main

ContributionsDevelopmentGame DevelopmentIndieOnlinePostmortem

AVION Checkpoint: Recreating a Childhood Game In-Between a Regular Job

November 20, 2014 — by Industry Contributions

feature.jpg

It was 1983; Ivan Mandic was in the second grade, when his father’s friend knocked on the door and came in with something wrapped in a blanket. A small black box with a rainbow, lots of cables, and a cassette player. A few days later, Ivan decided to create his first game. 10 circle(10,10,x), 20 x = x + 2, 30 goto 10 – a sample code from Spectrum BASIC. “There were no books about ZX SPECTRUM in book stores, but Father found one and rewrote it by hand. The whole book, including images,” Ivan recalls as he shares the story of his game, AVION Checkpoint.


avion.exe

In a few years, I bought a computer with my own money (no Commodore 64, it was Amstrad CPC 464). And in 1993, I turned on my first PC (286, 20MHz, 20MB HDD, 640KB RAM, Hercules), and this is very important: on the hard drive, I found a game called avion.exe. For a few days, I had been enjoying flying a small airplane up and down and throwing bombs on enemy bunkers. But while playing around with MS-DOS commands, I deleted command.com. It was a disaster. Darkness. End of humanity. No one knew what to do and whom to ask. After a couple of days in service, everything was fine, except one crucial thing: there was no avion.exe on the C: drive. From that day, I had the game in my mind. I’ve tried to make something similar in Pascal, Modula2, Delphi, C, C++, Java, JavaScript…no luck…

Знімок екрана 2014-11-05 о 03.16.25
Sopwith, David L. Clark, 1984., 29kB, MSDos, long-lasting inspiration for AVION Checkpoint, http://sopwith.org

Because of 3D

In February 2014, for the umpteenth time, I decided to download Unity3D, but this time, I was prepared. I had downloaded some gamedev tutorials from YouTube (about 60GB) and already knew what a game object was, a prefab, how to rotate it, how to get input from keyboard, what Vector3D was, and how to make a Space Invaders clone. I had read anything and everything about Unity3D and game development. I got lots of experience with 3D Studio, so already had a clue about 3D objects, transforms and textures, which was all helpful.

In a few months and countless number of versions, I came up with a game similar to the one you can try on Google Play. Many people asked me: “Why the hell 3D, all good and successful games are 2D, have you tried Angry Birds?”, and every time my answer was: Because I need to make it 3D. It has to be 3D. I see it in 3D. End of story. This was the only time I didn’t want to listen to good and friendly advice.

development-drawings
Some details of level design

Problems: Variety of Devices, Sound, iOS, and Lack of Time

The first problem was testing the game. Thousands of devices, OS versions from 2.3.4 to 4.4, small screens, big screens, screens bigger than my home TV, processors from single to quad core, memory, and skins…I decided to buy test devices: a SONY Xperia ARC as the low level device, SAMSUNG Galaxy TAB as the middle one, and Nexus 7 2013 as a high-end device. My goal was to make a steady 30+ FPS which would run easily on all test devices. But they don’t represent the whole world. On most phones older than four years and with processors slower than 1 – 1.5GHz, the game was unplayable. I didn’t find a way to check processor or memory on devices and to scale resolution/number of objects/textures to achieve at least 30 FPS, so I decided to limit the Android version to 3.0 or higher hoping that newer devices will have the higher OS version. This turned out to be a good decision.

game1
The Android version has been limited to devices with the newest OS version to make sure it works on all.

The second problem was sound. I have worked in various areas of the IT industry, but I never touched sound. My biggest involvement in computer sound was to play Pink Floyd or Metallica in WinAmp. Luckily, my brother Vladimir Mandić is an MC in Czech Republic, so he sent me a sample which is now the main theme in the game. But this is a favor done for a family member. I still need proper sound effects and background music, and I can’t ask my brother to make changes or try something else a few times a day. So I will need to sit down and learn something about melody and harmony.

“My biggest involvement in computer sound was to play Pink Floyd or Metallica in WinAmp”.

The third problem is the iOS version. Exporting from Unity, importing to xCode, compiling – and all I get is hundreds of errors. Maybe an OS X version? Downgrading the iMac from Yosemite Beta to Mavericks, and xCode to 5.0 didn’t produce any visible results. So the iOS version will have to wait for better times. Sorry Apple fans, it’s too complicated right now.

The fourth issue was time,  and it was the BIGGEST problem. I realized that my one-man team wasn’t big enough when my friend Davor Končalovic started testing all versions up to the smallest details and telling me what he thought should be changed, added, or deleted. I’ve spent hours fixing bugs, making changes, and adding improvements. Testing is essential, but I have a regular job from 9 to 5, and kids are waiting at home with their own wishes and demands. So the only free time I have is at night. The first night was easy, the 10th one was sleepy, and the 100th night was hard. Now I’m used to less hours of sleep. Sometimes a short nap in a chair is refreshing enough for some more hours of work. I’ve heard that Nikola Tesla slept 15 minutes several times a day (and the maiden name of Nikola’s mother was Mandic, so you can see the connection).

game2
Ivan runs a small advertising cogency, so the only time he has to work on the game is at night

Another obstacle is the need to cover so many different areas in order to make a good game: programming, design, modelling, rendering, optimization, sound, usability, gameplay… And I’m losing time because of the constant need to re-learn and remind myself how to do something – for example, add a new object or select all linked vertices in Blender. Of course, Google knows. But it’s 10 minutes lost! 10 minutes here, 10 minutes there… It’s too demanding to work on multiple projects/jobs at the same time.

blender-level-design
Re-learning and reminding himself how to do things, for example, in Blender, is the reason why much time is wasted, Ivan admits.

