cosmetic purchases became more attractive. This approach often leads to healthier long-term monetization because spending is driven by expression and identity, not pressure. The Bigger Lesson Animal Company highlights an important shift in how games grow today. Successful games are no longer only designed to be fun to play. They are designed to be fun to watch. When gameplay naturally produces moments that players want to share, marketing becomes embedded directly into the experience itself. Instead of competing purely through paid acquisition, games can grow through the creativity of their own communities. For developers, this raises an important design question: If a player records 30 seconds of your game, would someone else want to watch it? If the answer is yes, you may already have the foundation for organic growth. If you're interested in learning more about designing games that generate shareable content, I’ll be discussing practical strategies and examples in my upcoming session: Craft Social Content Players Want to Watch | Growth Series, Part 2 📅 March 13, 2026 🕚 11:00 AM PDT Looking forward to seeing other developers there and continuing the conversation around how games and creators can grow together.I Built My Social Media Wrong — 10 Mistakes XR Developers Should Avoid
When developers ask how to grow their social media, they usually expect tips about algorithms, marketing tricks, or viral content. But after working on several VR projects and building communities around them, I realized something much simpler: Most growth problems don’t come from the lack of marketing knowledge. They come from basic mistakes developers make early. I made many of these mistakes myself. So instead of sharing “best practices,” I want to share the mistakes I see developers make most often when trying to build a social media presence around their game. If you avoid these, your growth will already be much easier. Mistake #1 — Starting Social Media Too Late A very common pattern: Developers spend 1–2 years building a game in silence, then suddenly create social media accounts when the game is almost finished. At that point, they realize something uncomfortable: No one knows they exist. Building an audience takes time. Sometimes, a lot of time. Social media works much better when the audience grows with the project, not after it. Sharing development early allows people to feel like they are part of the journey, not just customers at the end. Mistake #2 — Trying to Be Everywhere Many developers try to be active on: Twitter / X YouTube TikTok Reddit LinkedIn Discord Very quickly, this becomes overwhelming. The result is usually: inconsistent posting burnout abandoned accounts It’s much better to focus on one or two platforms first and build consistency there. Growth usually comes from depth, not from being everywhere. In my experience, initially, TikTok and YouTube are enough to get started. Mistake #3 — Only Posting Big Announcements Another common mistake is treating social media like a press release channel. Posts look like this: “We are excited to announce our new update!” Then nothing happens for weeks or months. Social media platforms reward consistency, not occasional announcements. Small updates often perform better than big ones: a new mechanic a funny bug a quick gameplay clip a design question These small posts help build ongoing momentum. Mistake #4 — Posting Like a Company Instead of a Human Developers sometimes try to sound “professional.” Posts become very formal and corporate: “We are pleased to introduce our latest feature update.” But social media works differently. People follow people, not companies. A much more engaging approach is simply sharing the real development experience: things that worked things that failed experiments funny bugs Authenticity is far more powerful than polish. Mistake #5 — Waiting Until Things Are Perfect Many developers hesitate to post because something feels incomplete. The UI isn’t final. The animation is temporary. The mechanic still needs work. So they wait. But in reality, social media often rewards process over perfection. Players enjoy seeing: prototypes early ideas weird experiments development struggles These moments make the project feel alive. Mistake #6 — Ignoring Short-Form Video Today, many discovery systems are driven by short video formats: TikTok YouTube Shorts Instagram Reels For games, these formats work extremely well because gameplay is naturally visual. Even very simple clips can perform well: a 10-second gameplay moment a surprising mechanic a funny physics bug Short-form