Thanks for another great Ludum Dare!  Here is the results of my game, Bat Cave.  I previously posted a Post Mortem of the game, and when I get some time I may post a comparison of my 3 LD scores, but these results are a great improvement from my last LD.

The Project:
I was the programmer for a 3 man team (programmer, artist, soundbro) that made the entry Dino X for Ludum Dare 36. I used C++ with SDL to write the code, and started from only a small math library (clamp, v3 functions, etc.).
Overall:
Overall I think our project was pretty successful. We managed to complete something that at least resembles a game. Unfortunately, some basic elements of the game never made it in, such as scoring, due to fatigue.
What went well:
  1. Physical Location.
I really wanted to get people together in one location that wasn’t anyone’s house so there weren’t any distractions. This worked very well and we were all productive the entire time we were there. I think a large part of our success, at the very least my success, was due to simply not working from home this time.
  2. Game Idea
This went by very quickly and we adjusted it a bit once we got started before locking it in. Going into this we wanted to do some kind of platformer, but we brainstormed once the theme was announced and came to a few possibilities, waited a bit, and then all agreed on the final game style that we went with. Once programming and art started, we decided to change it a little bit after a few experiments.
  3. Experience
This was an excellent experience. I learned a lot, as did my teammates. The time pressure is a really nice thing for getting stuff done and learning how to do things. I think that not spending a lot of time to try and figure something out and instead just doing it actually helps a lot for the first time you’re doing something. You can learn how to do it properly later, but the experience you get from at least doing it at all helps inform the learning later.
What went not so well:
  1. Preperation
I didn’t properly prepare ahead of time, so i was starting from approximately zero code, and I’m not quite fast enough to work from that far back.
  2. Bottlenecks
Our workflow wasn’t set up very efficiently, so a few times we stalled as people were waiting on me to add features so they could see if adjustments needed to be made, and we didn’t have stuff planned out with enough detail that people knew what to do while their current work was stalling.
  3. Timing
We had some issues getting to the location and back, mainly to do with stuff going on in people’s lives at the time. This cut into our start time a bit, but it ended up not affecting a lot because of the next point.
  4. Endurance
We started on late Friday/early Saturday, and by mid-Sunday we were mostly out of steam. Most of us had really done something like this before, save for my very much failed attempt at LD33. I think the main issue was that we were just mostly starting out from bad places in terms of how rested and focused we were.
  5. Requirements
This was probably the biggest issue. We didn’t have clear standards for how we were going to do assets/gameplay stuff from the start, which ended up causing some wasted work and loss of motivation when some stuff had to be thrown out.
  6. Completion
Due to fatigue at the end, some basic things that woudln’t have taken long to implement and would have gone a long way to making the game better weren’t put into the game.
  7. Gameplay
The game has issues in the gameplay department. The scrollspeed is somehow too fast and too slow at the same time, the game is too difficult but also easily breakable in terms of difficulty, there is no way to win the game and no score tracking, etc. It has the core gameplay and literally nothing else, so it was half successful, but not fully.
Takeaways
Having a dedicated physical location where everyone in the same room that isn’t anyone’s house was a huge boon, and I will be doing this if I participate in the future.
You have to have clear and well-defined standards from the start, especially for art assets. Things like perspective and size shouldn’t be something that anyone has asks about.
Make it so that people always have something to do. We couldn’t afford to have people stalling with no work because of the time constraints, but we did. The original plan was to have the game easily playable and tweakable by the other people when they weren’t working on something, and I had most of that implemented, but not the final parts to make it work.
Having the game be in a playable, mostly-shippable state since about midway through the competition should be a requirement. Putting easy stuff off until later to come back to is simply risking not having those things in the game. We tried for this and I failed to deliver, which affected the final game a lot.
Conclusion
The event was a success. Our main goal was to get something running and running decently, so gameplay itself was a secondary concern this time. The game was only a semi-success, since the gameplay didn’t quite come together and it feels like an incomplete game, which it very much is. The experience and feedback that we have gotten will help us a lot on our next LD, when we will now have the skills and speed to try and make a game with good gameplay that feels complete.