Duke Dashington – Post-Mortem

Posted by (twitter: @@AdventIslands)
August 28th, 2013 12:02 pm

LD Banner Duke Dashington
Play the game here!

Ludum Dare compo is over and now that I’ve rated near 1oo entries, I’m going to start working on post-compo version of my entry Duke Dashington! But first, let me tell you a bit of the development progress.

Duke Dashington has been my 5th Ludum Dare entry and I started it largely unprepared. I had a hunch 10 Seconds would win, but I had not come up with any game ideas for it. It didn’t help that I was increasingly nervous if I could top my previous entries Persist and Mad Princess, which had both been huge hits. Most of the first LD day went being in panic about what I could make a game of. At first I wanted to make a top down style Zelda game, where you need to save the world in 10 seconds, and you could do various things that halt the time passing temporarily. I scrapped the idea because I felt it was too similar to PSP and PC game Half Minute Hero, and it wasn’t all that original, so my motivation wasn’t very high.

I have made many platformers in my game dev career, and I wanted to avoid making yet another one like a plague, and I wanted to use the 10 seconds in some clever way instead of of just time limit. But what do you know, that plan didn’t work so well either, I had already spend near 6-8 hours pondering an idea, I thought “screw it”, and started to design the character.

However, while my games might be bit samey, I always try to do something new with them all. While drawing the first character design, I remembered how few days ago my fellow irl game designer friend complained that while he found my previous LD entry Persist very good, it was too slow paced for his tastes. Then I got an idea: what if I made a game where you had to be constantly on the move instead of careful hop and bobbing? What could be a good way to keep the player character constantly moving? Then it hit me: Dashing! I got super excited and started to design the character around that ability. The player character could start dashing to left, right and up, but he could not alter the direction while moving, so player would need to navigate mazes by using the dashing cleverly and bouncing from walls.

Duke Dashington, the first and final design comparison

Duke Dashington, the first and final design comparison

I started to think how I could use the 10 second theme in my game, and came up an idea of bumbling gentleman adventurer, who accidentally causes the temple he is going to explore to collapse before he has even entered it! I originally didn’t want to use the 10 second theme as a time limit, but it was simply too fitting to the character to just pass. His brawn is inspired by one of my favorite video game series Wario Land and his clothing was inspired by Disney’s Tarzan antagonist Clayton. His British look and mannerism is based on one of my favorite Nintendo DS games Henry Hatsworth. Unfortunately his British gentlemanlyness doesn’t really show in the game, because I had to rush in the what little dialogue there is to get the game finished on time.

As seen on the first sprite, I usually make my sprites have completely black lines, but this time I wanted to try something new, and tried to keep the outlines dark, not black, and any outlines inside the character sprite would be light colored. The end result looks super nice in my opinion, very smooth. I’m definitely keeping this style when pixeling characters from now on.

After I had made some music and finished the character’s final design, my good friend BFlorry, who has been suggesting me to stream my game developing for couple of Ludum Dares now, asked me once again to stream, offering to help me set it up. At first I was reluctant, because I’m not very social person: I’m shy, I don’t like huge crowds of people and I don’t like being center of peoples attention. But after he insisted I gave in and thought: why not? I have never streamed anything before, might as well as give it a go, it could be interesting.


After setting the stream up, I was super nervous at first, my friend shared the link to the stream and more and more people started to follow it. All these people watching me draw and fight with bugs when programming! At first I was going to close the stream, saying I can’t handle this, because in real life, I can’t focus on drawing if I know someone is watching me, and at first the stream was really hampering my progress. I fought against my nervousness and kept going and eventually it was starting to feel kind of nice.

People were really excited of what I was doing, praised the pixel art, gave me good feedback and suggestions, were really encouraging telling me they couldn’t wait to play the finished thing. I had nice time chatting with bunch of cool people and and their enthusiasm was really inspiring, motivating and a huge confidence boost. I’m used to working alone, but chatting with all these people while making a game was really fun and nice change. I’m definitely going to stream again on other Ludum Dares now on!

Dink Dashington 6

Because I was making a whole new type of gameplay I had never done before, and programmed a highly complex dashing based control system, I was of course prone to run into some huge obstacles. The first one being was very weird collision bug: if you are already colliding against a wall, then try to dash against it, nothing happens! which is normal of course, but right afterwards if you dash to the other direction, Duke would slide on the ground with this regular idle animation. I spend way too many hours fighting this  bug, only to figure out a simple fix that took maybe a second to do: change his Idle and Dash stances collision box sizes. I felt like an complete idiot for not figuring it out earlier. I blame lack of sleep.

I was using the very latest version of game engine Stencyl, which just had rolled huge update which changed many of it’s internal workings to improve game performance. Old stuff like animated tiles stopped working because of it, and I am not sure if it was caused by the update, but game scrolling wasn’t working either. I wanted the game to originally have one screen and two screen long big rooms, but making the camera follow the player caused huge and ugly distortion in the graphics and looked hideous. I ended up having to scrap the scrolling levels, but I think it worked in the games favor: I tested a new way to switch game scenes and included old Zelda style room switch, which really adds to the feel of the game! Also designing smaller rooms was highly entertaining and easier to test.

The last problem was the number of levels. Every room is designed to be completed in under 10 seconds or the ceiling collapses on you, so I had to design lots of different short rooms, which wouldn’t had been a problem, but I had wasted too much time trying to come up with this game idea and fighting with the collision bug, that I didn’t have enough time to design more then 11 levels, and I didn’t want to add too much easy filler rooms. 4 hours before the deadline I decided that I would rather want a short but finished game, than long but unfinished one, so I started to work on the game’s ending, and wrapped the game up 2 hours before the deadline.

One of the not so serious problems was the game’s name. Originally his name was called Dirk Dashington, but then I found out that Dirk Dashing was already a name of some other developers super agent character. I then changed the name to Dink, after one of my childhood favorite pc games Dink Smallwood. It’s only after I had gotten good long sleep when I realized I could have named him Duke, which fits his character and sounds badass. I changed the name of the game and his name from game descriptions, but felt little bad that most people, especially the ones who had followed the stream, who had known him as Dink and were suddenly offered Duke.

Some of the game's tiles, drawn in 16x16 size.

Some of the game’s tiles, drawn in 16×16 size.

Overall, I think this was a very successful Ludum Dare, I had created perhaps my most polished game jam game yet and had got over of my horrible stage wright. I streamed for the first time in my life and it was very positive experience, over 1000 unique visitors and at the best times near 25-40 constant watchers. I’ll be working on Duke Dashington post-compo soon, and I will be streaming it again, this time aiming to create somewhere around 20-30 rooms total, depending how many ideas I can come up with. I want to give special thanks for BFlorry, my stream watchers, and anyone who was encouraging me and looking forward to the game! I could go as far as say this would not have been possible without you.

Many people have pointed it out and I noticed it too while looking at the various screenshots of the game on my phone: This game could work pretty well on mobile phones! After I have finished the post-compo version, I might look into mobile version of the game. More different ruins for Duke to explore (and accidentally destroy)? New levels, tiles traps? I need to do some control tests and experiment…

Leave a Reply

You must be logged in to post a comment.

[cache: storing page]