video has become one of the easiest ways for people to discover new games. Mistake #7 — Having No Clear Identity Sometimes developer accounts post a mix of unrelated content: random screenshots occasional updates unrelated thoughts sporadic announcements From the outside, it’s hard to understand what the account is about. Clear identity helps a lot. For example: a VR physics developer a social VR sandbox creator An indie experimenting with weird mechanics When people understand what you are building, it becomes easier for them to follow the journey. Mistake #8 — Forgetting the Community Loop Social media becomes much more powerful when it is not one-directional. Instead of only posting updates, invite players into the process. Simple questions can create engagement: “Which vehicle should I add next?” “What is missing from this map?” “Which mechanic feels more fun?” These interactions help players feel like co-creators, not just spectators. Mistake #9 — Not Creating a Community Space For games, especially, social media is often just the beginning. Platforms like Discord allow players to: give feedback share ideas create content connect with each other Without a community space, many players disappear after discovering the project. With one, they can become long-term supporters and contributors. Mistake #10 — Expecting Fast Growth This may be the most important one. Social media growth is usually slow at the beginning. It often looks like this: Month 1 → a few followers Month 6 → a few hundred Month 18 → real traction Growth compounds over time. Consistency matters much more than quick results. Final Thought If there is one lesson I learned while building games and communities online, it’s this: Social media works best when it reflects the real development journey. Not just the highlights. The experiments. The mistakes. The weird prototypes. Ironically, the moments that feel the least “polished” are often the ones people connect with the most. If you are building something in XR right now, I’d be curious to hear: What social media mistake have you made while developing your project? Let's discuss!I Almost Overdesigned My VR Game to Death
There’s a phase in game development that nobody really warns you about. It’s not the “I can’t code this” phase. It’s not the “I ran out of money” phase. It’s not even the “no one is playing my game” phase. It’s when your own ideas start overwhelming the game. That’s where I found myself recently. I have a Social VR game currently live on the Meta Horizon Store. And this is my story about how design — not bugs — became my biggest struggle. The Dangerous Kind of Productivity After publishing my game (3 months ago), the early months were manageable. There were bugs to fix. Core features to improve. Community expectations were still forming. But as time passed, growth slowed. And I felt stuck. Not because I had no ideas. Because I had too many good ones. New abilities Leveling systems Advanced control modes More immersive camera options Dynamic AI creatures Lore layers Progression trees World events Each one is exciting. Each one is defensible. Each one “adding depth.” And each one is making the game heavier. From the outside, it looked like progress. From the inside, it felt like friction. VR Makes It Worse In VR, every feature multiplies complexity. A new ability isn’t just a new mechanic —it affects comfort, cognitive load, UI clarity, and social balance. A new camera mode isn’t just visual —it changes perception and can introduce motion discomfort. A new progression system isn’t just numbers —it affects motivation, fairness, and retention. Everything touches everything. And when you stack systems without tightening the core, the experience starts to blur. The Subtle Identity Drift The scariest question I had to ask myself was: What is this game actually about? Is it skill-based? Is it social? Is it progression-driven? Is it a sandbox? Is it competitive? Is it experimental? When you add features faster than you refine your foundation, your game slowly loses its center. Not dramatically. Just enough that every new decision becomes harder. That uncertainty is exhausting. The Ambition Trap Overdesign often comes from passion. You care. You want your game to stand out. You want depth. Growth. Surprise. So you build. And build. And build. Until one day you realize you’ve created something impressive… but unclear. Complexity Feels Like Depth — But It Isn’t This was the lesson I had to learn: Depth comes from mastery of a strong core. Complexity comes from stacking. They are not the same thing. A