The fifth problem was game design decisions. Should I focus on gameplay, or are graphics more important? What is the path to a successful game – nice and shiny or addictive? I know that the best games have this all, but how can I achieve this? I decided to focus on the gameplay – making it addictive and creating more levels, leaving the menu designs and options for the better times.

Marketing: Ask What is Wrong and Get Feedback

In the beginning, I thought, I will make a game, post it on Google Play, and that’s it. What a mistake! Making the game is a small part of development. Having posted the first public version, I decided to announce it locally. I sent nearly 400 emails to my friends and people from my business contact list with brief information about the game and a download link, and asked them to help me by testing and posting impressions. What I received were 4-5 replies and maybe 10 more downloads. WTF? OK, I thought, maybe they don’t have time for testing, or don’t know how to tell me that the game is not good. So I posted the link on lots of Android game forums – and nothing again. I started thinking that other devs didn’t want to encourage my development because of competition. In 10-12 days, I decided to openly ask: “Hey guys, what is the problem. Am I a black sheep? Do you hate me for some reason? Do I write in a language you don’t understand? Please say something, even that my game is the worst one ever and I need to format my HDD and go to work on a corn field. Anything, but don’t ignore me, please!”

And BOOM! Hundreds of replies in the next few days. People said it’s a great or excellent game. I got nice ideas, people tried AVION Checkpoint on so many devices, providing detailed reports, lots of them offered help. What a great feeling it was to receive nice feedback!

acrylic_planes_1
AVION Checkpoint acrylic figures

I’m now working mostly on optimization and new levels. I’ve made 12 new levels and 3 more airplane models, which will have their own flying properties. The already available levels will soon be remade with better textures and baked shadows. I made one level endless, for the most passionate players. I’m also making a better online score system: all scores will most likely be nulled every week so new players will have the opportunity to see their score in the top 10, even if for a few minutes. I didn’t forget about the iOS version, but right now, I don’t have enough time for it.

facebook-cover
Already available for Android, iOS version needs to wait for the better times

I can’t wait for the day when I will be able to tell myself: it’s time to leave everything behind and start fulfilling the dream of your whole life – BECOME A GAME DEVELOPER.  I have energy, great support from family, and believe that day will come for sure, because I’m feeling butterflies every time I start Unity or Blender.

Ivan believes AVION Checkpoint has great potential, and promises to do his best to make it even better (and nicer), though he admits he will need help in marketing and promoting the game. It can already be downloaded in Google Play and from some other websites with Android games, and played online on Wooglie and Kongregate. Ivan adds that he hasn’t thought of monetization yet, since his day job takes too much time and strength, but plans to change that after meeting some publishers in the nearest future.

 

ContributionsDevelopmentGame DevelopmentIndieOnlinePostmortem

Own Kingdom: A Game Remake that Built the Team

September 22, 2014 — by Mariia Lototska

feature7.jpg

In July 2011, Eldwin Viriya took a leave of his job as a lecturer of basic algorithm and data structure for a semester to take a GRE test for the master’s degree. Having passed it successfully, Eldwin discovered he had a lot of free time. He decided to use this time for self-development and made DragManArds in 1.5 months. This Flash game really sparks the light of game development spirit in its author. Later, his company, Own Games, created DragManArds’ remake Own Kingdom, a fantasy medieval strategy game where you need to protect the kingdom from waves of monsters. He describes it as an experience of tower defense games with a taste of war games.


Get the Taste of Making Games

When I first created DragManArds, I used MochiAds for monetization, since that was the only monetization option that I knew at that time. I didn’t even know about Flash sponsorship back then! The result turned out interesting: I got a lot of feedback from real players in Kongregate, some fan messages and suggestions, and also managed to earn more than 200 USD in the first month (which was cut down to only a quarter in the following month, and to almost nothing for the rest of the month).

It felt amazing to actually experience the thrill of launching a game, but the best part was when DragManArds dragged me into the gaming ecosystem of Indonesia. Groups such as Gamedevid allowed me to get to know game developers of the country, as well as big companies like Blackberry and Nokia.

Own Games Team - left to right - Raynaldo - Jefvin - Eldwin - Okky - Agustian
The current Own Games team: Raynaldo, Jefvin, Eldwin, Okky and Agustian

Remake DragManArds: More Features, Better Graphics

In late 2011, Nokia held a game developer competition for their feature phone platform. I asked Jefvin Viriya, my brother (who was still in high school) to help me make the game in time. Having submitted a mini game named Beyond the Well, we came out as the third winner in the competition, and since then, we continue developing games together under the name of Own Games.

We started attending local gamedev events here in Indonesia, one of which was Game Developer Gathering. After this gathering, Kris Antoni from Toge Productions invited me to a meeting with Mochi Media. I got a chance to show DragManArds to their representative and received good feedback about the game. He said he was interested in being contacted again if there’s any sequel to the DragManArds. This meeting made me believe that my game has a lot of potential within.

The meeting with Mochi Media made me believe my game has a lot of potential.

