LD22: Personal post-mortem

Posted by (twitter: @EKTOutie)
December 20th, 2011 7:00 am

With LD22 – my first Ludum Dare – over and me having caught up on sleep, I’ve had some time to reflect on how it’s gone. Personally I consider this weekend both a success and a failure in different ways.

It’s a success in that I have proven to myself that I can create something at least playable within 48 hours – the reason I signed up in the first place. I’ve also managed to learn a new toolset.

It’s a failure in that the design is pretty crude and the end result isn’t nearly as “finished” as I would have liked it to be. This is due to various issues, some of which popped up very late in the compo.

Nonetheless I’ve submitted the last build, for the sake of honesty. This is what I could manage in the time given with the self-imposed restriction of using new tools rather than familiar ones. I’ll get better. I’m still not sure if it’s in a state I should have submitted it in, but it’s a bit late to change my mind now.


What went right

Not Alone

I spent the weekend with a couple of friends who were also doing LD. This gave me a good deal of motivation to not slack (too much) and made the breaks a lot more pleasant. It also meant that when it came to having food, one of us could make something for all three, saving the other two some time. Even when working solo, I’d greatly recommend being in a group.

Distraction management

As an extension of the above, I managed to mostly stick to the project with little distraction, even though I had the internet and some games right in front of me, and even though I am normally The Procrastinator.

Learning a new toolset

I decided earlier in the week that I’d try doing the game in Game Maker, to see what the toolset is like, and the art in Bottleship, to try its suitability for repeated rapid art production. Both of these I now have a reasonable grip on, and both of these are fairly nice to work with within their limitations.


What went wrong

Learning a new toolset

I kept accidentally overwriting existing sprites in Bottleship when I tried to do animations or derivative tiles. This cost me a bit of time redoing assets I’d done before. I found out that the Mac version of Game Maker has some significant pitfalls; not only is it still v7 which is now approaching five years old, I’ve also to this point been unable to get the Publish-to-Web function to work. Additionally I used the free version of GM, which meant I had to make some compromises on tech (no alpha transparency, in particular).

Solution: do more research into the new toolset next time or use one I’m familiar with.

Non-programming skills still a bit crap

The design I came up with is quite crude. If I’d had enough time to implement more enemy behaviours, it might have turned out more interesting. I’m planning to continue working on Headphones On and see if it turns out reasonably fun. Also, my pixel art is a bit rough and due to my nonfamiliarity with making music I was having trouble getting anything at all down on that front.

Solution: practice more.

Lack of inspiration techniques

This is part of the cause for the crude design mentioned above. “Alone” is not exactly an easy theme to work with I thought, but the fact that I have no methodical approach to finding inspiration didn’t help.

Solution: No idea. Any suggestions welcome.

Late-dev issues

I ran into several problems very late in development; we’re talking an hour before submission time. For one thing suddenly a crash bug popped up relating to key events, so I rewrote my entire input system in script rather than using GMs event system. This then resulted in having to rewrite collision detection, which resulted in having to re-test the level. This kept me busy for just more than an hour.

When it came to submission, I found that the Publish-to-Web function on GM/Mac seems to be broken, so I had to run to the website and buy the Pro version just to be able to make any build to submit.

Solution: Not much to be done about late-dev bugs. I was building and playtesting a lot and genuinely don’t know where that crash bug came from. The publishing issue I could have figured out if I’d researched GM4Mac more thoroughly, rather than just reading up on how it works.



In summary, while I’ve pretty much achieved the goals I set for myself for this LD, I’m not really satisfied with the product thereof. The main problems I was having were down to not having researched potential issues with the new toolset enough before the compo; the others are mostly things I expect will get better with practice. Very much looking forward to trying more jams and LDs in the future


Leave a Reply

You must be logged in to post a comment.

[cache: storing page]