RB Whitaker's Entry: Gem Roll

Summary:

Some details on the game I'm making for the competition, as well as progress updates.

page 1 of 212next »

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 19 May 2014 17:38

I got a chance over the weekend to put together some prototypes/mockups during our little informal 8 hour challenge that now has given me a bit of confidence that I'll be able to make the game I've got in my head.

Achievements Unlocked

Time Machine

Leveled Up

Prototype

Moral Support

The Tinker

Are We On?

It's Not Rocket Science

Competitive

The Prototypes

Here's what I prototyped:

• Random terrain generation (which may not ultimately be used, but made it possible to test out some of my other stuff).
• Being able to calculate the height and normal for terrain.
• "rolling ball" physics, to emulate a ball rolling across uneven terrain.
• Rendering the scene to a TextureCube object so that things can reflect the live environment around them, instead of just a pre-made skybox.
• A glass/gem shader, which is an improvement on my reflection shader from my tutorials that includes reflection (fixed, because the tutorial is showing the exact opposite side) refraction, and a bit of glass tinting to make cool looking gems for my game.

The Game

And now a brief description of what I want to make:

In a sentence, I could describe my game as Marble Madness meets Icebreaker, though of course most people probably aren't familiar with Icebreaker.

In a nutshell, you'll control a glass marble that can roll around on a floating platform/surface. Your objective is to collect all of the gems that are scattered throughout the level without falling off the platform to your death, and then escape through a hatch. Assuming I have time, you'll also be trying to avoid enemies, navigate through different types of terrain (slippery snow or ice, sticky mud, and maybe water.

Playing pre-made levels solo is probably my starting point, but multiplayer cooperative and competitive, as well as randomized levels for replay value may happen as well, but all of this depends on how much time I get. There will probably be time pressure of some sort in the game.

First Weekend Goals

I'm hoping to have a simple prototype working by the end of the first weekend. I want that to include being able to roll around (possibly on a flat surface for now, since I already know I'll be able to handle uneven terrain), pick up gems, and go down the escape hatch. The playing area will probably just be a big rectangle, and I'll probably place gems, the player's starting point, and the hatch randomly across it. Death by falling or running out of time would be nice stretch goals.

(Oh, and I'm defining the first weekend for myself as including Memorial Day, since I won't be at work.)

Last edited on 02 Jun 2014 04:05 by rbwhitaker Show more

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 24 May 2014 05:50

I didn't expect to get a whole ton done tonight, and sure enough, I didn't. But I did get something interesting up and running. I've got my little marble/ball rolling around!

A video would be more illustrative, but I didn't want to do a video capture for something so little. Here's a screenshot instead:

Re: RB Whitaker's Entry: Gem Roll

PiscesMike 24 May 2014 06:17

Looking good, RB!

---

"May the mercy of His Divine Shadow fall upon you." - Stanley H. Tweedle, Security Guard class IV, The League of 20,000 planets

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 25 May 2014 07:00

Well Day 2 has come and gone, and I think I made a lot of progress!

In fact, in a lot of ways, the goals I had for this weekend are basically done at this point. Here's what I've added:

• The ground (it's flat and boring, but it's there).
• Randomly placed gems.
• A hatch.
• You can collect gems (and get points!).
• When all gems are collected, the hatch opens.
• When you fall in the hatch, you win.
• When you fall of the edge, you die.
• Shift key makes you move faster.
• Control key makes you move slower.
• Beginning of Xbox gamepad input.
• A shadow under the ball to make it look like it's not floating. (It's just a 3D model with a black, blurred texture.)
• The game times you, but you still don't run out of time.
• Movement is more physics based than yesterday, which is an important step to make it so you can roll down hills and get air on ramps and such.

There's a screenshot below. Again, a video or a download would be better, but it's late, and I'm done for the night.

A few things that are on my mind right now:

• Keyboard input sucks right now. The game pad input is a lot more fluid. I think there are ways to make this better.
• The code is ugly. I need to spend some time cleaning it up before moving forward too much. I can't afford too much code debt this early on.
• There's a bug where sometimes the marble just disappears. Dunno why yet. Probably something to do with a NaN value.

Re: RB Whitaker's Entry: Gem Roll

Marcus42 25 May 2014 14:46

Wow! You've accomplished quite a bit very quickly, RB, and I'm impressed! I didn't actually expect anyone to get a prototype done by the first weekend, much less by Sunday, but it looks like I was wrong.
Keep up the good work!