At that time, working on a new Flash game would have been really hard for us. Firstly, Own Games already had a good amount of players from Nokia Store, and we want to keep them happy with our creations. Moreover, I was also busy with my day job as a lecturer, and my brother got overwhelmed with his high school final exams (not to mention that he didn’t understand ActionScript at all). So we continued our life as usual after that time.

A few months later, Nokia launched Lumia, a Windows Phone smartphone. Until this day, Own Games was focusing on feature phones only. We were working in native J2ME and were not really familiar with modern game engines. Then I noticed that one of my juniors had graduated from the bachelor program, and I invited him to work together in Own Games. The first thing he did in the company was port DragManArds to Lumia. The results turned out great: DragManArds  got a gold medal in the Lumia Apps Olympiad in December 2012.

DragManArds Gold Medal from Lumia Apps Olympiad
DragManArds’ gold medal in Lumia Apps Olympiad in December 2012.

Then I finally decided to quit my job to completely focus on Own Games. On April 1, 2013, Own Games transformed into an official company. Agustian, a 2D artist, also started to help us out. It was really a big move for us: before he joined, we were short on manpower and, what is more, he had a degree in arts and experience in making games. The first objective became clear: remake DragManArds with more features and better graphics.

Learning From Mistakes and Feedback

DragManArds already has a lot of versions: Flash, Windows Phone, Blackberry 10, and even J2ME. Having received a LOT of feedback, we planned a lot of stuff that we wanted to implement in the remake. It turned out to be a lot of tasks. But as Agustian is a talented artist with experience in game industry, I could fully dedicate myself to improving the gameplay and user experience, and our programmer had proven himself successful in making the Windows Phone and Blackberry version of DragManArds, we believed we’ll make it.

Own Kingdom Banner
The DragManArds remake, Own Kingdom, needed much more effort than expected.

I was too optimistic back then, and set the deadline to 3-4 months (DragManArds was made in 1.5 month by me alone, right?). But we weren’t able to finish everything in that time. As any new startup, we faced many challenges, both technical and not. I often argued with Agustian about how he used a lot of time to draw some tiny details that cannot even be clearly seen in the final game. Meanwhile, our programmer had to work remotely from another city because his father had a serious illness. In the end, he realized that he didn’t have enough time to develop anything and left Own Games. So we lost our programmer, our art assets production took more time than planned, and my entrepreneur’s soul was still on a very early development stage. I used to get a salary each month, now I had to pay salaries each month – It feels totally hard in the beginning even though you are already aware of the risk.

I used to get a salary each month, now I had to pay salaries each month. Feels hard in the beginning.

A few months after our programmer left the team, we met Ray Naldo, a former junior in the university where I worked. But we didn’t want to give him the pressure of developing a game as big as Own Kingdom for his first time. So we decide to make Eyes on Dragon, a 3D endless runner. During its development, we also got some help on 2D art assets from Okky, Agustian’s junior.

Eyes on Dragon Logo
Eyes On Dragon: project for the new programmer to adapt.

Meanwhile, Jefvin was learning C++ and tried to make Own Kingdom for Windows Phone 8 using Cocos 2dx. The WP8 version eventually became the finalist of the Indonesia Game Show. During our presentation at the competition, the judges called Own Kingdom’s gameplay a unique and promising one, but pointed out that the program was crashing and the buttons weren’t working smoothly. Even though we didn’t win the competition, this encouraged us to go on with Own Kingdom. But, sadly, once again, we had to put development for WP8 on hiatus when we realized that Cocos 2dx for WP8 didn’t support mp3 files.

Back to an Abandoned Game

A few more months had passed. Eyes on Dragon was published. We were happy with what we made, and decided to go on with the development of Own Kingdom. Ray started learning Unity 4.3 for 2D, Agustian and Okky made more art assets for the game, and Jefvin and I kept improving the game design, level design, and also the whole gaming experience.

The second development phase was not easy, but definitely better than the first one. Continuing the game that was once abandoned is for sure not an easy task, since most of the courage is gone. What is more, there were two desires we struggled with: to make the game better but, at the same time, finish it as fast as we could. Yeah, that’s shameful. Nevertheless, coming back to Own Kingdom had positive sides, too: we already knew that the game is worthy and that a lot of people wanted to see it completed. What is more, now we had a bigger team and some experience. Eventually, we managed to finish Own Kingdom in April 2014.

own kingdom gameplay 1
In April 2014 Own Kindgom was ready.

The development of Own Kingdom is a long journey, and we realize that it has not ended yet. But we are really happy with the growth of each of us. Agustian has started to become more efficient and effective at allocating his energy to finish the work in time. Ray got a lot of experience in making the game using Unity in both 2D and 3D, which opened the possibilities to reach more platforms. I became more familiar with project management, and got a whole new experience in leadership. But the most valuable thing that makes me really grateful is how Own Kingdom turned Own Games into a more solid and powerful team.

own kingdom gameplay 2
Own Kingdom turned Own Games into a more solid and powerful team.

Own Kingdom is available in Windows Phone Store and Nokia Store (Nokia X only), and has recently been launched on Android.

ContributionsDevelopmentGame DevelopmentIndieOnlinePostmortem

Company of Tanks: Bringing an Arcade Multiplayer Tank Experience to Mobile

February 26, 2014 — by Mariia Lototska

CoT_feature.jpg

Based in Kajaani, Northern Finland, Critical Force Entertainment is the town’s first independent game company. Tim Spaninks was brought in as producer and lead designer to direct a young team in the development of a cross-platform game: Company of Tanks. In this article, Tim shares his experience of working with a fairly inexperienced team resulting in a very successful outcome.