single mechanic refined to excellence will carry a game further than five half-polished systems competing for attention. Especially in VR, where clarity of experience is everything. The Turning Point My shift wasn’t about cutting ideas. It was about asking a harder question before adding anything new: Does this strengthen the core loop? Not: “Is this cool?” Not: “Will players like this feature?” Not: “Will this make the game deeper?” But: Does this make the core experience clearer and stronger? If the answer wasn’t obvious, it didn’t belong — at least not yet. The Real Struggle Isn’t Technical Most developers think the hard part is engineering. In my experience, the real struggle is restraint. It’s saying no to good ideas. It’s choosing focus over ambition. It’s realizing that sometimes your game doesn’t need more mechanics. It needs a sharper identity. The Second Mistake: Retention Here’s something even harder to admit. After refocusing the core, I made another mistake. I didn’t give players a strong enough reason to come back. Clarity alone is not enough. Players need: progression competition meaningful goals something to improve at A strong core gets them in. Retention systems keep them returning. Balancing simplicity and long-term motivation is the real design challenge. If You’re Feeling Overwhelmed If your project feels heavier every week… If every feature you add creates two new design problems… If you keep “improving” the game but feel further from clarity… You’re not alone. You’re not bad at design. You might just be overdesigning. And that’s usually a sign you care. What I’m Learning Simplicity is not a lack of ambition. It’s disciplined ambition. I almost overdesigned my game to death. Now I’m learning that the strongest games aren’t built by stacking ideas. They’re built by protecting the core — and then carefully layering systems that support it. That lesson might be the most valuable part of this entire journey. If you’re building something in VR right now: What are you struggling with the most? Clarity? Retention? Scope? Motivation? Let’s talk.Design a Community Leadership Pipeline: Setup Your Discord Community
So, your game is growing. Your Discord is active. Players are talking nonstop. But moderation feels harder every week? When developers talk about community growth, we usually focus on: Player numbers Engagement Events Social media traction But here’s the uncomfortable truth: Growth without structure becomes chaos. Most communities don’t fail because they lack moderators. They fail because they lack a system. Your Discord Is Not a Chat Room It’s your future leadership system. At the beginning, everything feels manageable: You answer questions. You resolve conflicts. You know most members personally. But as the community grows: Messages multiply. Conflicts increase. Support requests pile up. Toxic behavior finds gaps. Without structure, the founder becomes the bottleneck — and burnout follows. Why “Just Add Mods” Doesn’t Work Many teams respond to growth by promoting moderators reactively: Someone helpful gets promoted. Someone active gets permissions. Someone available gets the responsibility. This creates hidden problems: Authority without preparation Inconsistent enforcement Internal drama Moderator burnout Loss of community trust A strong community isn’t built by adding moderators. It’s built by designing leadership. What a Leadership Pipeline Really Is A community leadership pipeline is a structured path where authority is earned over time. Think of it as a progression system: Member → Contributor → Moderator → Leader Responsibility increases gradually. Trust becomes observable. Power is distributed intentionally. Instead of reacting to problems, you cultivate leaders continuously. Step 1 — Define Authority Levels Clarity prevents politics. Without defined roles, people guess: Who can enforce rules? Who can speak officially? Who resolves disputes? A clear progression might look like: Member Active Member Contributor Junior Moderator Moderator Senior Moderator Community Lead When expectations are explicit, drama decreases dramatically. Step 2 — Make Trust Observable Promotions should never feel arbitrary. Not: “Nice person” “Friend of staff” “Online a lot.” Look for behaviors that demonstrate reliability: Consistent positive activity De-escalating conflicts Helping newcomers Responsible reporting Creating useful content Trust should be earned publicly, not privately. Step 3 — Increase Responsibility Gradually Authority should grow in stages, not leap suddenly. For example: Level 1 → Can guide and