itch.io

Here are some general tips. Keep in mind that I'm just a humble game designer. I'm not claiming to be an expert on the topic, and I want you to treat these more as guidelines than as rules. If anything I say strikes you as just flat out untrue, then feel free to ignore it.

1. Make a plan. Don't just dive right in and hope that you'll make something good. Spend at least the first hour of the game jam brainstorming ideas. Usually, the best idea that I come up with isn't the first one that pops into my head.
2. Manage your time carefully. Make sure that your game is playable and that the core idea of the game is fleshed out by the end of the first day. I know that sounds like a lot to do, but you'll be so glad you have that extra 24 hour buffer of time when you're trying to add artwork, sound effects, etc.
3. Pick something simple. Don't try to make the next smash hit MMORPG. You only have 48 hours. That goes by way faster than you might imagine. On that note...
4. Don't do a 3D game. 3D games require more texture work, more lighting work, more edge cases while programming, more difficult level design, more difficult camera work, and the list goes on from there. Unless you're a very experienced game developer and you can hammer out 3D games faster than sonic can run, just stick to 2D. You'll thank me later for that one.
5. Consider the 2:44 ratio rule that indie developer Vlambeer came up with. I use this rule for every game jam I participate in, and it always works out in my favor. The idea is that you have 48 hours to make a game, so you want to manage your time well. Vlambeer suggests spending the first two hours brainstorming, deciding on the right idea to pursue, and then actually finishing a playbale prototype of your game idea. Make sure it works, make sure it's fun, and make sure that you're happy with what you have after the first two hours. If you are, then you can spend the other 44 hours working on game feel, artwork, music, sound effects, designing levels, and polishing the mechanics you've already implemented. Additionally, if you're unhappy with the result after 2 hours, you can just scrap it and work for another two hours on a different idea. In the end, you only lost two hours, and you'll still have 42 hours for all the other good stuff!
6. Don't burn yourself out. Take good care of yourself physically. Get sleep. I know you only have 48 hours, but if you don't get between 6 and 8 hours of sleep at night, your brain can't work efficiently. You'll be so much more productive if you sleep well than if you try to work for 48 hours straight. Also, don't eat pizza and Mountain Dew. Eat fruits, vegetables, and drink lots of water.
7. The right tool for the job is whatever you're most comfortable with. During this jam, you might hear a lot of people tell you to use this engine or that engine. Don't listen to their advice. There's a good argument for using just about every engine in existence. The best engine for this game jam is whichever one you're most comfortable with using, and whichever one you think you'll be the most productive with. Speaking of engines...
8. DO NOT WRITE YOUR OWN ENGINE. Writing a game engine is an extremely complex task, and is often a multi-year project that requires lots of maintenance as the user base expands. I don't know anybody who could write an entire engine over a single weekend. Don't be afraid to rely on tools that other people have made. The goal is to make a game here, not to re-invent the wheel. If you're unsure about which game engine to use, at least start with one of the free ones: Unity, UE4, Xenko, Godot, Game Maker Studio, Construct. Xenko and Godot especially are free and open source, so there aren't any features stuck behind a pay-wall.
9. Make sure your game compiles and runs as a stand-alone application at least 24 hours before the deadline. There's nothing worse than spending 47 hours working on a game just to find out that you can't build the project into an executable during the last hour and spending that whole last hour trying to get it to work. There are too many submissions each year that simply don't run or crash immediately when you open them, and I just think that's tragic. Read up on the documentation for your engine of choice and make sure you understand everything you need to do to get a build working. It will save you a lot of stress. Additionally, most engines don't just deliver a stand-alone application when you export your project into an executable. They often create an executable *and resource packages* containing all your game's assets and other data. You need to include all of these when you submit your project, or it often won't run. On the note of creating executable builds:
10. Consider making a browser-playable version of your game. This is often well within the realm of possibility for a 48 hour game-jam game. Most modern game engines these days support exporting your game to HTML5 with little-to-no effort, and itch.io supports HTML5 games pretty seamlessly (even being able to pick up on which engine you used). The nice thing about browser games is that people have no excuse not to try your game. It's right there in browser. No downloads, no installation, no resource packages. Just open the page for your game and start playing. I think you're more likely to get feedback if your game is playable in browser.
11. Make (or download) sound effects for your game. SO MUCH of what makes a game feel good to play is the sound design. Unless the game being silent is specifically part of the design, silence will make your game feel lifeless. Most sound design tools are pretty expensive these days, but you can use free tools like BFXR to generate sound effects (hell, I use BFXR in almost every game jam I participate in). Also, consider using freesound.org, where you can download hundreds of royalty-free sound effects for free and use them in your game with no problems. Finally...
12. Remember to enjoy the process. Presumably, you're a game designer because you enjoy doing it. Don't be so focused on the end result that you forget to have fun during these crazy 48 hours. I'm not going to lie to you: your game will probably not make it into Mark's video. The GMTK jam is always the largest game jam on itch.io every year. I think there were well over 1,000 submissions last year. Odds are, Mark won't even have time to play your game. So, my advice is not to focus on that. Just focus on making a game that you're proud of, and enjoy the hell out of it, dammit!