Megabonk looks like a joke until you check the receipts: a solo developer built a 3D Vampire Survivors-like in Unity, launched it 14 days after Silksong, sold over one million copies in two weeks, peaked at 117,336 concurrent players on Steam, and then withdrew from a Game Awards nomination because it did not feel right. The short answer is that Unity was the only sane bet for that story.
Vedinad started Megabonk in August 2024 and shipped for Windows and Linux on September 18, 2025 with 20 unlockable characters, 70+ items, and 200+ quests layered over procedurally generated 3D maps with verticality, sliding, jumping, and speed boosts. That scope — third-person 3D, auto-attacks, XP gems, random upgrade offers, Silver meta-progression — maps directly to what Unity is good at: cross-platform 3D with a C++ runtime and C# scripting, an editor that a single person can actually finish a game in, and instant ports to PC without hiring a port team. Unity describes itself as a democratizing engine for indie developers, and Megabonk is almost a literal illustration of that pitch.
The comparison that matters is not Unreal Engine 5. It is Vampire Survivors. Luca Galante started Vampire Survivors in Phaser while unemployed, proved the auto-combat loop, then had to port the game to Unity from version 1.6 onward to handle performance as the team and content grew. Megabonk skipped the painful middle step. By starting in Unity, Vedinad bought headroom for hundreds of simultaneous enemies, 3D physics, and vertical maps from day one, at the cost of a heavier base build than a web-first Phaser prototype would have had. For a solo dev planning 20 characters and 70+ synergistic items, that trade was correct: one rewrite avoided is a year saved.
Brotato is the other useful contrast. That 1-million-copy survivors-like bet on Godot and GDScript for fast 2D iteration and 20-90 second waves. Megabonk bet the opposite way: Unity for 3D space, longer runs, and movement verbs. Godot optimizes for iteration speed in 2D; Unity optimizes for not rewriting when you add a Z-axis, slopes, and jump arcs. If your game can be drawn on graph paper, Godot is lovely. If your dodge depends on a ramp, Unity pays for itself.
What Unity bought, and what it cost
What Unity bought was time-to-3D. Procedural maps with verticality, third-person camera, CharacterController-style sliding and jumping, and auto-attack projectiles all exist as documented Unity workflows. Vedinad did not need to write a renderer, a physics broadphase, or a Linux exporter. Starting in August 2024 and shipping 13 months later with daily-buildable Windows and Linux versions would be borderline impossible on a hand-rolled C++ engine for one person outsourcing only music.
Unity also bought content throughput. Twenty characters with unique abilities, 70+ items with synergies, 240 in-game quests cited on Steam, and Silver currency earned via challenges, silvery-blue pots, and Tome upgrades are a data problem, not a graphics problem. Unity's prefab, ScriptableObject, and C# tooling culture lets a solo dev treat characters and items like spreadsheet rows and test a broken combo in seconds. That is why Megabonk feels overpowered in a fun way: the engine lowers the cost of adding one more ridiculous synergy.
The cost was ugliness and suspicion. Eurogamer's first impression was blunt — rough-looking, mobile-like padlocks gating powers — and Steam reviewers initially read the crude art as cheap. Unity's democratization cut both ways here: players have learned that a rough Unity survivors-like might be asset-flip shovelware. Megabonk had to overcome that prior with feel, not screenshots. The second cost was performance risk. Unity can push 100k concurrent players' worth of clients, but only if the solo dev does the unglamorous work of pooling projectiles, culling off-screen hordes, and keeping the 3D physics step stable. The 117,336 peak is proof the work was done, not proof it was free.
Silksong makes the timing legible. Silksong is also a Unity game, released September 4, 2025 for nine platforms and selling over seven million copies by December 2025. Megabonk was originally planned for September 4 and slipped to September 18 after Silksong claimed the date. Most commentary reads that as fear. I read it as distribution judo.
My take: verticality is the difficulty slider, movement is the weapon
Here is where I depart from the usual Megabonk coverage, which stops at derivative-but-fun. Most survivors-likes scale difficulty by adding more HP and more numbers. Megabonk scales it with slopes.
In 2D Vampire Survivors, positioning is defense: you kite in circles while builds do damage. In Megabonk, a ramp is both cover and trap. Enemies path differently up and down inclines, XP gems roll, line-of-sight breaks, and a slide down a hill can save more HP than any armor item. That means Unity's Terrain and physics are not set dressing; they are the balance spreadsheet. Vedinad moved part of the design burden from item numbers into level geometry, which is cheaper to tune for one person than rebalancing 70 items every patch. If you think of verticality as a free difficulty slider that never needs a tooltip, Megabonk's 3D choice stops looking derivative and starts looking economical.