Без названия1

Initial goal: a high-quality tank game for the mobile platform

When I was brought into this project, World of Tanks (which is a massive online game developed by Wargaming.net) had fairly recently become insanely popular and managed to open up a completely new market segment for a new sub-genre called tank games.

Anything put on the Google Play Store featuring the word ‘tanks’ seemed to gather up to hundreds of thousands, or even millions, of downloads, seemingly regardless of the quality of the game. Therefore, the initial goal of Company of Tanks was to “create something like World of Tanks for mobile devices” with a quality that would outmatch the existing competition.

From left to right: Ville, Lassi, Sampsa, Mikko
From left to right: Ville, Lassi, Sampsa, Mikko

Sampsa, Mikko, and Lassi came to Critical Force Entertainment as interns, and managed to get a simple prototype up and running very quickly.
Right after I joined the team, we brought in our artists Ville and Thanabodi a.k.a. Viola from Thailand, and I knew we needed a change of direction.

Every Game Should Have its Own Identity

Right from the start, I didn’t agree with the mentality or the spirit of the project. Firstly, I believe that every game should have its own identity and bring a new experience to the player. It wouldn’t feel right to try to duplicate a game’s experience, even if it’s on another platform. Secondly, we were mainly developing for the mobile platform, which has a completely different target group and lends itself to different gaming experiences than PC and consoles.

Без названия3
At first we were going for a more slow-paced, realistic style game

At first, we were going for a more slow-paced, realistic style game, but then decided to adapt the gameplay to the platform: the game would become faster and way more arcade-like to have shorter game sessions with more action. We also decided to change the visual style accordingly. The game was to become stylized to set the right expectations: it’s not a tank simulator and sure as hell isn’t a World of Tanks clone. This way, we would still appeal to a very large market segment aching to play 3D tank games, but at the same time differentiate ourselves from the competition in terms of style and gameplay.

It was now time to prototype, test, reflect, prototype, test, reflect, and so on to find the right way to make the game as fun as possible!

Без названия4
The game was to become stylized to set the right expectations: it’s not a tank simulator and sure as hell isn’t a World of Tanks clone.

Saying No

Often in a game’s development, it’s the designer who says “Yes!” and the producer is who says “No!” to gameplay and feature suggestions. One of my toughest personal challenges in this project was to have to take both of these roles at the same time. Many awesome-sounding or even almost crucial features such as an in-game chat, friend lists, clan support, ranking lists, player stats, or simply the ability to completely customize your tank by drawing on it, placing emblems, etc. had to be put on hold or scrapped completely in favor of finishing the game on time. I was only going to be in Finland until mid-December, and the entire team would end up only working part-time on the game shortly after: we had to release a playable Android version before that time.

Prioritizing was essential, and through continuous debate and feedback, we were able to pinpoint what needed to be done to get everything ready on time. This often meant going for the absolute minimum viable options. No fancy customization and putting together your tank of parts collected throughout the game, but simply a very basic upgrade system.

Sometimes, it’s demotivating not to be able to make the game as awesome as you dreamt, but knowing that we could keep adding features and content after release and strive to make the game as perfect as possible is something that made it bearable.

Remember Who the Game is For

Без названия5
Showcasing our game at the Northern Game Summit in Kajaani and DigiExpo in Helsinki was an extremely enriching experience

When we had our first playable version ready, we got some great opportunities to receive crucial feedback to pinpoint what aspects of the game we needed to work on. Showcasing our game at the Northern Game Summit in Kajaani and DigiExpo in Helsinki turned out an extremely enriching experience. I’ve learned a lot participating in the Northern Game Summit conference’s pitching competition. And winning the €5000 prize for the development of our game allowed us to speed up, acquire some needed licenses, additional testing devices, and invest in a custom-made soundtrack and sound effects.

The most beautiful moment of the entire project for me was at the DigiExpo 2013 event in Helsinki. A young kid picked up our tablet and immediately understood how to play, and got completely immersed in the game. At some point, he glanced over at his friend next to him with a grin and said “hyvää peli!” (meaning “good game!” in Finnish). This nearly broke me. This kid stayed at our booth playing the game for nearly an hour! When getting lost in the development process and reaching your deadlines, it’s easy to forget what you’re actually doing it all for. This was the moment when it became tangible for me that after all of our hard work, we were actually making this for someone. And that someone really loved our game! This is why I love my work, and those tiny moments make it all worthwhile.

Без названия6
This is why I love my work, and those tiny moments make it all worthwhile.

The rest of DigiExpo was sort of a blur of wonderful moments with many people (mainly kids) playing the game. Besides providing us with a lot of feedback to pinpoint what aspects of the game needed work, it was a massive motivational boost for the rest of the project.

Player Base in the Beginning: No One to Play With

There was one thing throughout development that I was dreading the most: how are we going to get players? It’s known that it can be hard for smaller online indie games to gather enough people because nobody wants to play a game that doesn’t already have an established player base – which complicates things even more.

Before our Android build was ready, we decided to Beta test and soft launch our game on the web platform using Kongregate and Facebook. This would allow us to build interest and gather some players without any marketing budget and get some valuable feedback at the same time.

Без названия7
There was one thing throughout development that I was dreading the most: how are we going to get players?

