[Nekorosu]

So Spelunky was made with GameMaker. It was extremely enjoyable to play which led to the development of the XBox version.

It contradicts your statement "it's unfortunate that the process doesn't result in games that are particularly enjoyable to play.", doesn't it?

Also I'm sure the dev tools don't play such a big part in the gameplay's quality of the result as you think.

[derefr]

Game Maker is scriptable in a Turing-complete language, so of course you can make anything in it--it's just a matter of blind determination. VBA is also Turing-complete, which means you can write your game in Powerpoint. (It's a lot like Flash, in fact.) But I don't see many people doing that, and for good reason. :)

But I think another, more devious way to phrase "it's unfortunate that the process doesn't result in games that are particularly enjoyable to play", is:

"It's unfortunate that the process doesn't prevent more bad games from getting made."

This seems needlessly cruel at first, but seen another way, it's immediately sensible: languages with strong typing prevent bad code from getting compiled. We prefer compiler errors to something that runs, but buggily. Maybe we could rely on tools, of sufficiently high level, to point out things like: "this finite-state-machine representing a game mechanic has a very high connectedness; this could indicate that your game mechanic requires the player to make repeated choices with too many options, which will induce choice fatigue, making them quickly bored of playing."

[Nekorosu]

I can see your point but I can also see a clash with reality because a lot of successful mobile games are built with lua-based game engines. Maybe the clash is due to the fact bad code has nothing to do with the quality of the gameplay and rather simple game with a top-notch gameplay can be build with duct tape and bubble gum.

From my experience the tools which enable makers to create better gameplay are the tools which enable a very fast iterative development process (think live assets reloading, parameters tweaking without recompilation and ultimately code hotswapping).

Talking about strong typing it does help if you need to scale your simple game into something more complex. It also helps a lot if you are going to build something complex right from the start.

I think this link is absolutely relevant to both of our points: http://elm-lang.org/blog/Interactive-Programming.elm

[derefr]

Note that I didn't say anything about the programming quality of the resultant games. Gameplay is a matter of design quality. We can build tools that help us make better games, but they have nothing to do with programming tools (tools that help us make games, better.)

[Nekorosu]

My fault. I skipped the "Maybe we could rely on tools..."

Now I can see your real point.

I tried to read a book on game design which abstracted out the game mechanics into formalized simulation but it was too technical that was taking out the fun from the process so I stopped reading it. The name of the book is Game Mechanics: Advanced Game Design.

But yes, maybe there'll be tools advanced enough to help not taking the fun out.

[Widdershin]

I've made tons of games in Game Maker. GML was my most proficient language up until mid last year.

The tossup with making games in Game Maker is how much you hate the language versus how much you like the standard library. I have encountered absolutely nothing that has the ease of use and breadth of the Game Maker stdlib for making 2d games. Maybe Unity's 2D support will change that, but we'll see.