Digital Sun Games, the studio behind Moonlighter, has scaled back the first playtest for its next game. ReVamp’s test was meant to be public. It will now run on the studio’s Discord from 29 October 2026, with players let in over several waves. The studio hopes to open a public test in January. ReVamp is due on Steam on 25 March 2027.
The reason is new, and it affects anyone planning a demo or playtest this year.
Why Digital Sun pulled back
In a statement on Steam and Bluesky, reported by Game Developer on 19 October, the studio said recent cases worried it. Games have been reverse-engineered, cloned and published before the original came out. With launch still months away, putting a large public build out felt too risky. The studio said it hopes platforms keep strengthening protections for original work. It also said early showings help indie games find their players, which is why the call hurt.
80 Level’s write-up gives the same reason and adds the practical details. To join, you pick the Playtests role in the studio’s Discord, and testers are added in announced waves.
Digital Sun isn’t alone. Game Developer reports that another team cancelled its Steam Next Fest demo the day before for similar reasons.
What’s changed
Clones aren’t new. Threes and Ridiculous Fishing were both copied on mobile more than a decade ago. What’s changed is the cost. Earlier this month we covered AI agents rebuilding whole games from their compiled binaries in weeks (AI can now decompile games. What that means for indies). Game Developer also notes developers on Reddit reporting that people have fed demos, trailers and screenshots into AI tools to rebuild a game’s core loop.
There’s also a legal gap. Copyright generally protects your code and your art. It does much less for your mechanics. A cloner who rebuilds the loop with new art can be hard to stop, even when the copy is obvious.
So the risk is real. It’s also uneven. A cloner needs a game that’s simple to copy, has a clear hook and will sell under someone else’s name. Most indie games never attract that attention. For them, the bigger danger is still that nobody plays the demo at all.
Your options, from most open to most closed
You don’t have to choose between a public demo and no demo. There’s a range, and each step trades reach for control:
- Steam Next Fest or a permanent demo. The most visibility you’ll get, and a proven wishlist driver. The build is public and anyone can download it.
- Steam Playtest. A separate free app attached to your store page. Players request access and you decide how many to let in, so you can open it slowly. The build still ends up on strangers’ machines.
- Keyed or Discord-gated tests. What Digital Sun chose. You know who has the build, and you can revoke access. Reach is limited to people who already follow you.
- Supervised sessions. Testers play over a stream, or on a build you host and they control remotely. Almost nothing leaves your machine. It’s slow and it doesn’t scale.
Each step trades reach for control. Figure: Indie Tooling
Whichever you pick, a few build habits make copying slower and leaks easier to trace:
- Ship release builds only. Strip debug symbols and remove dev consoles and cheat menus.
- Pick the harder target where your engine offers one. In Unity, IL2CPP produces native code that is harder to read than the default Mono build. It won’t stop a determined team, but it raises the effort.
- Cut the demo to the hook. Show the first 30 to 60 minutes. Keep late-game systems, progression tuning and your best content out of the build.
- Watermark keyed builds. A per-tester ID baked into the build tells you where a leak came from.
- Add an end date. A demo that stops working after a set date limits how long a copy stays useful, though a skilled cracker can remove the check.
A smaller test has an upside too. A few hundred keen players on Discord often give better feedback than thousands of strangers who download a build and drift off. You can ask follow-up questions, run quick polls and see the same people come back for each new build. What you lose is wide-audience data, such as how many newcomers finish the first level. That’s the data Digital Sun said it wanted from a public test, so its January build still matters.
What we’d do as a solo dev
For most solo developers, we’d still run a public demo. The wishlists and feedback from Steam Next Fest are hard to get any other way, and a game nobody has heard of is a poor clone target.
We would change what goes into it. Keep it short. Ship a release build. Hold back the systems that took longest to design. If you’re on Unity, switch the demo to IL2CPP. Keep a dated record of your development, such as source commits, devlogs and store page history. If you ever need a platform to take a copy down, proof that you got there first helps.
The maths shifts if your game’s appeal rests on one mechanic that’s easy to describe and quick to build. Small mobile-style games and short arcade loops are the classic examples. In that case, a staged approach like Digital Sun’s makes sense. Test with your community first, keep the reveal close to launch and open the public demo later.
Digital Sun has 147 days between its first playtest and launch. That gap is long enough for a fast copy to appear, so the studio has chosen to show less for now. If your gap is shorter, or your game is harder to copy, you can afford to show more.
Top comments (0)