answer questions Level 2 → Can mute disruptive behavior Level 3 → Can kick users Level 4 → Can ban Level 5 → Can oversee moderators Gradual power increases reduce mistakes and prevent abuse. Step 4 — Give Leaders Identity Leadership must feel meaningful, not invisible. Recognition systems reinforce responsibility. Effective signals include: Role badges Private staff channels Public recognition Monthly spotlights Early access privileges People protect roles that give them identity and purpose. Step 5 — Write the Rules Systems scale. Memory does not. Every stable community needs a documented structure: Moderation guidelines Conflict resolution procedures Escalation paths Code of conduct If expectations live only in conversations, they will drift — and conflict becomes inevitable. What Happens When It Works A properly designed leadership system changes everything: The founder steps back from daily firefighting. Moderators coordinate independently. Community norms reinforce themselves. New leaders emerge organically. Your community becomes self-sustaining. Common Leadership Killers Even strong communities collapse when these mistakes appear: Promoting too quickly 🚩 Granting full authority immediately 🚩 Lack of written expectations 🚩 Publicly undermining moderators 🚩 Ignoring burnout 🚩 Leadership systems fail when ego replaces process. Building the System in Discord Discord provides powerful tools — but tools alone don’t create structure. Effective setups use: Role hierarchies Permission layers Private staff channels Moderation logs Ticket systems Reaction-based access gates Design the system first. Assign people second. Community Leadership Is a Product Decision This isn’t just a moderation topic. It directly impacts: Player retention Safety Content creation Brand reputation Long-term growth Healthy communities amplify games. Chaotic ones drain them. Final Thought Because the strongest communities aren’t managed. They are structured. 💡 LEARN ABOUT THE META HORIZON START PROGRAM The Meta Horizon Start program provides intermediate and advanced developers with the resources, hands-on support, and expert guidance needed to accelerate their app development. Join a thriving community to get the tools and go-to-market guidance you need to successfully deploy and grow your app on Meta Horizon OS. Apply to Start today: https://developers.meta.com/horizon/discover/programs/startRetention in Social VR: Why Most Indie VR Games Struggle (And How to Fix It)
So, you built your game. It is FUN. You drive players daily. But most do not come back? When we, indie VR developers, talk about growth, we usually talk about: TikTok / YouTube Store visibility Influencers Launch day spikes But here’s the uncomfortable truth: | Growth without retention is just expensive churn. I had the same issue...And over time, I realized something important: Retention in Social VR is not a metric. It is a campfire. Retention is a Campfire 🔥 Think about a campfire. People gather around it because: It is warm It is safe It is alive It gives them a reason to stay If you stop feeding it, it dies. If there is no structure around it, it spreads and burns out. If it’s too small, nobody gathers. Retention works the same way. Social VR Is Not Just a Game — It’s a Place In traditional games, players chase wins. In social VR, players chase: Belonging Visibility Identity Status They are not looking for “levels.”-They are looking for warmth. Retention in VR is driven by social capital, not just progression. The Retention Triangle (How You Feed the Fire) Over time, I simplified retention into three pillars, which are the logs you place into the fire. 1️⃣ Direction (Quests & Challenges) When a player logs in, they should immediately know: “What do I do next?” If your spawn area says nothing…The fire weakens. Direction means: Clear daily goals Short-term progress Visible next steps Daily challenges are not “gamification.” They are fuel. Without small logs, the fire fades. 