I asked a few weeks ago for a suggestion on what to write a simple game in, and it seemed SFML was a good choice. Started making a little 2D concept engine and it's been going fairly well... It has a pretty simple map, texture and sprite loader. I can move around a player object (sprite) in the middle with the arrow keys. The movement is pretty clunky as the player moves tile by tile. What should I work on next? Should I mess with collisions or animation first? Neither? Started working on it a bit today and realized I had no plan of action... I have the weekend to myself and want to put a good chunk of time in...

Help please holy computer spell writers!

Also Happy Halloween.

Happy Halloween to you too.

At this point I usually do collision. Which is pretty easy when you do tile by tile movement; just check to see if the tile type is one the character can walk on.

For fun, make special items that let you walk on exotic tile types. Although you'll have to do inventory at the same time to get that running, that's easy too.

How about making the character be draw at a pixel location (pixelX, pixelY) instead of a tile location (tileX, tileY), and then add collision to it?

Note: If you decide to just keep it at tile movement, collision is way easier. So maybe just add collision and stick with tile positioning.

Thanks guys, about to start on collisions.

How about making the character be draw at a pixel location (pixelX, pixelY) instead of a tile location (tileX, tileY), and then add collision to it?

Note: If you decide to just keep it at tile movement, collision is way easier. So maybe just add collision and stick with tile positioning.

I'm very tempted to draw pixel by pixel for the player. Obviously it's smoother and I like smooth. In all honesty I haven't ever worked on collisions. Is it going to make it much more difficult? If not, then I'll change it right now.

EDIT:Didn't realize you answered everything I just asked haha!

Sorry long day. Thanks again!

I'm very tempted to draw pixel by pixel for the player. Obviously it's smoother and I like smooth. In all honesty I haven't ever worked on collisions. Is it going to make it much more difficult? If not, then I'll change it right now.

It's significantly more difficult for your first game. Collision is no longer a mere matter of checking the next tile to be moved into with an if, but involves a bit of math.

You can smooth out tile to tile movement by animating between (have the character take a couple steps into the next tile).

There are advantages to stopping on tiles, in that you know exactly where the character is for interactive animations between it and enemies/the environment.

Other games fix this by forcing the character into a certain position before beginning an interaction, but this is yet another thing to do.

It's up to you, but I'd stick with tile movement first.

I'm very tempted to draw pixel by pixel for the player. Obviously it's smoother and I like smooth. In all honesty I haven't ever worked on collisions. Is it going to make it much more difficult? If not, then I'll change it right now.

It's significantly more difficult for your first game. Collision is no longer a mere matter of checking the next tile to be moved into with an if, but involves a bit of math.

You can smooth out tile to tile movement by animating between (have the character take a couple steps into the next tile).

There are advantages to stopping on tiles, in that you know exactly where the character is for interactive animations between it and enemies/the environment.

Other games fix this by forcing the character into a certain position before beginning an interaction, but this is yet another thing to do.

It's up to you, but I'd stick with tile movement first.

Thanks will do! Grabbin' that two o'clock coffee and going for it.

You can also use the type of tile that the character is moving into (and/or out of) to determine the animation to be played (like climbing if it's a ladder, or swimming if it's water), as well as the amount of time it takes to move into that tile.

It's nice to provide more complex collision, where moving on a certain material is less of a yes/no proposition, and more of a strategic decision of speed and movement efficiency. This can add a lot of nuance to your game, particularly when you bring in combat, and it's relatively easy to accomplish.

There's also one-way collision, which in a game like this, might be short cliffs which can be jumped off of but not up onto.

You may also want to get into tiles that push the character onto another tile (blowing wind, flowing water, or falling down a higher cliff).

But, that also depends on what kind of game you're planning to make.

Personally, I like to put more work into the interaction with the environment because it gives you a lot of bang for the buck.

I agree that tile based collision is a good next step. But I suggest maybe holding off on smoother movement for a bit.

- First maybe you could add some other people or critters of some kind and have them move about randomly but not through things they shouldn't. It just generally adds some life to your game.
- Maybe after that you could add a number of coins or something to random places on your map. If the player collects / collides with one remove it from the game. If all are collected, game over. No real need for inventory just keep a count of coins collected.
- Maybe after that you could have the critters you added earlier move towards the player if within say 4 tiles. If they collide with the player then game over.

I'm thinking that at that point you have a really basic game to continue building on. It really doesn't take much to getting sucked into doing feature after feature with no gameplay of any kind in your project (which is the problem I have at the moment).

I have two suggestions.

I find that sometimes, especially when making games as an act of learning, simply having a vague idea of what you want your game to be will sort of push you into the direction you need to go to learn something else. Let's say you eventually wanted this to be a zelda-esque type game, like others mentioned you'll want to add some more characters, then learn how to handle collision, then perhaps learn how to calculate projectile spells. You'll want to add in attack and damage. You'll want to learn how to save/load the game, etc. But, everything you learn, you'll find something else you'd like to add to it as well, and if you ever run out of things to learn, well, then your game is likely done

It's a little haphazard, to be sure, and obviously having some sort of idea what order to learn things is a pretty good idea, but I find that flows kind of naturally. Obviously, you'd want to add NPCs before implementing collision, and set up your attack/damage functions before adding in projectile attacks. You should be pretty able to determine the order of what to learn, and it becomes apparent pretty quick if you're learning things out of order. It's pretty difficult, in my opinion, to learn things out of order, as if you don't know the prerequisites for how to do something, you'll know you need to step back and learn those first. I've spent many a weekend relearning mathematical concepts I'd forgotten

It doesn't have to be the most original or mind-blowing idea for a game, but perhaps flesh out a vague outline for how you'd like it to end up. I'm still in this stage myself (the learning stage, I mean), but just giving yourself an end-goal of sorts will direct your learning process to a pretty good degree and ensure you're never out of things to do next.

The second thing is to keep a TODO list of sorts for your game and add things to it as you come across them. When you find yourself with time and unsure of what to do next, you can go over it and pick something that fits your mood I find sometimes I'm in the head space to tackle a difficult challenge, but one that doesn't have much of a visual reward (optimizations, things heavy in math and such), and other times there are really simple things with a large payback game-wise. But, I find it's really easy to forget what things you can work on if you don't keep track of them.

Anyhow, welcome and happy Halloween!

I didn't get to spend as much time as I would have liked this weekend but I still have a few solid hours left. Last night I was able to get some simple tile collision to work, so at least some progress was made. The end goal that I have in my head, although vague, is exactly what has been guiding my todo lists direction! Funny you mentioned almost exactly that. I'll start penning some more solid end goal ideas out... Thanks for the feedback!