Art Post Mortem for “My Little Dungeon”
Simon here, with my late post mortem for the LD23 Jam. I was responsible for the graphics for our Ludum Dare Jam game called “My Little Dungeon”. I recommend you play it if you haven’t already… and rate it. 😉 Thanks!
So this is an overview of the 3 days and what we did from my own perspective. This means it is my perspective and Stefan and Gavin may have different views on things. 🙂
There were two first time things for me with this game:
- It was my first Ludum Dare game for me.
- It was the first time I did pixel art animations at all.
The tools I used:
- GraphicsGale (All animations and most of the pixel art)
- Photoshop (Some art/logo and some reworks and cutting/layer-work)
- PyxelEdit (I used it for previewing the tilesets)
So you want to know what went wrong and what went right? Well then, lets start after the break:
What went right:
- It was a last minute decision for us to enter the Jam. Stefan asked me back in February the first time, if I wanted to join the Jam with him together. But up until the day before the start of LD 23 I was unsure if I have the energy to go through a 3 day crunch time, even if it is self-inflicted. This was because I just had some rough times behind me in my professional life and just started to recover. But in the end I joined and it was by far the best decision I made in a long time, and it was a really funny, creative and productive time. Even though I had to take a break a little bit more often and sleep a little bit longer then Stefan. 🙂 Additionally, Gavin joined us and helped out with all music and sound-fx tasks, which was a really big help. This was great especially as we asked him on really short notice. Thanks alot :). Anyway we finished and we are proud that we did finish. This was a big positive.
- The tool choice was the right one. Even though I had not used all the tools before under production stress it worked out well and they helped me more then they kept me from finishing anything. It was a really smooth transition between the three tools and was a great combination. Especially PyxelEdit and GraphicsGale in combination was great for getting stuff done early before I could see them ingame.
- Art style: I had a vision of what I wanted to achieve right from the beginning which really helped alot. My biggest fear in the beginning was that I would have a creative block, but hey guess what: nope didn’t happen. 😉
- The simple art style also helped me to get my second biggest fear out of the way: Pixel art animation. I had never done that before. Normally I do 3D work and pixel art is a new field for me. I only just started to explore pixel art with our last game “Pixel Towers” but that game had no animations at all. Also GraphicsGale really helps you to get your animation job done. Love it! In the end I was happy how the character looked and how they moved. 🙂
- The communication between the three of us worked really well even though we were sitting in 3 different countries and only had Skype for communication. But we are used to work that way anyway since we’ve always had the problem of different locations, from the day we started working on games together.
What went wrong:
- I was unhappy with some colour schemes I used in the graphics. I also lost tons of time fiddling with the palette and trying to get the colours right. Next time I will plan a little bit better ahead and create the palette before I start making the art itself.
- Even if the game itself worked as we thought it would, there were definitely some big holes in the balancing and the things the player can do.
- For example the game is way too slow. But as we only had one hour left at the end for balancing and tweaking we couldn’t solve that anymore. (or maybe our brains were just too slow after three days of working that we couldn’t imagine the game in a faster mode ;))
- There were not enough things for the player to do, making the lack of speed even worse.
- There were some bugs and problems we just couldn’t solve anymore in time, like that you can get stuck or that the AI sometimes just loses interest in doing anything and you die from starvation. And we thought it was not that bad, but as it seems it is happening far more often than we thought. Sorry! 🙂
- In the end, I had lost myself way too long in some art work. For example I took 4 hours and three tries to get the endboss room done. That was way too much time for a rather small asset. Also the logo kept me busy for another 2 hours which was also too much in my opinion.
These are some iterations it took until I had the final endboss room. Took me too long in the end…
- We failed to make it clear how we used the LD 23 theme. The dialogue and the endboss-room were just not enough and you really had to play through the game to really get it which takes way too long and can be an impossible task due to bugs. So next time we will spend a little bit more time in thinking about how we can integrate the theme in a nicer and more fitting way.
- The menu and the controls are one of the things we rather failed at. First, we started way too late (last day afternoon) with the user interface. So we could not integrate a real interface but more a statistics window. This was because we originally thought, that we wouldn’t have time to get to the interface done at all. In the end it would have helped the game if we would have invested in a more working UI. One thing we also failed to implement was a simple main menu with the possibility to restart the game. But it was just too late already when we thought about it. 🙂
Overall it was big success for us as a team and for me personally and we are proud that we made it to the end. The best bonus of the whole thing: We now have a prototype for one of our next games. Because we are going to extend this as well as move it to other platforms in the future.
Can’t wait to get going again with the next Ludum Dare, and thanks for reading this rather long post mortem. 🙂