I've been working on the still untitled game for the visually impaired, referred to as the blind game for short, for nearly a year and a half. During this time, this game has taught me so much about game design, at it's deepest level.
As noted in a previous post, when dealing with a game that has no visuals, you lose some of your most basic tools as a designer for conveying information to the player. Where one might otherwise just put something on the hud, or just put a visual queue in the terrain, this is impossible in Blind game. The information is still important and needs to be conveyed somehow though, so you need to come up with clever ways to give the player the information they need to play the game.
This is what is at the core of this game, and any game really. The gathering of information. The difference between a good game and a great game is the effectiveness of the methods used to convey information to the player.
Think about old NES games for a moment, the ones that were less good and usually extremely hard didn't do a good job conveying information to the player. Some of them depended too much on instruction manuals to tell the player how to play the game, rather than through clever design. Others didn't even make an attempt to convey the information at all. Granted, some of these games were bad and hard because of other things, but often a lot of the problem just came from the player being unable to understand what the developer was trying to convey. Can you think of an old game that you were good at as a kid, but can't for the life of you figure out as an adult? This might be because as a kid, you were patient enough to sit there and figure out what the designer wasn't doing a good job conveying to you without even knowing it.
Or think about modern games. Some of these have the opposite problem, where they overload the player with information, whether it be through excessive exposition of the story, or through conspicuous, overwhelming tutorials that cause the player to tune out, making them just as bad as not having tutorials at all.
Expressing information in clever ways is absolutely vital to the design of a great game. It's sorta like sending an encrypted secret message to someone. Make the cipher too easy to crack, and it will be too obvious and easy to crack, and if you encrypt the cipher too well, then no one will get it.
In this case, making the message too obvious is like having annoying tutorial popups and really excessive plot exposition, while too encrypted would be hiding the instructions to playing your game somewhere outside the game entirely like in an instruction manual that, as time passes, inevitably gets lost.
The key to this is evenly spreading just enough information through the game in such a way that the player can always find it if they need it. Use clever design to give them just enough to figure out how to play the game, then allow the player to find more advanced tutorials in the game, but make them optional.
The Beginners Halls in Final Fantasy VII are a great example of this. The visuals of the game are easy enough to understand that you can make it through the first bombing mission without any trouble, with NPCs dropping hints only where you need them, then if you want everything explained to you in detail, you can go to the Beginner's hall in Sector 7 to learn more. Even the Materia tutorial on the second day is completely optional. Just enough information is given to the player to get by, with more being available on request. Another example would be Metal Gear Solid with the Codec. You would recieve manditory calls, but the information recieved there was important, and if they player wanted more information, they could call for it at any time.
In the case of the Blind game, this is a much harder balance to strike. The entire game is information gathering without sight, so almost every action you take feeds back information on some level. Finding the balance point where the player is getting enough information is much different. After all, a picture is worth a thousand words, but all I have to express ideas to the player is words, and no pictures. This skews the balance.
What this means is making all methods of information gathering as tight as possible, making sure the player only gets feedback that is information dense and useful. Again, ideally this is the case in any game, but this is even more true in Blind game.
Right now we're dealing with balancing this, making it so the player has all the information gathering tools they need to complete the game, without overloading them. In recent testing (the game works brilliantly on OSX, by the way), I thought the game had enough information based on testing I've done so far. I found that this might not have been the case, and me and my friend talked it out for a while and came up with some great new ideas that I'm pretty excited about figuring out.
Besides all this philosophical musing, we recently straightened out a bunch of little bugs in the code, and now the game more easily transitions between levels. Level editing has also become significantly easier. I've found that a lot of optimization I did over the summer is now paying dividends, making level editing much easier, and saving me a ton of time.
Finishing the first level (which still probably needs some tidying up quite frankly) took well over a year to complete. There is now a second level that is nearly done, and including bug fixes, that has only taken about 2 weeks. I have ideas for two more levels as well, so we're really starting to make some progress.
I'll continue to update here as I continue to make progress.