2️⃣ Identity (Badges & Visibility) Badges should not track progress. They should broadcast who the player is. In social VR: Cosmetic visibility > invisible XP Titles > hidden levels Social proof > private stats When players can show who they are, they protect their place in the circle. Identity makes the fire meaningful. Without identity, it’s just heat. 3️⃣ Rhythm (Live Ops & Weekly Anchors) Retention dies when time feels flat. A strong campfire needs regular fuel. Strong social VR games create rhythm: Daily resets Weekly rotations Limited-time cosmetics Community events Rhythm keeps the fire alive. A game without rhythm feels abandoned — even if it isn’t. The Stone Circle (Systems & Moderation) A campfire without stones spreads. A community without structure collapses. One of the biggest mistakes indie VR developers make: Moderators added too late Events built too late Systems implemented after chaos Unmoderated growth = chaos. Chaos = churn. Systems protect retention. The Social Layer Multiplier Here is where it gets powerful. When players see: Creator tags Event hosts Ranked titles Rare cosmetic holders They don’t just play. They aspire to sit closer to the fire. Aspiration strengthens the circle. And that is when retention becomes natural. What I Changed in My Own Game In my own social VR game, I realized: | Fun mechanics were not enough. So I focused on strengthening the fire: Visible daily challenge boards Public tech tree branches Social nameplate titles Weekly mode rotations Creator spotlight systems Not more content, but more fuel. Retention Is a Design Philosophy You don’t fix retention with a patch. You design your world like a place people want to gather around. You build: Direction Identity Rhythm When those three align, players don’t just visit your game. They sit down. They stay. They return. Final Thought When a player logs out today… Does your campfire still feel warm? Or does it go dark? If the answer is not obvious, your retention system isn’t strong enough yet. — Tevfik Ufuk Demirbaş VR Entrepreneur & Developer & Start MentorGet Started with 6 New AI Building Blocks in Meta SDK v83 | Mentor Workshop
Meta SDK v83 gives developers a faster path to production-ready AI and hand tracking features through drag-and-drop Building Blocks in Unity. In this Mentor Workshop, Start Mentor Quentin (Valem) walks through each new feature with live demos so you can see apply them right away. Every Building Block covered in this session is available now and ready to drop into your projects. 💡 By watching this video, you will learn: How to set up and connect the six new AI Building Blocks for speech-to-text, text-to-speech, LLM, and object detection How the new Fast Motion and Motion Upsampling optimizations automatically improve hand tracking accuracy for high-speed VR interactions How to implement controller-free locomotion using micro-gestures, walking sticks, climbing mechanics, and NavMesh teleportation How to use the Throw Physics Tuner and its preset profiles to stabilize thrown objects without custom physics code Recorded live in January 2026 as part of the Meta Horizon Start program. 🎬 CHAPTERS 👋 INTRODUCTION 🕒 00:00: Welcome & v83 Overview 🕒 01:42: AI Building Blocks Introduction 🤖 AI BUILDING BLOCKS IN UNITY 🕒 09:28: Implementing Text-to-Speech 🕒 11:26: Integrating Large Language Models 🕒 16:00: Multimodal AI with Image Input 🕒 18:50: Building a Conversational AI Manager 🔍 OBJECT DETECTION & RECOGNITION 🕒 23:33: On-Device vs. Cloud-Based Detection 🖐️ HAND TRACKING IMPROVEMENTS 🕒 27:14: Fast Motion & Latency Optimization 🕒 29:21: Locomotion with Micro-Gestures 🕒 33:24: NavMesh, Vignettes, and Comfort Settings 🕒 36:00: Walking Sticks Locomotion Mode 🕒 37:57: Climbing Mechanics & Architecture 🎯 THROW PHYSICS 🕒 43:10: Throw Tuner Profiles & Setup 🕒 46:18: Live Demo with Physics Behaviors 🧑💻 MEET YOUR MENTOR This session is led by Quentin (Valem), one of the most recognized voices in the Start community since joining the program in 2019. In that time, he’s helped thousands of developers master Unity and VFX through his Valem Tutorials YouTube channel, while launching games like Detective VR on the Meta Horizon Store through his own studio. 🌐 VALEM ON YOUTUBE ➡️ Valem Tutorials: https://www.youtube.com/@ValemTutorials 📚 RESOURCES ➡️ Developers Blog: https://developers.meta.com/resources/blog/ ➡️ Meta Quest Developer Hub: https://developers.meta.com/horizon/documentation/unity/ts-mqdh/ 🔗 CONNECT WITH US ➡️ Sign up to get the latest news from Meta Horizon: https://developers.meta.com/horizon/newsletter 💡 LEARN ABOUT THE META HORIZON START PROGRAM The Meta Horizon Start program provides intermediate and advanced developers with the resources, hands-on support, and expert guidance needed to accelerate their app development. Join a thriving community to get the tools and go-to-market guidance you need to successfully deploy and grow your app on Meta Horizon OS. Apply to Start today: https://developers.meta.com/horizon/discover/programs/start
Building the Foundation of a Social VR Game — Movement & Multiplayer in Baby VR Episode 2
