CarveOut 2160 charges you for boost and only sells it for risk

It is a snowboarding game set off Earth, and the pitch is a tournament with a body count. Humanity has bases in space, the quadrillionaire funding the programme wants a spectacle, and the prize is whatever riches the winner asks for. The developer is blunt that this one is not primarily about winning — it is about surviving the field.

The verbs are the ones you remember from SSX, 1080 and Cool Boarders: carve, grind a rail, launch, spin, chain it into a combo, land it. What CarveOut 2160 does with them is make the trick system pay a bill. Landing something risky earns boost, and boost is what you spend to burst into a shortcut, break through a barrier, or ram a rival off the course entirely.

That is a tighter loop than it sounds. In most arcade racers, style and speed run on separate economies — one feeds a score screen, the other decides the race, and you can ignore either one and still place. Here the score screen is the fuel tank. Ride conservatively and you arrive at the barrier everyone else went through with an empty meter and no way past it.

Then there are the gadgets — grappling hooks, rocket thrusters, more promised — which push it further from a snowboarding game and closer to a combat racer that happens to be on a board. And it is built for a room rather than a leaderboard: four players in splitscreen, eight online, across races, trick events, time attack and more.

Downwell’s Dual Purpose Design argues that the strongest systems hold down two or three jobs at once, and that the overlap is where the speed, the risk and the hard choices come from. Boost here is fuel, key and weapon. Every time you spend it you are answering three questions with one input — go faster, get through the wall, or take somebody out — and the meter is too small to do all three.

How Game Designers Protect Players From Themselves is the tuning problem sitting underneath that. Given a safe, reliable route to victory, players will take it even when it flattens the game, so the exciting route has to be the attractive one. A trick economy only works if the risky line genuinely beats the clean one. If a cautious rider can podium, the stunts become decoration and the whole thing quietly settles into a slower racer.

How Games Use Feedback Loops is the one to watch in an eight-player lobby. Tricks earn boost, boost wipes out opponents, and wiping someone out clears your line and buys you the air to land more tricks. That is a positive loop, and positive loops snowball a lead until the race is settled long before the finish. Whether the person in eighth is still playing a game depends entirely on what pulls the other way — and a splitscreen game lives or dies on that answer, because the loser is sitting next to you.

CarveOut 2160 is made and self-published by [byteparrot](https://www.carveoutgame.com), coming to Windows, with the release date still to be announced. It is listed under Action, Racing, Sports and Indie, which is a fair description of a game that wants to be three things at once.

The multiplayer list is the part worth reading twice: single-player, shared-screen for four, online for eight, both PvP and co-op, plus Steam achievements. Shared-screen support is a deliberate cost — it is real engineering that no store page rewards — and it is also the reason anyone still talks about the snowboarding games this one is descended from.

If the appeal is a trick system that has to carry you somewhere instead of just filling a score meter, Kick is the sibling on this list — a game where the whole job is keeping a football alive on the walk to school, the trick is the travel, and dropping it is the failure state.