Re: RB Whitaker's Entry: Gem Roll

TheGeniusDev 25 May 2014 15:47

Congrats, RB! You already have a prototype, and I haven't even begun working on my project (I have an idea, I just haven't talked to my teammates yet)!

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 26 May 2014 05:03

Well here it is, the end of Day 3!

If you haven't started yet, or if you've only barely started, now's the time to jump in and start making progress! There's still a ton of time left, and if you're in the US, you've probably got tomorrow off. Just jump in! For me, all it takes is forcing myself to start writing code for 25 minutes, and after that, I usually have some momentum going and I can keep going for the rest of the day.

With a basic prototype behind me yesterday, I've begun to move on to the most challenging part of my game: the physics.

I've never been afraid of math nor physics, but the physics in this game is haunting the back of my mind. It's a simple game, but there's a lot of math in it that I've never done before. It's the biggest unknown, and the biggest risk.

Because of that, I'm making sure it gets addressed right away. Maybe you're like me. My natural tendency is to take the hard parts like this and find clever ways to avoid doing it. I find other productive things to do, or spend my time researching. I know myself well enough to know that if I don't force myself to get underway with the hard stuff, I may never get to it…

So in light of that, my goal for the end of next week (Sunday night) is to more or less have the game's physics sorted out. It might not happen, and ultimately, if it slips beyond that, it won't necessarily be a deal breaker. But at the same time, I don't feel like that's necessarily unrealistic for a goal.

With my prototype done, I hit the physics hard (no pun intended… aw, who am I kidding? It was intended…) and I've made good progress.

Here's a screenshot of where things are right now:

That doesn't look terribly different from yesterday, so let me explain what I've accomplished.

• Fixed the bug where the marble randomly disappears. Turns out, my rotation matrix was just unstable when the ball is barely moving, which created a bunch of NaNs and blew everything up.
• Did a bunch of code cleanup.
• Added in crates! The model sucks; deal with it.
• Wrote a bunch of code to determine if a sphere and an axis-aligned bounding box are intersecting or not.
• Made it so you would crash into the crates and stop,.
• Made it so you would bounce off of the crates, but still with a lot of bugs…
• Made it so you could make the marble jump. This is probably a temporary feature so that I can actually test the physics. I don't think being able to jump is really going to be a feature of the game.
• Began doing a lot of work to reimplement how the marble interacts with the ground. In the past, I've just said basically, "you're on the ground if the marble is in the range <0, 0> to <300, 300>, and if you're not, you're not." With the crates, there will be more than one platform, and they're not going to always be at y-coordinate 0 (or even flat, for that matter). I need an improved system that allows for general purpose platforms that can hold the ball up, of which the ground is just one of.

Anyway… there's one more day left in the weekend for me. I'm hoping it's as productive as the last couple of days have been.

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 27 May 2014 04:45

Here it is now, the end of Day 4, and the end of the first weekend!

It's been a good weekend, though I can't say what the workweek will hold. There's a tendency for things to get busy, and sometimes there's no time for game dev, and sometimes, going a couple of days without working on a project will derail me. That would actually be really bad for me at this particular moment, as I've got a lot of math/collision/physics stuff in my head. If I walked away from my game for a week right now, I could be in big trouble…

The bulk of my time today has been in massive updates to the way collision detection works. It turns out, I need a lot more than box/sphere intersection detection, because I need to be able to know what face/edge/corner was hit and bounce off of it in an intelligent way.

So hopefully I can still get at least some time to work on this during the week to keep the momentum going.

I tried to make a video today, but the capture I made didn't seem to have that great of quality, and when I uploaded it to YouTube, it seemed much worse. Dunno what to make of that. But no video capture yet. Maybe another day.

Here's another screenshot that looks a lot like yesterday's, only with some artifacts that I used as a sanity check as I was debugging some problems with my collision detection:

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 28 May 2014 05:52

Well not surprisingly, today was less productive, game development wise than the weekend. There's just not the same amount of time to work on it.

I also spent part of the evening talking with PiscesMike and Swatacular in the chat room about collision detection, and I wouldn't trade that for more time on my game. It was a good discussion.

