[In this honest indie postmortem, Wadjet Eye Games' Gilbert discusses the creation of adventure game sequel The Blackwell Convergence, looking at everything from promotion through art direction and engine to analyze what went right -- and wrong -- in the game's creation.]

The Blackwell games are point-and-click adventures made by my small indie game studio, Wadjet Eye Games, that center on an awkward writer named Rosangela ("Rosa") Blackwell who, upon her aunt's death, inherited the family ghost. The ghost in question is Joey Malone, a sardonic spirit from the 1930s. Their task is to assist lost souls and investigate supernatural goings on.

The Blackwell Convergence marks the third game in the series, and after three games I have realized that sequels are a very funny thing. On the one hand, you have the opportunity to enlarge a franchise and keep your characters in the public eye.

You've learned from your mistakes, and can apply your newfound knowledge to making the next game better. But you also run the big risk of alienating your old fans in the hope of courting new ones. Is change always good? Well, read on, and see how I fared.

What Went Right

1. Evolving Rosa's Character

When I wrote the first Blackwell game, my goal in creating Rosa Blackwell was to make a character who was hurt by her past and thrust into a situation she had no interest or desire to be in, but made the best of it and emerged a better person because of it.

For the most part, people liked Rosa. They liked her awkwardness; they liked the fact that she was a relatable human being with flaws. Unfortunately, because of these qualities, they also found her very difficult to play. She was often so socially awkward that she became a hindrance rather than a help.

There was a key moment in the first Blackwell game where Rosa needs to speak to a woman named Nishanthi. Nishanthi is in the park, playing her flute in front of a crowd of onlookers. A normal person would just walk up to her and say "Excuse me" but Rosa is too shy to approach her in front of a crowd. Instead, you have to solve a puzzle in order to draw Nishanthi away from the crowd so Rosa is comfortable enough to talk to her.

As a character study, it worked. It's a natural thing for Rosa to do, and it really hammered home her sense of isolation and cemented her as a social misfit. But for a player involved in an interactive experience? It was frustrating. he original game was littered with moments like these and I wanted to fix that for Convergence.

But, how was I supposed to "fix" Rosa without completely changing her character? Simple. I didn't change her. I just put more focus on her positive traits than her negative ones. Rosa is the bookish sort. She's a writer and a reporter, which makes her intellectual and very observant. She knows when people are acting suspicious or when they are lying. This is a side of Rosa that we didn't see in the first game, so I made a point of showing these traits in the sequel.

Plus, the sequel takes place six months after the first, so she has had the time to mature and get used to her new supernatural abilities. Rosa is still very awkward in the game (and it provides the game with some of its more funny moments), but the awkwardness takes a back seat to her other qualities which get a chance to shine.

When the game was released, the change in Rosa's character was instantly noticed by reviewers. Some criticized it for being too much of a drastic change, while others felt it gave her a much-needed edge. I don't think either is entirely accurate. For me, she is still the same Rosa Blackwell -- just viewed from a different angle.

2. Using AGS as a Development Platform

It's a very good time to be an indie, as there are countless free third party tools that you can use for development. As for myself, I use AGS. Short for Adventure Game Studio, it is a third party engine geared toward the creation of old-school adventure games. It is a system that has been tested and refined for almost 10 years by many users, so it one of the more reliable tools out there.

Using a mature existing engine like AGS took a ton of the grunt work out of development. Being familiar with the system I was able to do 99% of the programming myself, saving a lot on development costs.

It also enabled me to prototype game events very quickly so I (and QA testers) was able to see what worked well. If something didn't work well, it was a quick matter to make adjustments. The only disadvantage of the system is the lack of portability, so Convergence will never play on a Mac anytime soon -- but the benefits more than made up for it.