The only pre-compo works that will be in my submission is just setting up my environment and creating a shell for the game. This includes the splash screen, menu screen, level screen, and end/credit screen. They have no content in them; they are just empty shells to begin the competition with.

If this will be problem, just say so and I’ll start from scratch.

Source for the empty screen classes can be found Here.

UDPATE: Screw java webstart; I’m also having all sorts of last minute issues with my engine choice; gonna play it safe for my first LD48…

Language: Java (LWJGL + Slick2D + marteengine) GameMaker HTML5

IDE: Eclipse GameMaker IDE

Graphics: Inkscape + Gimp

Audio: Ardour + sfxr

My original plan was to use Java(LWJGL+Slick2D+MarteEngine) for the compo. I made it through all last week getting my environment all set up and getting ready to go. Less than 6 hours before the start of Ludum Dare 23 I decided to update Eclipse and move to Java 1.7. In the process I completely botched my environment and could not get it put back together again in time to be dependable (and who wants to spend the entirety of a compo troubleshooting non-game issues?).

So about 2 hours before the start of LD#23, I decided to go see what other things were in my toolbox that I could pick up quickly. I first went to my various Torque engine licenses. I’m quite comortable in the Torque engines, but one problem; no real web deployment options. And since my Java environment wasnt’ stable enough for me to depend on Webstart, I decided to keep looking. I had once tried a demo of GameMaker 5+ years ago and decided to go see how far along it has come. And wouldnt’ you know it, they have an HTML5 friendly version now. So I whipped out the credit card and decided to give it a shot.

I spent a good chunk of last night and this morning just getting used to the product. By mid-day today I felt comfortable enough to start giving my game a solid go.

For now I have a controllable character; some basic physics; and my artwork started. I’ll keep banging away today until I finish the character logic and get started on the baddies. That should leave tomorrow for sound and polish (and time permitting: music).

–homepage–
 

I find myself routinely “playing” with my game when I’m working on other problems. I decided to see how many AI I can drop on my world and maintain 30FPS. It seems that magic number is ~2500. I suspect that number may go down one my AI learns perform actions other than wondering the globe.

Facts

• My first competition ever
• This is the furthest I have gotten in developing a game, ever
• I had planned to use Java(LWJGL + Slick2D + MarteEngine) for webstart capabilities but the my dev environment didn’t survive my meddling with version updates on Friday. I ended up switching to GameMaker:HTML5 2 hours before the start of the competition
• I had less than 10 hours of experience with GameMaker spread over 6 years prior to the start of LD23; ~6 hours 6 years ago, and 2 hours the last two hours before the start of the competition
• I just started using Inkscape 1 week prior
• I’ve been on the fringe of gamedev for many years, but have never really done much more than quick prototyping/PoCs
• The increasing popularity of time-lapses of game-jams are what drew me to participate in LD23.

What Worked

• Surprising, switching engines 2 hours before the compo started was the best thing I could have done.
• I was able to pick up and get comfortable with GameMaker quickly. What I didn’t know was easily googleable(word?) with only a few rare exceptions
• I loved the work-flow of GameMaker. Being able to put concepts together in the project tree and to visually build game logic was a nice way to prototype idea. The transition from taking those drag and drop items and moving them into local and global game scripts worked very well for me.
• Inkscape is a fantastic tool, even with it’s bugs.
• s/bfxr SORT of worked, although it getting newer generation sounds was a bit tough. It’s still a fantastic tool and great for 8/16bit sounds.
• WolframTunes (even though this isn’t really composition and moreso luck).
• Using the game environment for scrore feedback (lights on the flag bases) rather than puting in GUI overlays.
• Using OneNote-type tools to track concepts and helpful web links with notes (I used Evernote for this, but OneNote via skydrive is nice too).
• Staying social during the competition (I kept on top of twitter, IM, the LD site, my own site, and Reddit; all with focus on GameDev and LD23).
• Letting the theme drive the project. I did no brainstorming at all before the competition and allowed concept drive my creativity once it was released.
• Music without vocals. I listened to a lot of Clint Mansell, General Fuzz, and John Williams(classical guitarist) throughout the competition. I didn’t try to stay awake beyond reason; had I done so, I would have been listening to Infected Mushroom really late.
• Submitting the game in a buggy/incomplete state. The feedback on even broken games will make you a better dev.
• Drawing tablet. Even if it’s a cheap tiny pad, your developer art will be faster and look better than mouse art.
• Lots of PGTips Tea; ~20 cups.
• Having a very supportive and understanding significant other

What Sucked

• Finding bugs/nuances in GameMaker, but not knowing the engine enough to know it was a bug.
• Breaking my AI in the last few hours of development.
• Not being able to have the whole basic concept working in time for submission
• Not realizing I had another hour of development available because I thought the game had to be submitted by the end of the 48 hours and not just that development had to stop.
• Being so fatigued by the end that I had stopped saving my GameMaker project about 3-4 hours before the end; so I do not have a saved state of the project for the final game.
• Not having time to put in GUI’s

Lessons Learned

• Make sure your dev environment is ready before competion and LEAVE IT ALONE
• Know your engine/language so well that you can have a fully functioning basic game in the first 24 hours of the competition so that you can spend the 2nd half on art, polish, and
• Have enough discipline to see a problem through, but not so much that you spend TOO much time on it.
• Take regular breaks and stretch

Submition Page – My Blog

So I’ve uploaded and submitted my entry for the Charity Game Jam. It was a fun experience and I learned quite a bit, but the game is far from complete. But considering I didn’t know about the competition until about 8 hours after it started, I think I did quite well.

