← the game the redshift search bench

The redshift search bench

The graph from the game, with the guard rails off.

You do not need data of your own to be here. Pick one of the four cases below and it loads a real galaxy from the study this game is built on, with its real frequencies and the real windows it was observed through — then drag the redshift until something locks. Nothing is scored, nothing is unlocked, and the answer is not written down anywhere: it is the thing you are looking for.

If you do have data, that is what the panels are for: type in the frequencies you detected, build the tuning you have (or the one you are thinking of proposing), and read off what fits. Every tone, candidate and colour comes from the same engine the chapters use — a port of the paper's own Python, RSG.py, checked value by value against the original. Copy link puts your whole setup in the URL, so a question is something you can send to someone.

Turn the sound on — the button is at the top right — and drag. A predicted line approaching a detected one beats against it, slower and slower, and goes still at the match; release there to lock it. Once you are on a candidate, the button under the graph offers to check the other lines that redshift predicts. If it predicts something loud and you never saw it, that candidate has just died.

What this does not check

The bench trusts you, and the things it trusts you about are most of what an observer actually has to know. It does not check that your frequencies are real. It does not check that your windows could be recorded in one pass — a receiver can only be tuned so many ways, and a set of windows that looks fine here may be several separate observations, or impossible. It does not check that the atmosphere would let you: the game's oxygen line at 118.75 GHz is one closed door among many, and above about 350 GHz the sky is only open on good nights. And it does not know how faint your galaxy is, which is why it asks you how deep the observation was rather than working it out.

Somebody has to carry all of that, and that somebody is the astronomer. Which is roughly the job: the tool does the arithmetic, and you are responsible for whether the question was sane.

And afterwards

Say you pin a redshift down here. In real life that is a beginning, not an end: it tells you where in the history of the universe this galaxy sits, which is what makes every later measurement of it mean anything — and it is usually the argument for asking for more telescope time, which is a written proposal, judged against everyone else's by a committee, for a telescope that is oversubscribed several times over. The RSGplot() line above is genuinely how a figure in that argument gets made.

Method: Bakx & Dannerbauer (2022), MNRAS 515, 678 — the paper this game is built on, and yes, astronomers publish the code alongside the writing. Line frequencies, the CO ladder, the candidate search and the exclusion argument are the paper's; the ALMA tunings come from its giveALMA. Scoring a tuning against a whole population of galaxies is a different question, and that one is Chapter 9 of the game.