Testing the game on Kongregate revealed one massive problem: when a player wants to start an online match, he is thrown into a lobby to wait for enough players to start the game. Because we started out with a non-existent player base, as soon as someone tried to start a game, they found that nobody (or not enough players) was in the lobby, and simply disconnected. This led to a situation where there were continuously one or two players online who didn’t have enough people to play with.

Arguably, the biggest mistake made in the development process was the way we dealt with this. We figured that if we decrease the minimum player requirement to two and interest in the game would pick up later, the problem would vanish. Surely, when the Android downloads would start streaming in, the problem would fix itself? Well, it didn’t.

Shortly after the Android launch, we noticed the problem still existed, and released a patch changing it to a drop-in, drop-out kind of system that would throw the player immediately into an already running game. Thankfully, this worked and we now have an active player base!

Additional Tweaks

We’ve just reached over 240.000 downloads and with around 7000 downloads every day, the project has been a tremendous success for such a young team so far. Based on the numbers and received feedback, it’s safe to say that many people are playing and enjoying our game, which is a fantastic feeling!

In our eyes, the game is far from finished though. It’s lacking end-game content and goals to strive for. There are many features and content to be added and many in-game tweaks to be made. We’re working hard to implement metric systems to collect tons of data using various analytics plug-ins to determine where our focus needs to be. So far, we’ve basically been working in the dark, and shedding some light on the impact of changes we make will allow us to work more efficiently and improve the game where it counts.

In the meantime, we’ve applied for Microsoft & Nokia’s AppCampus program to be funded with €50.000 to create a Windows Phone version of the game with custom content for the platform. We plan to use those funds to further tweak the game and get ready for a later iOS release!

ContributionsDevelopmentGame DevelopmentIndieOnlinePostmortemPR & Marketing

Pretentious Game: From Flash to App Store Featured

January 6, 2014 — by Mariia Lototska

feature1.jpg

Keybol is a one-man indie game developer company from the Philippines. Bari Silvestre has made a bunch of popular flash games, some of which were showcased at game conferences around the world. Bari is now venturing into mobile, and his first big release happened to be a hit in the form of Pretentious Game.

Contest Entry Becomes Viral

The setting is about a blue block who is in love with the pink block at the other end of the screen
The setting is about a blue block who is in love with the pink block at the other end of the screen

Pretentious Game started out as a small flash game created for Ludum Dare 23 accelerated game development contest. The setting is about a blue block who is in love with the pink block at the other end of the screen. He must reach her in any possible way, even if this means breaking the rules set for the game, or even introducing a new mechanic. The ending is also a bomb waiting for every unwitting player.

The game turned out to be a hit! It went viral and topped the Reddit gaming subtopic. I understood Pretentious Game was a hit not only among gamers, but also within the indie games industry, when it got showcased twice at Casual Connect and was featured at Rock, Paper, Shotgun, and IndieGames. When the game was presented at the Indie Prize Showcase at Casual Connect San Francisco, it won the Director’s Choice award and was nominated for Best in Storytelling. Pretentious Game was also on the front page of big flash game sites such as Kongregate, ArmorGames, and Newgrounds.

Casual Connect
The game was presented at the Indie Prize Showcase at Casual Connect San Francisco and won the Director’s Choice award and was nominated for Best in Storytelling.

The following chapters, Pretentious Game 2 and 3, were received positively as well, since the second one has a surprising ending, along with improved gameplay. The third part is interconnected and awes many players with how witty the whole story is.

I think the game gained popularity thanks to its story, but the gameplay wasn’t lagging behind too. Players liked the fun of solving new puzzles while discovering the whole story. They also found the title amusing, and some thought it was a sort of parody.

screen
I think the game gained popularity thanks to its story, but the gameplay wasn’t lagging behind too.

When Publishers Harbor at Forums

Since the reception for the franchise was overwhelming, I decided to port Pretentious Game to reach a wider audience. To promote the game, I needed a trailer to later post it to the TouchArcade forum. I found a very catchy tune at AudioJungle, and imagined a good trailer script with the positive reviews from Rock, Paper, Shotgun, JayIsGames, IndieGames and Mike Bithell himself – the creator of Thomas Was Alone. Rishikanth Somayaji, a developer I met at Casual Connect Asia, volunteered to make the trailer, and he did an awesome video with his own vision.

The trailer caused a brief discussion, and then Bulkypix, a French publisher, contacted me with an offer on that forum. They reminded me that mobile games need visibility to take off, and promised to help with that. Bulkypix was really giving me benefits, and I couldn’t reject their offer anymore. The most memorable thing in our discussion was when they told me what their Lead Programmer told them: “If we don’t take this one, we won’t take any game!”

After additional quality testing and 10 more languages later, the game was published on December 5, 2013.

The Friday When Bari Woke Up Successful

It was another Friday morning here in the Philippines, when I woke up and decided to check the Best New Games. Pretentious Game was one of them! I thought I was dreaming, I almost immediately thought of hitting the jackpot.

As for the reviews, they were great. I was overwhelmed to see my game in such prestigious sites as Touch Arcade, App Advice and Pocket Gamer. In a week, even more reviews followed. The players liked the game too: more than 1900 reviews averaging 4.6 on Google Play and mostly 4 and 5 stars in App Store.

screen 2
“Concealing a deeper meaning, Pretentious Game is an enjoyable platformer with a touching message.” – 148Apps.

