Yes, I realize all of us are very excited on this project, and yes we do have some very nice ideas, but before we get any more excited, I think for the benefit of this project, we need to limit the scope of this project as far as the technical constraints are concerned.
I would like to put it through a programmer's point of view, since we don't have much technical people in this project, unless someone else is profficient in object oriented programming concepts such as Java / C++ which is the fundamental skill based to use Game Maker efficiently.
Whatever core mechanic of whatever game we might think of, if you cannot explain it in simple terms using a straight forward verbs, then most likely the algorithm to implement it will be a bomb.
For instance, say a simple scroller like mario brothers.
Mario can:
- Walk in two directions (left and right)
* If keyboard is left, then move left, draw animation
* if keyboard is right, then right, etc
- Jump in three directions (upper left, up, and upper right)
* if jump + left is pressed, then upper left jump
* etc
- Dash while moving left or right
* if dash + left is pressed, then draw dash animation, increase speed
* etc
- Dash while jumping to get extra height
- Shoot fireballs (if you get the powerup)
- Jump on enemies,etc
Now for each actions, the game engine needs to check:
- Collission detection with walls, empty space, enemies etc
- Collission detection of fireball
- Coliission detection of Mario's Jump on the enemy, etc etc
As you can see, even for a simple scroller with "walking around and jump" as the core mechanic, when it comes to actually implement it, be it with game maker or from scratch, it requires a lot of work and polishing up.
Now imagine if we have a writing recognition as the core mechanic of the game. Game maker only provides simple keyboard/ mouse events, such as key pressed, mouse pressed etc.
While we can use the mouse events for drawing, the algorithm for writing recognition involves comparing each pixels drawn on screen with the intended matching object, for instance if you want to draw the letter A on a canvas sized 40 x 40 pixels, then for each of the pixels drawn, i have to iterate how many percent of it lies within the area I consider necessary to form the letter A. If the percentage is more than say 90%, then i consider the drawing as the letter A. If not, I will reject it. Also, for each objects, we need to keep a database of how they made up the 40x40 pixels for comparison purpose. Everytime the player draw something on the screen, this whole process of calculating the pixels drawn and comparing each pixels with a database of objects, could be quite expensive in terms of game performance. Now in case you don't understand what I am talking about, then you get the idea how complex it is to write a seemingly simple action into a fully functional program algorithm.
But isn't writing recognition just another way of input? I mean, Instead of writing recognition, the same effect could be perfomed by, key sequences.
I remember playing tekken with their 10 hit combos, involved a very long key sequence like XXOXOXOXOXOAAALRLR etc. That puts me off, and I stick to my simple 3 hit street fighter combo. The mechanic of kicking and punching is the same in any fighting game.
What I am trying to say is that, whatever ideas we put, please keep in mind that at the end, we need to actually program it.
Programmers are not Magicians that can pull rabbits out of the hat. In other words, if you can write in paper step by step how each actions are to be performed as detailed as possible, then the programmer's job is to convert these steps into mathematical formulas necessary to simulate the process and come up with the results. This is the very reason why computers are not good for tasks that require human senses, like cooking, face recognition etc, and I won't go into that in details.
So yeah, although Game maker brings us back to the age of SNES, I believe that even with the time and manpower constraints, we can still come up with nice, simple, innovative gameplay, which is sellable, event if it doesn't have superior graphics, story, or innovative control systems. In fact, If we can pull it off, it'll be quite impressive i guess?
Just my two cents =)
