Out of the 38 games I produced in the 1990s, I only finished 30. (If you want to get technical, I finished 31 then unfinished one.) Of the eight unfinished games, four had an actual chance at completion. I had the whole game planned out, and most of the infrastructure in place; all I needed to do was build the levels.

One of those four is a Blaster Master tribute called Rōdïp: Rover of the Deep. It came out of a sort of competition with another Game-Maker user, to design the best Blaster Master pastiche. His game hewed very closely to the source material. I took the basic concept and a variation on the vehicle design, and I wandered.

The plan was for six main levels, five bosses, and four vehicle upgrades. I finished everything except the levels, the bosses, and some elements of the presentation. Then I got distracted, and the game sat on the shelf for 15 years.

A few months ago I dug it up again, and I realized that the game ain’t half bad. For a Game-Maker game it controls unusually well. The existing monsters and background tiles are distinctive and meticulously built. The existing levels were pretty good, if rough around the edges — and certainly unlike something I would design now.

So I tidied up the levels, making sure that platforms and monsters were placed sensibly, and then I began to design a new one. The slot was already there; I just had to fill it.

In the game’s structure, level 3 is the first new level after the player gains the hover upgrade, allowing much more free movement through the terrain. It made sense, then, to capitalize on that new element and create a free-flowing map, not so dependent on platforms as on environmental hazards and barriers.

There is a contrast, though. For all the player’s new freedom, in a way the new map should be more constrained than ever. One concern is to give the level a sense of structure. It means nothing if the player can go anywhere, if there is nowhere specific to go. So throw in some long vertical corridors, or awkwardly connected rooms, to underline the potential of the player’s new mobility.

The other concern is to give the player a challenge that offsets the new ability. You don’t want the game to get easier just because the player is more powerful. Sure, earlier sections become a cinch — but a new rule should change the game’s focus, and give the player something new to master just around the time the player starts to feel comfortable. So the new level should present new sorts of problems that can only be solved with the new concepts at play.

None of this was a big priority. I have articles to write, a life to attend to, skills to learn. I don’t need to spend too much time messing with old, abandoned projects. So maybe once a week I would spend an hour or so tinkering with the map, adding another screen or two. It has taken a few months. The other day I finished the level. And here it is:

The entrance is toward the lower middle; the boss door is toward the upper left.

It may not be the strongest level in the world; much of it was improvised, rather than planned out deliberately. Still, that improvisation was informed by certain principles and, I think, a pretty good sensibility. Which is to say, I rather like it.

I’m uncertain whether I ever will finish the game; it’s been a decade and a half, so there’s no rush. And now that I’ve finished this level, I think I may have satiated my interest for the moment. Further discouraging me is that, after all, this is a pastiche. If I could take the rover out of the game, and turn the project into something wholly original, then maybe I would feel less reluctant. And yet then, the game would lose much of its identity — so there’s no point.

Again, though, I think the game is pretty decent for what it is — even half-finished. You can play it here, if you like. Press F6 and select the appropriate slot to skip straight to level 3.

Alternatively, here‘s a level 3 playthrough on Youtube.