A tenth of a second is the whole clue
Songless gives you a tenth of a second of a track and nothing else. No intro, no chorus, no title screen. Either the timbre lands or it does not, and the game is built on how often it does.
A tenth of a second carries no melody and no lyric, but it still carries production, vocal texture and mastering.
Songless removes everything except those first impressions and asks you to name the record from them.
A clip that short holds one attack, one timbre and one room. You are not hearing the song, you are hearing how it was recorded. That is why a 2010s pop track and a 1970s rock track are usually separable instantly even though neither melody is present.
The library is large enough that random guessing is useless, so the game lets you narrow it first. Choosing a genre, an artist or a decade changes which songs can appear in the run. Picking a decade you know well is the difference between a fair test and a coin toss.
Extending the clip would turn identification into a different skill. At a tenth of a second you cannot wait for a chorus or a signature riff, and you cannot search by lyric. The restriction is the game: it rewards the part of memory that stores sound rather than words.
Solo runs move at your pace and let you sit with a clip as long as you like. Live rooms put the same clip in front of everyone at once, so the answer is a race rather than a test. Both use the same library, which means a song that stumps you alone will stump other people too.
The library, the filters and the replay are the whole toolkit.
The clip loads instantly and replays as often as you want. Repeating it rarely helps, but the button is there for the tracks that feel familiar without resolving.
Genre, artist and decade narrow the pool before a run starts. A narrow filter makes the game a memory test; a wide one makes it a lottery you can still win.
Because no melody survives the cut, the game leans on vocal grain, drum sound and mix character. Headphones expose those details far better than a phone speaker.
In a room, everyone hears the identical fragment and answers at the same time. The score reflects who recognised it first, not who was given an easier track.
Pick a pool, listen, answer.
Choose a genre, an artist or a decade. Leaving every filter open makes the clip almost impossible to place.
A tenth of a second plays once automatically. Replay it if the timbre is familiar but the title is not.
Pick from the list and the run moves to the next clip. The next song is drawn from the same filtered pool.
Three players on the same short clip.
The clip is too short to hum, so I stopped trying to sing it back. Listening for the drum sound alone doubled what I get right.
I filter to one decade before a run. With the pool at a few hundred tracks the game becomes a real test of what I actually know.
Losing a room by half a second stings more than losing solo. Everyone heard the same fragment, which removes every excuse.
What players ask before the first clip.
Pick a decade, listen once, and see whether the answer arrives before the second replay.
Try a clip