I'm currently developing a good ol' 2D tile engine with the aim of using it for a platform game and it's coming along nicely. So nicely, in fact that I think I should set about deciding what the game requirements will be before I get too deep into the engine. As a result, I've been bashing around a few ideas and I'd be grateful if any of you guys could throw in your opinions. No design doc yet, just a brainstorm. Inspired by a previous thread started by Saluk I like the idea of an Indiana Jones style setting for the visual style (hope you don't mind mate). However, in terms of gameplay I'm looking for something a little different and gained a spark of imagination from elastomania. The idea of that game is to traverse maps on your motorbike as quickly as possible. This is where I thought, "why not have a level full of traps and hazards with a fairly non-linear layout. The players aim is to traverse the level as quickly as possible to reach the goal. No monsters, no pickups, just a race against the clock." Now, I'm fairly happy with this idea, unless someone can suggest a major flaw, but I'm still mulling over the finer details. What makes elastomania so addictive is the possibilities the physics provides so I need something which is similarly 'easy to learn yet hard to master'. I've had the following thoughts and would love some criticisms / suggestions. - Player movement. Simple left/right/jump jobby but a sprint key is provided. Just like track and field, repeated battering of the key makes you run fast in the direction you are currently facing. Better players can make bigger jumps / go faster. Should I limit the use of this, would it just be annoying or would it add tension/excitement? - Grappling hook. The first game I ever made was a simple two-player platformer using grappling hooks. They were basic but added a huge amount to gameplay and emergent properties. I've never attempted spring physics in a game but would be interested in trying if the idea could work. If not, then perhaps just a simple circular movement representation. Activated by pressing jump in the air and it shoots at a fixed 45 degrees. - Wall jumping. The more abilties avaiable the more possibilities. Think Mario Sunshine. - Gliding. Not sure on this one. But the ability to glide for a short length of time. - Landing jumps. When falling from a large height the player will stumble upon landing. However, if they press sprint a split-second before landing, the character will roll and continue running un-affected. With these inital thoughts in mind I can envisage some fun to be had. Taking a massive run at a spike pit by hammering sprint, launching into the air, gliding to the edge, wall-jumping off, grappling a nearby ledge and swinging to safety - all with the pressure of time on you. Any thoughts would be appreciated folks. cheers. Extra thought: Maybe I could just sack off the timing stuff and make the challenge be navigating the free-form levels?

I am right now working on an article, that I will hopefully send to Gamedev, dealing with "creating difficulty" in a 2D platformer. You might find it interesting, if I ever get it done.

ToohrVyk

Sounds like a great idea, with the main focus being on tight control of your character and the many things you can do, combined with good physics. If you get that element down, the content side should be pretty easy. The physics could be difficult to work out though, I think you definately want them to be a bit more arcadey than elastomania, because speed is such an issue and you don''t want it to be too rediculously difficult; but they have to be substantial enough that unexpected things can happen. SUch as if you bounce off a wall just right, you might go into a crazy unexpected spin. I''m not sure about the tap throttle, but if you make it so you don''t need god-like tapping skills to go full speed, it might work. THat could be part of the difficulty options perhaps.

Sounds like good fun though.

Hmmm, yeah. The idea of of the track and field style sprint would be as an extra to standard movement so you can always run at a normal speed without it. Still not sure as to quite how well it work work though. I suppose I''ll have to take the old Molyneux approach - "The best way to see if a component works is to get it in your game as soon as possible."

Maybe I''ll just wing it. Cheers for the input chaps.