Game About Squares

Getting Game About Squares to actually run without wasting your afternoon

I spent about three weeks with this thing before I figured out the basics, mostly because the official documentation is either missing or actively misleading depending on which mirror you're reading. The core loop is straightforward enough—arrange colored squares into target patterns within a grid—but there are several system-level behaviors that trip people up. Here is the download situation first since that is where most beginners get stuck. The game does not have a single authoritative distribution point. The Steam version works fine if you can get it, but the standalone executable from the indie publisher portal is lighter and runs better on older hardware. The itch.io listing has the Windows build, Linux port is community-patched and sometimes broken after updates, and the macOS version tends to have rendering issues on Apple Silicon unless you enable Rosetta. If you are just testing the waters, grab the itch.io Windows build and skip everything else until you are invested.

How Game About Squares actually works in practice

The grid system uses a coordinate-based layout. Each level gives you a set number of square pieces, each with a fixed color and a maximum placement range. Your job is to position every piece so the final arrangement matches the target pattern shown at the start of the level. You can rotate pieces by pressing R, and you can swap two pieces by selecting both in sequence. That is most of the interface right there. What people miss initially is that piece range matters more than piece color. A red square placed at the edge of its range behaves identically to a red square placed at its center. The only thing that changes is whether adjacent pieces can interlock properly. I learned this the hard way on level fourteen, where I spent twenty minutes trying to force a perfect color match and could not solve it, then moved one piece four pixels toward the edge of its range and the entire puzzle snapped into place. The game does not tell you this. It just judges the final output against the target. There is also a hidden constraint most guides overlook. Once you lock a piece in place, it cannot be unselected until you restart the level. This means your ordering matters. In later levels with twenty-plus pieces, placing the most restrictive pieces first—usually the ones with the smallest range and the most unique colors—saves significant time compared to working randomly.

I ran into a specific edge case on level forty-seven that almost made me quit. The target pattern included a color that matched one of my available pieces exactly, but the piece had a slightly smaller range than what the visual layout suggested. I kept trying to force it into position and failing repeatedly. The workaround was to place all the other pieces first, leaving that problematic square for last. By then, the remaining space fit its restricted range perfectly. This happens in maybe three levels throughout the entire game, but if you hit it, remember that last-piece strategy. The difficulty curve is not linear. Levels one through thirty ease you into the mechanics with generous piece counts and large ranges. Then levels thirty-one through fifty suddenly restrict your pieces significantly while increasing the target complexity. This is where most people stall. The sixty levels after that are mostly variations on established patterns with minor twists, so if you push through the middle section, the endgame is relatively manageable. Performance-wise, the game runs at a consistent sixty frames on integrated graphics, but the Linux port drops to around forty on the same hardware due to how it handles the rendering thread. Not a dealbreaker, but noticeable if you are playing for extended sessions.

There are no multiplayer features, no leaderboards, and no in-game hints. The hint system is purely external—there are Reddit threads and Discord servers with spoilers for specific levels. If you are the type who likes to solve things independently, ignore those. If you are stuck on a level for more than twenty minutes, checking a solution is usually faster than continuing to fiddle with piece placement. The game is priced at eight dollars, which is fair for roughly ten hours of content. It is not long, and it does not reinvent the puzzle genre, but it executes its narrow mechanic cleanly. If you enjoy tight spatial reasoning games with minimal friction, it is worth your time. If you need hand-holding or progression systems to stay motivated, you will probably find it too bare-bones.