I did do some cleanup of my code, and I think I understand why the marble gets caught on the edge. Screenshot below, though it's not terribly different from earlier versions:

The problem only arises when you're on top of the box, and that's the only place that you can approach a corner from the wrong side. I think what's happening is the sphere is basically crossing the polygon that forms the top of the box a little bit before it crosses the edge or corner. Technically, that top face ought to pick up the intersection detection first, instead of the edge or corner.

What it sounds like I need to do is yet another overhaul on the way the collision detection works. (Third time's the charm?) I probably need to implement it as a swept sphere vs. a polygon, instead of just a plain old sphere. Then I can figure out what order the collisions happened in, and only take the first one.

These are the things I think about in my morning commute…

Well at least I'm getting closer!

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 29 May 2014 04:49

Day 6: The updated physics is coming along pretty well. I've got a swept sphere vs. a point and a lot of progress on a swept sphere vs. a line segment. Both of those are key to making a swept sphere vs. a polygon. Perhaps when I get done, I'll share the code in a tutorial somewhere.

As I started getting into this stuff this morning, I realized really quickly that it was going to be a lot of work, and I wanted to be sure it was going to fix the problem.

So I made a tweak to how the game recognizes that the ball is sitting on a surface (so gravity doesn't take effect) and then set up my physics loop to update 1/10 of the distance, 10 times in each actual game loop update. This has effectively fixed it (now on very rare occasions, the marble still gets caught on the edges and corners). Seeing that, I'm now quite confident that the swept sphere stuff is going to do the trick…

Here's a screenshot I made to test out the physics in a little more depth: floating boxes!

Re: RB Whitaker's Entry: Gem Roll

PiscesMike 29 May 2014 05:16

Wow! Looking awesome RB, cant wait to play!

---

"May the mercy of His Divine Shadow fall upon you." - Stanley H. Tweedle, Security Guard class IV, The League of 20,000 planets

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 30 May 2014 05:47

There's not much to report today. I've been working on the collision detection code, which is progressing, albeit slowly. I just haven't had a ton of time to work on it with work and everything.

In lieu of a screenshot of my game, which actually still looks like the last screenshot, here's a screenshot of Visual Studio with my failing unit test:

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 02 Jun 2014 04:52

How did it get to be the end of Day 10?! We're just about 1/3 of the way through this competition! I honestly don't understand how this happens…

My goal for this week was to get the physics up and running. There was a lot of head-hit-keyboard action, especially over the last couple of days. I was actually pretty nervous that I'd be able to get it working. And to be fair, there is still some little "quirks" (see how I got out of admitting to still having bugs there?) that need to be sorted out, especially on non-flat surfaces.

But they physics is essentially working! And I'm quite proud of that!

For the time being at least, I'm meeting the goals I've set for myself. ("Quirks" aside.)

It really is time to give you guys a demo to play with. Marcus is the only one who has done that so far… I should join in the fun. But it won't be today.

Let's see… what else is there to report?

I've claimed a few more achievements.

Moral Support: Given for leaving constructive comments on other people's game threads. I think my comments are constructive. :D They're not meant to be destructive, that's for sure.

The Tinker: For those times when you spend forever tweaking code before you finally get it right. That's definitely been my experience with this physics stuff. I've been going in circles. I feel like I've reimplemented this whole thing a dozen times now, but I think it's working quite well now, and shouldn't require any significant changes.

Are We On?: For participating in chat. That was easy. I would have had that at the beginning, if it had existed then. (Add more achievement suggestions! We could use some more!)

It's Not Rocket Science: For adding non-trivial physics. Check.

Competitive: You've done more than one competition here. Yep. I did the last one too…

Goals for Week 2

With the physics at least mostly under control at this point, I'll be moving on to other things.

My goal for this week is to make a more "formal" game experience. Adding in things like a menu, high scores, gameplay modes, etc. That means I've got to polish up my GUI framework and try to turn it into something useful.

I have a feeling that might take more than one week, but I guess I'll just have to see where it goes.

My other goal is to get a playable version of the game out for people to play. I want to start getting feedback. I'd prefer to get one out early, and then maybe another one out at the end of the week, but I'll call it good enough if I get at least one playable version out for people.

So there we have it! Let's get Week 2 started!

Re: RB Whitaker's Entry: Gem Roll

rbwhitaker 02 Jun 2014 04:53

Oh, oops. I meant to also post another screenshot, in lieu of a demo:

page 1 of 212next »