I just submitted my LD25 Warm-Up entry, Robozerkertron. It’s a quick little game that plays similar to the old Berzerk and Robotron arcade games.

Head on over to the entry page to play and feel free to comment.

It’s time for another LD! While I wanted to use LÖVE for LD25, I’ve been elbows-deep into GameMaker for a few weeks, so I will be sticking with that.

Tools:

• Engine/IDE: GameMaker:Studio Pro+HTML5
• Graphics: The Gimp, InkScape
• Sound: BFXR
• Music: Sunvox
• Timelapse: Chronolapse

Coding Playlist includes: Clint Mansell, General Fuzz, Jack White (Raconteurs, The White Stripes, The Dead Weather, Third Man Records Library), The Black Keys, Asylum Street Spankers, Chet Atkins & Les Paul, Diablo Swing Orchestra, Glen Hansard, Miles Davis, Dave Brubeck, The National, Iggy Pop, as well as some game soundtracks (including: Bastion, Borderlands, Portal 1 & 2).

Just a quick update. I started about 8 hours ago. I’ve got the level building OK with a few objects still yet to be created.

Here’s a quick screencap of the level so far:

The game field is in place and the click menu is starting to take shape and function.

Next step is to work out the inheritence components of the dynamic menu system and then start using it to create entities.

Well it wasn’t exactly the MOST efficient day; but I did make quite a bit of progress…

What concerns me about tomorrow is actually having a game running. As of right now I’m still working on play mechanics. I know I should have a working game on day one; but we’ll have to see how tomorrow goes.

What works:

Level loading, mob spawners, minion spawners, movement.

What’s needed:

Combat, resource system, level win/lose/transition, sounds.

Nice to haves:

Animation, multiple mob and minion types, title/help/credit/splash screens, and music.

Let’s hope tomorrow is focused and productive….

I feel like this time I got much closer to a really fun game. While the game that I made: Dungeon Overlord: Master of the Queue is missing some important balance components, Most of the “must have” features made it into the game.

Submission Page

Tools

• Engine – Game Maker: Studio Professional + HTML5
• Graphics – The GIMP
• Sound – BFXR
• Timelapse – Chronolapse

This was my second LD that I used Game Maker on. While I’m beginning to get comfortable using the engine, there was still a lot of research needed during the whole process to get features working. I’d say I spend 3-5 hours during the compo just fighting with my lack of knowledge with the engine.

Graphics:

I had no real plan for graphics during the development, they just basically evolved from one item to the next.  I wish I had spent the first few hours sketching out artwork and a color palette. That would of course lead to using Inkscape to develop the art and then export to bitmap for final finish in The GIMP.

I also wish I had made time for doing animations as well. My first bit of animation was just manipulation of the x and y offset to reflect stricks and hits by the players; and that was way things were going to be until after 3-4 hours of developing my animation base on those offsets I discovered that the HTML5 runner for GameMaker didn’t like how I used local vars for this. As I didn’t want to try and port all of that into something that worked, I ditched the mechanic and side-tabled animation to be picked up later (frame animation coupled with particles + sound… which didn’t actual happen due to time constraints).

Sound:

I really only had time do the first stab at game sounds; had I made more time for it, I would have had layered sounds so that they didn’t sound so weak.

Music:

My plans were, if I had time, to bang out a tune in Sunvox; that generally takes 1-2 hours, so I had to sacrifice that time to get the game to  a playable state.

Gameplay:

All through development I wanted to get the game to where it was at submission time, then take time out to develop additional content. After that I would have balanced out the gameplay to something enjoyable. I ran out of time there.

What worked:

My toolchain

What didn’t work:

My time management skills

Conclusion:

As has been said time and time again, get a working game in your first 6-8 hours. If it takes longer than that, you will be late. The problem was that I let the bithead in my try to use engine mechanics that I had never used before; so that resulted in inflated development time.

Credits:

• Idea – Squidi’s Tiny Crawl 3.0 – Dungeon Defense
• Timelapse Soundtrack – Isoteric (with permission).

Timelapse:
Youtube

It’s been about 2 years since I’ve given LD a serious go. So time to clear out the cobwebs and give it a shot.