Related idea: movement is the real weapon. Auto-attack removes aiming, so most survivors-likes risk reducing the player to a build picker. Megabonk re-injects skill through slide, jump, and speed boosts. A well-timed jump over a boss charge or a slide through a gem cluster is DPS, because faster pickup is faster leveling is faster offers. In other words, Vedinad used Unity's CharacterController as a damage item. That is a subtle break from Galante's design, where movement speed is mostly a stat. Here movement execution compounds with stats.
Anonymity worked the same way. Vedinad spelled backwards is Danidev, voice comparisons to missing YouTuber DaniDev fueled Reddit threads, and Vedinad declined a video call and later declined to answer Are you Danidev directly. Most coverage treats this as gossip. I think it is infrastructure. A solo dev with 920,000 copies at about £8.50 cannot afford a community team. Mystery offloads lore, marketing, and even QA onto volunteers. Like a Unity Asset Store plugin, the community fills a system the dev did not have to build — in this case, narrative and hype. Silence became a server.
The Game Awards withdrawal fits that logic. Nominated for Best Debut Indie Game for 2025, Vedinad withdrew, saying Megabonk is not a debut because he made games under other studio names, urging voters to pick another debut title. Host Geoff Keighley honored the request and removed the game. The press framed it as humble. I think it is credentialing. By refusing debut status, Vedinad signals veteran status without doxxing past work. In a market where Unity shovelware has trained players to distrust newcomers, I am not new is more valuable than a trophy. Losing an award on purpose bought more trust than winning it would have.
Silver is the quietest smart system. XP floods you; Silver starves you. Earned slowly via challenges and pots, Silver forces quest variety and elongates the game — the exact padlocks Eurogamer first disliked. Idle games shower meta-currency to retain; Megabonk throttles it to prevent binge-and-quit. On Unity, where it would be trivial to sell boosters, choosing scarcity is a design stance: brake, not accelerator.
What indie devs can steal
Start on the engine you will need at 100,000 enemies, not 100. Vampire Survivors paid a mid-life port tax moving Phaser to Unity; Megabonk prepaid it by starting in Unity for 3D from August 2024.
If you are solo, move complexity into space, not stats. One ramp that changes pathing replaces ten spreadsheet tweaks across 70 items.
Give auto-combat a manual verb. Slide, jump, and dash cost almost nothing in Unity and restore skill expression without breaking auto-attack accessibility.
Treat scarcity as content. Silver-style throttling plus 200+ quests stretches 2 maps further than a third biome would.
Do not fight Steam traffic; surf it. Shipping 14 days after Silksong looked suicidal, but it placed Megabonk in front of millions of players already opening Steam that month. Delaying further into October AAA season would have buried it.
Browser trainers that teach the same muscles
You do not need to install Megabonk to feel its loop. These browser games isolate one Megabonk skill at a time and run instantly with no download:
For the 10-minute survivors loop — leveling, spell choices, and surviving escalating waves — try 10 Minute Mage. It compresses the level-up-every-30-seconds pressure into a short run, so you practice offer triage the way Megabonk demands it.
For 3D horde spacing with jumping and sprinting under pressure, try Against Zombies. It is a 3D shooter about kiting undead hordes while defending an escape car, with WASD movement, sprint, jump, and weapon switching. Use it to drill the slide-and-jump repositioning that matters more in Megabonk than raw aim.
For build synergy without reflexes — mixing items, skills, and traits into an overpowered squad — try AutoRPG Arena. Its turn-based auto-battler loop rewards the same collection-and-optimization thinking as Megabonk's 70+ item synergies and Silver unlock order.
Play one 10-minute run in each before your next Megabonk session and you will notice the transfer: faster offer reads, calmer kiting on slopes, and less impulse-spending of scarce meta-currency.
Boundaries and open questions
Unity version, render pipeline, and netcode details for Megabonk are not public; Vedinad has not published a tech postmortem as of November 2026. Sales beyond one million in two weeks and the 117,336 SteamDB peak are snapshots, not lifetime totals, and Linux/Windows-only scope means console port costs remain untested. The Danidev identity theory — Vedinad as anadrome, similar voice, overlapping August 2024 timeline, declined video call — remains unconfirmed because Vedinad stopped replying when asked directly, and should be treated as speculation, not fact. My verticality, movement-as-weapon, anonymity-as-infrastructure, and withdrawal-as-credential readings are interpretations built on verified systems and sales, not developer-stated intent.
If you are choosing an engine for a survivors-like today, the lesson is narrow: pick Unity if your fantasy needs a hill. If it needs only a horde, a lighter 2D stack will ship faster. Megabonk needed the hill.
