There was another test play yesterday for A Looking Glass Jukebox; I’ve lost count, but I think it was the 14th one so far. It was the first time there were five players, and it was the first time that a PC had been knocked unconscious. This was only the second time there had been brawling; both times, the players resorted to violence when confronted with robot dopple gangsters who had stolen their identities.

Test play helps refine the rules and teaches me how to GM the game and write scenarios for a musical comedy game. These three aspects feed into each other: rules must be written so that scenarios can be tested, and scenarios must be written so that rules can be tested. One of the earliest scenarios I wrote for the game now plays better after the rules have been refined. Generations of character sheets have been created as the rules are refined.

I’ve played a few comedy roleplaying games in the past, including Paranoia and Toon, but most of my GM and scenario-writing experience is with fantasy games. What do players expect to do as a rock band? What do they enjoy?

“What, you haven’t memorised the rules?” says a player, mocking me as I look up a rule. It takes me several sessions as a GM before I can really run a new game properly, especially a game as different as A Looking Glass Jukebox.

It hasn’t just been testing the rules; I have also been learning how to write what I do in a session as a homebrew GM into a scenario that others can run. This has also been a new experience, even though I’ve written a few convention scenarios; I’ve never played a scenario I wrote with someone else as the GM before. Scenarios rarely survive player contact and must be rewritten after the playtest.

I’ve done a little bit of test playing with some of my regular players, and I’ve run sessions at a convention. At a convention, I managed to recruit two random local guys for playtests, Mark Moncrieff and Michael Monahan. I want to give a bit of a shout-out to these guys, and to anyone searching for their names. They are great playtesters, with a quick grasp of new rules and their implications; they were dedicated to playtesting, not just playing the game. They pointed out problems and suggested options for solutions. Every playtest I had with them left me with a lot of work on the game mechanics, which is exactly what a playtest should do. They would adhere to the game’s direction rather than do the conventional roleplaying moves. I can’t thank them enough, which is why I’m praising them here.

I’ve got three different test-play groups (not including the groups I ran through at Melbourne’s Chaosium-con). Two of the test-play groups are experienced role-players, the third aren’t at all. As I want this game to appeal to the non-core roleplayers, I managed to convince two old friends who hadn’t played any roleplaying games since university to try it. They created a band called Mallum Argentum, played some gigs, wrote and recorded some songs, got a manager, got trapped in a laundromat by sentient washing machines, and released a single; it failed to chart but didn’t lose money. The players had such a good time that they wanted to play another session to find out what happened to the band they created. In the next session, they recorded an album and narrowly avoided turning into a cartoon bubblegum band like the Archies.

More playtests are planned, but I now need to revise the scenario I ran and rewrite the music composition rules based on David Nissen’s excellent suggestions, supported by both mathematical and musical-historical arguments. I hope to finish a beta version soon.