Here are some reviews:

“I really like this game.” 4/5 stars — Touch Arcade

“I instantly fell in love with its challenging charm.”- App Advice

“Concealing a deeper meaning, Pretentious Game is an enjoyable platformer with a touching message.” – 148Apps

“I can say that it’s refreshing to see something like this come along.” – Arcade Sushi


Pretentious Game is one of those rare games on the App Store or Google Play Store full of originality.” – Pocket Gamer

I thought App Store’s feature rotation after a week of success would bring the number of downloads down, but instead we had entered the US App Store top charts!

I believe it’s because of the review from Touch Arcade and a video of 10 Fun Mobile Games by VSauce. It’s amazing how much this contributed to the number of downloads and sales. I can say I’m now more eager to move further through the US market. In fact, we reached 110,000 downloads in 10 days.

Promotion: Festivals, Contests, Press and Direct Suggestions

I’m now pursuing the game even more, since I’ve heard from other successful mobile developers that the first two months are the most critical, because what is being done now determines how long the tail of downloads to come will be.

screen
I’m now pursuing the game even more, since I’ve heard from other successful mobile developers that the first two months are the most critical.

I’ve met with directors from IGN Asia to tell them about the game, and reached out to local media to promote a Filipino-made game in my own country. Pretentious Game has also been suggested for a possible feature in Google Play, and is in process of being included in a mobile Humble Bundle – Bulkypix met with Google as part of their weekly roadmap, possibly pitching the games eligible for feature promotion.

I had also submitted the game to IGF 2014 before it was released, and hope they’ll choose it as one of the finalists.

Kickstarter-Style Monetization: Useful, but Unclear

We’ve tried a different approach to monetizing the game: a “Kickstarter-style” of purchasing the full game for a bigger price in exchange for more rewards like wallpapers and soundtrack. It was also meant for players to support the developer if they wanted to.

It worked, bringing some extra 30 percent to the earnings. But at one point after reading a comment from a player, I understood that it confuses other players. She said she can’t pay $4.99 to unlock the game, since she thinks it’s too much for a mobile game. I had to remind her that it will only cost her $0.99 to unlock the full game. The $4.99 is an option to support the developer and get an additional wallpaper and soundtrack. In the end, because we don’t have real stats to compare with (since, as far as I know, no other game has used this approach), we decided to leave it as is and focus on content instead.

Mistake in a Twitter URL May Have Cost Downloads

It’s a good thing that I manage the Facebook fan page of Pretentious Game, and can connect with the players to advise them what to do. Once there was a slight mistake in the URL of the link shared on Twitter, but that was easily fixed by the next week’s update. I am still wondering if it should have brought in extra downloads. I also noticed the Facebook sharing feature I put in the game did nothing! At least, that was the case when I checked for #pretentiousgame and saw a total of zero users having shared the message on Facebook.

Bari is now working on Circles with multiple personalities! He is planning to start a studio and gather a bunch of local talents in his provenance. If you want to check out more of his games, visit his Facebook Page.

ContributionsPostmortem

Cyberpunks vs. Syndicates: Exploring Another Life

September 6, 2013 — by Mariia Lototska

feature1.jpg
Mososh was created in 2010 by Christian Primozich when his previous company, Challenge Games, was acquired by Zynga. He is passionate about working in a creative environment that encourages everyone to try their best, and has a love of games and comics. He talks about Cyberpunks vs. Syndicates, his third MMO.
Office
Christian Primozich working in his office

Where Did CvS Come From?

Cyberpunks vs Syndicates (or CvS for short) is the third web-based MMO I’ve done after hitting the indie scene as Mososh. CvS has been in development for about two years, as I’ve been running and enhancing my previous title, Chronicles of Herenvale, in addition to doing the development on the new game. Mososh is largely me plugging away – days, nights, weekends – doing design, writing and development, as well as customer support, server administration, accounting and all those other things associated with running a business – gah!

The inspiration behind CvS can be found in old-school titles like Syndicate Wars, tabletop games like Shadowrun, books like Snow Crash, and movies like The Matrix. There’s something alluring about that dystopian image of the future where the hacker dons the role of hero to fight against corporate corruption. I myself am a hacker. I want to be a hero. CvS lets me explore themes that are personally relevant and inject my own perspective into them. That is the fun part.

What Makes A Mososh Game?

Early Concepts Sketches
A few early sketches of CvS

Finding the right artist is a key to the quality of my games. I suppose it’s a reflection of what appeals to me when I choose what game to play. I won’t work with anyone whose art I’m not looking forward to getting in my inbox. At my last company, one of the developers used to say getting art from our lead artist was like Christmas. That’s how the player (or me) should feel when they see new stuff, whether that’s exploring new areas, finding new gear, or fighting new enemies. It’s important I find an artist that really has the same eye for detail that I have, as well as a distinctive style that’s going to set the game’s visuals apart from (and hopefully above) games of a similar play style.

I’ve chosen to work on web-based MMO’s for the last six and a half years, because I like to combine aspects of adventure games with asynchronous PvP and co-op features like guilds. Players will often gravitate towards one aspect, but expanding beyond a single dimension gets more people involved. Multiple aspects also give them somewhere to go if they get bored with one part of your game (as opposed to loading up someone else’s game).

What’s With The Web?

One of the challenges of being an indie developer is distribution, which is why I love the idea of web apps because they are accessible. Developing on the web ensures you can reach the widest possible audience and take advantage of open publishing platforms like Kongregate and Facebook. Without platforms like Kongregate, I couldn’t do what I love to do.

