Phase 1 was the database phase. For a few years back, I tried to have a somewhat complete database. E.g. the weapon database was supposed to have spears, swords, maces - everything basic upon completion.
The original plan was to close 'phase one', then work on 'phase two' and so on, like proper software engineering suggests. But in my game-making experience, that's likely to mess up in the future.
But no.. the database changes too often, and even now, I'm adding a lot of stuff to it. Choosing to code the database in Lua was the right decision: MMF2 is horribly inept with on-the-spot changes to data, while Lua let me recently add on weird stuff like scaling, energy use, and how blows are deflected.
Had I coded it in MMF2, I'd have to put several versions before it was done. I would have to wrap up the database as a complete "phase", only to stick in several new additions and versions whenever I want to add a change. Lua let me change things on the spot.
What would be a problem is if I decided to change many different things on the spot - if I made a sword, a club, and an axe, instead of going along the entire code with only a single coded club, it means that I'd have to make three times as many changes every time I have to change something.
That said, "phase 1" is practically done, but it's only going to be properly closed once I get to around phase 4 or so without any trouble. I planned it out so as not to experience any trouble once I get past it, but I know from experience that I will and I don't want to redo it all over again just because of a small change.
Synopsis
In plain English, don't complete a plan just for the sake of completing it. Very important advice to the more ambitious game makers amongst you guys 
|
Advertisement
Muz 
Favourite