Back in 1994, my 17-years-old self started making his third game, inspired by Empire Software's game DreamWeb (whose demo he had deeply enjoyed) and by Kim Stanley Robinson's novel A Short, Sharp Shock (whose Italian translation was titled Anello intorno al Mondo, which Ring around the World is the literal translation of).
Like the previous two games (one of which has been remade from scratch and released in 2021 with the name Follix), it was made with AMOS Professional.
It remained unfinished: it was just too big and beyond my capabilities. The graphics looked ugly, the code was even worse, there were only a couple of horrible sound effects, only one map was (kind of) complete and two other maps were basically placeholders.
However, that game always held a special place in my heart.
Fast-forward to the fall of 2022 and one evening, while talking about books, I mentioned the novel and then the game to a friend (a former Amigan). The next day, a gust of nostalgia prompted me to fire up the game and wander around the playable - and terribly sparse - map. Immediately, the memory of a shameful graphical bug returned: after a few steps, I performed the action that would bring it up. Through the years, such scene repeated quite a number of times, but I never got around to fix the issue because it was caused by the conceptually wrong graphical system, so a solution required a major rework - not something an unfinished game was worth. Actually, in 2000, I did start a remake, but I stopped quite soon before completing the new scrolling routines because, when compiled, the game would crash - the AMOS Professional Compiler did not get along with the custom equates I had defined for the game. Disappointed (to put it mildly), I refused to remove the equates (which would make the code less efficient) and I simply abandoned the project. This time around, I could no longer suppress the urge to fix the defect for good, so...
Fighting the disgust, I waded through the entangled code, figuring out how it worked, reimplemented the graphical system and finally got rid of the bug. But I did not stop there: I fixed also other issues and refactored, optimized, cleaned up and made error-proof the code - and even standardized the naming conventions and removed the pieces of Italian scattered here and there.
I could well have been happy with that, but the ball was rolling and I could not stop it. I pushed things further and replaced the scrolling engine several times until I settled for one that moves the world view by tiles and substituted the quasi-real-time mouse-driven movement with automatic destination targeting - given that the protagonist's movement was locked to tiles by design since the beginning, the original 2-pixel per update scrolling was just a drag. Also, I redesigned entirely the interface - promising myself I would not touch the rest of the graphics.
Once again I could (and should) have stopped, but there was still a major issue: the graphics were in 32 totally wasted colors! There were two blacks (one for the screen border color and another one for the non-transparent black pixels of bobs), a color was reserved for the main bob, a color was unused and many colors were very close to one another. That annoyed me so much that I broke the promise and decided to reduce the colors to 16. I carefully defined a palette of handpicked colors, redrew the interface (multiple times) and recolored the rest of the graphics by means of an automated process.
The automated recoloring results were worse than expected - and that started one more iteration of the remake. I had to repixel the graphics: I knew it would be a lot of work, but it could not be avoided. While at it, I changed the screen layout entirely by making the inventory vertical and moving it to the right side of the screen, so that the world view could be bigger and square - previously, in fact, moving vertically was uncomfortable due to the limited height of the view.
Countless improvements and additions followed, until the game reached the current state - and became suitable for public release, which had not been even contemplated when the work started.
In the end, nothing remained of the original material and the last challenge was to answer the question: where to stop? The remake was not planned, stole time from other (and way more important!) projects and, to become a sufficiently large game, would have required months (or even years) of further development. Therefore, I decided to make it a mini-game restricted to the (questionable) map of the original work, keeping its spirit intact.