All of these languages are largely just tools. You can build crap with the right tool and something amazing with the wrong tool.

I wrote my first game on a TRS-80 back in the 80’s, but I really dove deep into development during the explosion of the Internet in the 90’s. I’ve written Java, C, Perl, Visual Basic, C# and most recently PHP. I’ve written stand-alone apps, libraries, and web apps. I’ve listened to debates rage about which language is the right language, but I don’t think that’s the point. All of these languages are largely just tools. You can build crap with the right tool and something amazing with the wrong tool. You learn that quality is largely not dependent on the tool when it comes to software.

So what are my tools? L-inux + A-pache + M-ysql + P-hp = LAMP. One of the best things about LAMP development is how free it is – no really, it costs zero dollars. That isn’t to say it doesn’t have its downsides, the biggest being the complexity (or perhaps perceived complexity), as most developers don’t have a lot of experience with web technologies. Running a web server, a database server and gluing everything together can be a bit daunting at first. The good news is services like AWS and Rackspace have made running servers in the cloud relatively inexpensive, easy to set up, and very test friendly in that you can launch and tear down servers quickly. There are also tools like MAMP that really help when getting your development environment set up. Personally, I use HTML and Javascript as my front-end. However, that can limit some elements of the game. Animation is challenging at best and real-time synchronous play can be difficult. You have to adapt the gameplay with these limitations in mind.

Why Did It Take So Long?

I started on the idea of doing CvS in 2011. I had been running my fantasy RPG, Chronicles of Herenvale, for about six months and figured it was time to think about my next title. Running a live game and working on a new one can be a real challenge, though. There are always bugs to be fixed, improvements to be made and new features and content to be added. I put together some rough ideas and set out to find someone to breath life into them.

Early Examples
Early concept designs of female characters in CvS

I found an amazing artist, Don Ellis Aguillo, through craigslist in the summer of 2011. Craigslist has worked really well for me in finding talented contractors in CA. I live in Austin, TX, and we just don’t have the density of art talent the West Coast does. He put together some concepts based on my descriptions, and we were off to the races! But it turned out to be a marathon, not a sprint. My original asset list doubled, then tripled as I didn’t have time to really develop the game – and I didn’t want to lose all of our momentum.

Focusing is crucial to building anything, but can be so hard in this tech-driven, interrupting world!

Early 2012 didn’t actually show much progress on my new game, as I was spending a lot of time in Herenvale. Through the summer and fall, I buckled down and concentrated on constructing CvS based on top of the framework I had built for Herenvale. I then started adding features and modifying other features until I was buried (again!) – so much for a late 2012 release.

Finally in the Spring of 2013, I set aside a lot of time to focus almost exclusively on CvS. I was still adding quests, items and other things to Herenvale in addition to answering support emails, monitoring the servers, etc. Days of essentially uninterrupted time were necessary to finish work on the new game. That meant putting things I usually took care of every day on hold for days at a time or just checking email in the morning, or any other means I could find to focus. Focusing is crucial to building anything, but can be so hard in this tech-driven, interrupting world!

By June, I was ready to run a beta of CvS for some of my hardcore Herenvale fans – it’s great to have players who are willing to help. By end of July, I was ready to launch! That last push is so hard, but so rewarding when you reach the finish line.

Final Character Art

Shall We Play A Game?

CvS can be played here. The game is also on Kongregate and Facebook. I love connecting with other game developers, and I’m happy to share my experience, especially if it’s helpful. It’s almost time to figure out what I’ll be working on next if I can fit something in between running two live games.

You can reach Christian on Twitter!

ContributionsPostmortem

Postmortem: Astroflux

July 29, 2013 — by Mariia Lototska

feature11.jpg

The Stockholm-based game studio Fula Fisken (which literally means “The Ugly Fish” in Swedish) was founded in 2010 by Magnus Lind, Stefan Svebeck, and Toomas Hommik. Fula Fisken has four games on their list: Astroflux, Världskoll (for client UN on iOS/Android), Kolleagues (for client TCO on iOS/Android), and Sudoku and Friends (a multiplayer sudoku, playable on the web and Android). Magnus Lind shares the difficulties and the rewards of creating Astroflux, a real-time 2D space shooter MMORPG browser game.

Creating the Game We Want to Play

Everybody on the Fula Fisken team are game nerds; Toomas used to be late for work after long Xbox gaming sessions, Stefan’s a hardcore Starcraft player and plays through Diablo 2 on Hell Mode at least once a year, and Magnus plays everything from Starcraft to Clash of Clans and has a passion for retro games.

When we founded Fula Fisken, all we knew was that we wanted to create awesome games. We all had years of experience with running huge web-based games with twenty-ish database/web servers, and we also knew how a strong community could make a game flourish.

We started out in a super small office, which we shared with a T-shirt designer that collected toy robots. It was kind of surreal to have our own game studio, an office, and the possibility to create whatever we wanted to. One problem though: we did not have an income. We really needed to find contract work to pay the rent and get a minimum salary. Luckily, we got an offer to create a game for the Swedish Trade Union TCO. We were to come up with a concept that “showed the upsides of collaboration in a work place”. This was a dream come true: we got to make our first game and we would get paid for it! The result was Kolleagues, made together with artist Sandra Löv.

We still needed to decide what our first company game would be. The ideas were many, but we all liked the idea of a 2D space shooter, possibly with multiplayer elements, so alongside Kolleagues, we started working on what would become Astroflux.

