Student Game Dev Part 24

Student Game Dev Part 24 - The End

Soan

New Zealand194 Posts

July 19 2014 05:24 GMT

+ Show Spoiler [Previous parts] +

The End - A Post Mortem
Hello again TeamLiquid! Welcome back to my blog about the development of The Adventures of Sam the Pirate, the 2D platformer I'm creating as the final game project for my Bachelor of Software Engineering degree. This post is pretty overdue, as I finished my degree, and thus this project, a couple of months ago now. Since then, I've been distracted with shifting house, looking for a job(recently started work at Outsmart Games), taking a bit of a break, and otherwise procrastinating. I'll be doing a bit of a post mortem about the project, as it has since finished.

What went right
Overall, I am pretty happy with how the game turned out. There are still a couple of minor bugs, but those who have played the game generally seem to enjoy it. It's not as complete as I might like, due to having to cut a lot from my original plan, but for a solo project it's worked out ok. Well, apart from the art side of things, but I'll go more over that later.

The programming side of things mostly worked out how I hoped it would. Some of the code is definitely super inefficient (each individual tile is its own physics agent! ugh, so bad), but that's more down to me trying to get things done fast to meet deadlines more than anything else. And it works, which is the most important thing. Of course I encountered various bugs, other things I hadn't coded quite correctly the first time, made mistakes, but I managed to fix pretty much all of them. There's quite a lot, probably nearly all of it, that I would also like to rewrite, both to fix inefficiencies, better ways of doing things I thought of while working on it, and a few other things.

I would particularly like to rewrite the rendering to use OpenGL. Currently it's just using SDLs built in 2D rendering, which was nice and easy to setup and to use, but I don't feel like I have many options in how it anything works. I'd also like to learn OpenGL at some point anyway, so this seems like a good project for that in the future. I'd also like to replace my custom physics with Box2D. My custom code works just fine, it's a little buggy in places, but it works no problem.

Speaking of the physics, it works just fine, but is a little buggy. For me personally, the biggest success around the physics was getting it to be calculated at a fixed rate, independent of the rendering frame rate. The rest of it was fairly simple, basic rectangle/rectangle collisions, detecting which direction and adjusting position appropriately. Originally, it was being calculated at 60fps, so every 1/60th of a second the physics would be calculated. Later on in the project, when I was putting together the levels in my editor, doing things quickly by making each tile it's own physics agent finally came back to bite me in the ass and caused some performance issues with the number of tiles I had on screen in a couple of levels.

I had a couple options to fix this, either cut out some of the tiles, or lower the frequency of physics calculations. I ended up going with lowering the rate of physics calculations to 30fps as it was the fastest change, and I was pressed for time. This worked out well, but did cause some of the collision detection and response to be a bit buggy occasionally, mostly when falling. If you fall far enough, it's now more possibly that you could go through the top tile and land on the second. It's not game breaking, or that common of an occurrence from what I've seen, but it does happen occasionally.

The level editor I created work out fairly well, allowing me to put levels together infinitely faster than having to lay out co-ordinates etc by hand. It's not particularly user friendly, and has its quirks (you can place tiles with the menu open), but it worked nicely for me.

What went wrong
Probably the biggest problem I had, was the lack or art. While I had a few artists express interest in working on the game during pre-production, apart from a few pieces of concept art, one set of tiles (which I ended up not using), and a single animations, I got nothing. Looking back, I really should have either pushed the existing artists to actually create stuff, or found others more interested in doing it much earlier. I ended up getting some art off OpenGameArt.org, and one of the art tutors helped out by providing a couple of animations, although I think I'm still missing one or two. The only upshot about not getting any art is that being a programming degree, what the final game looked like was kind of irrelevant as it didn't count towards any marks. Still would have been nice to have the game look a bit better though.

The other big problem I had was time. Namely, there wasn't enough to do everything I wanted to achieve in the project by myself. To be fair, I did kind of expect this, as for one person it was a fairly large project, so it wasn't really a surprise that I did not get to do everything I wanted. I just knew that I wanted to give it a shot, so that's what I did. In the end, I had to massively cut my original plan, down from 40 levels, to 20, to 10, then down to however many I could get done, cut the enemy types down to a single one, no bosses, there isn't even really an ending.

The Game
You can download the game here! It supports Xbox 360 controllers, and other controllers might work too, I haven't tested with any others. No guarantees on performance either, on slower/older machines it might be a bit crap.

What's next?
I'm going to be looking at redesigning my blog, WordPress was great for getting it set up quickly, but I'd like to write it myself. Not for any particular reason really, I just want to do it, so I will. Once I'm settled into my new job a bit more, I'll be looking at creating some small mobile games in my spare time, and seeing what happens there. I have a few ideas, so we'll see how that goes. I'll still try to write blog posts, but they'll be far less frequent than they have been during the production of Sam the Pirate.

Keep in touch
If you're interested in following any of my future work, you can follow me on Twitter, and like the Facebook page.

Peeano

Netherlands5685 Posts

July 19 2014 18:32 GMT

Looking forward to playing. ETA 40 min.
No mention of music or sound effects.
Would you like critics? Just wondering, because you said it is now finished.

Did you play Shovel Knight yet? Seems like you would enjoy that.

Peeano

Netherlands5685 Posts

July 19 2014 19:23 GMT

What did grade did you get?

I'm now playing and I'd say one little improvement you could/should have done, is add in controls in the main menu (at the start up of the game). The ones you listed in Readme.pdf

Edit: Attack+Jump spam to fly through the levels :D

BisuDagger

Bisutopia19369 Posts

July 20 2014 01:15 GMT

Hey you finished! I'll check it out tomorrow. Congrats man. I felt so rewarded when I finished my first game.

Soan

New Zealand194 Posts

July 22 2014 21:12 GMT

On July 20 2014 03:32 Peeano wrote:
Looking forward to playing. ETA 40 min.
No mention of music or sound effects.
Would you like critics? Just wondering, because you said it is now finished.

Did you play Shovel Knight yet? Seems like you would enjoy that.

Hmm yeah it seems I totally forgot to talk about music/sound effects. They also went pretty well, although I ended up with more music tracks than I could use.

Sure I'll take critics, feedback is good for possible future versions.

No I haven't played Shovel Knight yet, got too many games - not enough time as it is.

As for grade, I got a B+.

Please or register to reply.

Student Game Dev Part 24 - The End

Completed

Ongoing

Upcoming