So, I’m building Baby VR — a social VR game that I’m developing with the community on YouTube. In the first episode, I talked about where the idea came from and what kind of social sandbox I want Baby VR to become. But before we touch things like game modes, cosmetics, progression, or monetization, there’s something much more important we need to build first: The foundation. Movement and multiplayer. If these two systems aren’t solid, nothing else in a Social VR game really matters. So in this post, I’ll walk you through how I’m structuring Baby VR’s technical core — from modular architecture to Gorilla Tag–style locomotion and real-time networking using Photon Fusion. Let’s dive in. WHY FOUNDATION COMES FIRST In Social VR, players don’t just control a character — they are the character. Every movement, every hand gesture, every head turn becomes part of how other players perceive you. That means two things need to feel perfect from day one: * Locomotion — how you move through the world * Networking — how other players see you move in real time Everything else — game modes, progression, stores, and live events — is built on top of these systems. THE MODULAR ARCHITECTURE (FLAMECORE) Baby VR is built using a modular system I call FlameCore. Instead of one giant, tangled codebase, the game is split into independent modules, where each one does a single job. This makes the project: * Easier to build * Easier to maintain * Easier to scale over time Think of it like LEGO blocks. You can swap, upgrade, or rebuild parts without breaking the whole system. THE 5 CORE MODULES Here’s the high-level structure: * Core Module: Locomotion, VR Interactions, player rigs *Networking Module: Multiplayer sync, player authority, sessions * Gameplay Module: Game modes, rules, sandbox logic * Backend Module: Economy, store, platform services * LiveOps Module: Challenges, analytics, moderation, live events Social VR isn’t just a “game” — it’s a live service, which is why LiveOps is treated as a first-class system, not an afterthought. LOCOMOTION — MOVING LIKE A PLAYER, NOT A CONTROLLER Inside the Core Module, Baby VR uses Gorilla Tag–style arm locomotion.This system is based on an open-source project from the creator of Gorilla Tag, and it changes how players relate to the world: You don’t press a joystick to move You push against the environment You climb, swing, and launch yourself physically This creates what I like to call somatic progression — players don’t “level up” their character. They level up their own coordination and skill. For a social sandbox, this kind of movement makes every interaction feel more personal and more expressive. MULTIPLAYER — HOW SOCIAL VR ACTUALLY WORKS Every player in Baby VR exists as two representations: HardwareRig (Local Player) This is the version of you that lives only on your computer. It reads directly from your VR hardware: headset position, controller positions, and hand movement. Its job is simple: read reality. NetworkRig (Online Avatar) This is the version of you that everyone else sees. It syncs your movement across the network and represents you inside the multiplayer session. Its job is also simple: show reality to others. THE DATA FLOW Here’s the full loop: * You join a Photon Session (the game room) * Your HardwareRig reads your VR hardware * That data is packaged into custom network data * Your NetworkRig sends it to the session * Photon Cloud broadcasts it to all players * Other clients update your avatar in real time The key concept here is State Authority. You only control your own NetworkRig. Other players’ NetworkRigs are read-only on your machine. This prevents conflicts, keeps the simulation stable, and ensures everyone stays in sync. Your head and hands follow the exact same logic — they’re just smaller versions of the same pipeline. META STORE RELEASE CHANNEL To make this truly community-driven, I’ve opened a Meta Store Release Channel. Join here. This lets you access early builds, test experimental features, and see Baby VR evolve step by step. If you join, you’re not just a player — you’re an OG tester helping shape the game from the ground up. CLOSING THOUGHTS This episode — and this post — is about one thing: Understanding how Social VR actually works under the hood. Movement and multiplayer aren’t just features. They’re the language players use to express themselves in virtual worlds. In the next episode, I’ll start building on top of this foundation — VR interactions, sandbox systems, and game modes designed for controlled chaos. If you want your ideas to turn into features, leave a comment, join the builds, and be part of the process. Let’s build Baby VR — together. 👶🚀 -Tevfik
