Two Headed Disaster

Full of optimism, I took Totem Topple down to Brighton last month, and duly presented it at the PocketGamer Big Indie Pitch. The game did not win any prizes and was quite heavily criticised by the judges. Whilst maybe it shouldn’t have come as a surprise, it was still disheartening and led to some considerable introspection.

To give a little background, the Big Indie Pitch involves indie game developers doing a sort of “speed dating”, spending 3 minutes face-to-face pitching, presenting and letting play to one or two judges at a time. Then when the timer is up, swapping tables and pitching to another judge. At the end, the judges confer and pick a winner.

big indie pitch brighton 2015Scene of the crime

Things quickly unravelled once I got down to pitching, with a number of the judges wholly unimpressed. The problem was two fold:

Firstly, going into the event, the game was set up in production mode, so that once the tutorial had been completed, it remembered that fact and didn’t show up on next play through. After the first pitch, it was left down to me, in lieu of the tutorial, to try and explain the game as fast and succinctly as I could. All whilst the judges sat there, determinedly trying to play it.

A good tutorial accounts for the fact different people learn different ways, and some will always want to just jump in and start mashing buttons to see what they do. Unfortunately, this rapidly caused an issue with one of the game’s key features: It’s dynamic AI system.

The game is designed so that when the player is struggling, the difficulty of the enemies drops right off, with just a few especially weak enemies spawning. Provided a player can ride out the last of the difficult waves of enemies from before, they’re given a breather. A chance to recover before the difficulty begins to ramp up again.

At the time, the game did this by just counting the number of heads the player had left alive. The tutorial started players off with 3 extra heads, already putting them above this limit. However, the next play through, which is what the judges played, would only give one starter head.

Cue judges placing a single head and waiting to see what it did. Then when the starter head finally got destroyed, placing another head to replace it. And so on, at painfully slow pace, with the enemies never getting harder and the judges never getting beyond two heads at any one point. In some cases, leading them to write off the game completely as lacking player agency.

We’d had it pointed out before that this “trick” essentially nullified a large part of the game, letting players build risk free to infinity. But somewhat foolishly, we’d dismissed the idea on the basis that people would get bored with that tactic and default to just building quickly to match the “frantic” pace of the game.

Now, at the worst possible time, the same issue, but from a new player’s perspective of trying to learn the difference between each head, was painfully exposed.

We’ve since added in not only a new tutorial system, but also a sophisticated help system for those that like to skip tutorials. I’ll detail those changes in a later blog. And the AI system now has different checks for when the player is struggling, which should encourage players to naturally build up. (Though I’m not going to detail how as that would give the game away!)

And as for the pitch? Suffice to say, spending half my very limited time apologising for the lack of tutorial, and the other half telling the judges they were doing it wrong, didn’t go down well. It wasn’t a great experience, but we’ve learned from it, and the game is now all the better for it.

Android developer working for Crystalline Green www.crystallinegreen.com