Early Mockup
An early design of Astroflux

Getting Players into the Game

In May 2012, we were ready for our initial release of Astroflux. It was a rather quiet release, since we only shared the link with a few friends. Bugs were found and fixed. Eventually, Chris Benjaminsen at PlayerScale (now Yahoo), our backend service provider, tried the game and, if I dare to say, fell in love with it.

Chris played Astroflux more aggressively than any other player during beta and offered us advice. Having Chris helping us was awesome, and he pushed us into new territories, like finding publishers and monetizing strategies. Before Chris came along, we were mainly focusing on our game — but suddenly, we were also thinking of how to make it popular and profitable.

First Version
It was a rather quiet release, since we only shared the link with a few friends.

Too Many Players!

In October 2012, we released Astroflux on Kongregate. They put us in the “Hot New Games,” and we also ended up in the “Top Games This Month”. The influx of new players overwhelmed us. Within an hour, we had 300 players online and our server crashed.

The next few hours were horrible; it was the game’s best moment to date and the worst at same time. We simply couldn’t get the game going except for a few minutes at a time before it crashed again, and again, and again…Players rated the game one star, and we had to ask Kongregate to remove us from the “Hot new games” section. We realized that we needed to rewrite a huge part of our server code, because more than ~20 players online caused all kinds of bugs.

Bad time for a mission!
It was the game’s best moment to date and the worst at same time.

Getting Back on Track

The months following the release on Kongregate were spent fixing — everything. Astroflux needed the ability to handle more players, and we knew that just adding more servers wouldn’t help in the long run. Aron Pettersson, a robot engineer student with sick mathematical skills, was a great help in this process and with programming enemy AI.

We also received excellent feedback and ideas from Peter Eykemans at Kongregate, and we decided to make some changes. For example, rather than just having one spaceship, we added the ability to have a fleet, with different ships getting different stats. The thought behind this was to have more stuff for the players to upgrade, and to experience something else by playing with a different ship. The result was a success! The players loved it, and the incentive to play beyond a certain point grew significantly. This was a success for monetization.

In late May 2013, we were put on the front page of Kongregate again. This time, we were ready and we could handle the traffic. Our rating went up to 4.2 and it was great. We started to get decent payments and for the first time, it actually felt that we had something going for real. We got over 25,000 new players from Kongregate and peaked at 842 players online.

PvE
We got over 25,000 new players from Kongregate and peaked at 842 players online.

Battling Free2Play and Pay2Win

People in the business have told us that to make the game as profitable as possible, we should basically allow everyone to upgrade everything with our paid currency. They might be correct about that, but we’ve chosen a model that we’re more comfortable with.

First of all, the game is free-to-play. We have deliberately not added the super-mega-weapon for sale. Instead, we reward active players with quite a lot of our purchasable in-game currency. This way, the players get used to spending the in-game currency, and they also know that if they’re active, they will sooner or later be able to buy some of the cooler items. If a player gets curious about an upgrade or a ship, it’s possible to spend money to get it earlier. For us, the monetization process is very much about finding a balance between “free-to-play” and “pay-to-win”. These two often go hand in hand in games today. I don’t really mind it, but for a game like Astroflux, the balance is important for keeping the players happy.

By limiting a single part of the game (the player level), we’ve made sure that a player can’t pay their way to victory. A level 50 player that invests $100 in the game won’t be able to beat a level 80 player. Reaching level 80 is time consuming, but anyone can get there even without paying. Buying upgrades and a XP-boost will get you there faster, but you still need to invest quite a lot of game time. My experience is that players usually accept that you can get items faster by paying, as long as you still CAN get them for free.

Hangar
For us, the monetization process is very much about finding a balance between “free-to-play” and “pay-to-win”.

What’s Next?

We’ve got the game running smoothly and it monetizes decently, but we have some important steps to take:

1. Player retention

We lose too many players early on in the game — and we lose too many players because we don’t have enough incentives to play beyond a certain level.

2. Better PvP

Our current PvP system is inherited from an earlier Police vs Pirate-feature and does not work very well. We’ve made some quick fixes, but we need to re-think and re-make.

3. Improve Clans

We’ve got clans in the game, but they’re not really used for anything at the moment except showing that you are in a clan.

4. More content

Creating new star systems and enemies takes time. What we need is a way to allow players to create content or possibly dynamic content creation based on player level.

As you can see, we’ve got a lot of fun ahead of us!

A Bumpy Road

Even though we were experienced programmers when starting this project, we didn’t have experience in making real-time MMORPG:s. We certainly were in over our heads with Astroflux as one of our first projects. We probably would have been better off by making a couple of much smaller games to start with. We’ve literally spent months correcting our own newbie mistakes.

Real-time multiplayer also probably complicates things by a factor 10. It adds synchronization problems, lag problems, the need for servers, and much more. Simply don’t go there for your first game!

PVP Battles
Real-time multiplayer also probably complicates things by a factor 10.

Hands-On Tips

1. Don’t be cheap

Don’t be afraid of awarding active play with your purchasable in-game currency. Instead, make sure there is always something to spend it on.

2. Clans should not be expensive to create

We thought it was a good idea to make clans expensive to create and free to join, but after lowering the price to get your own clan, the engagement skyrocketed.

3. Last but not least: Think at least twice about real-time multiplayer

Even though I think having a running real-time MMO is super awesome, the extra complexity can really be a show stopper. In all honesty, there were many times during the creation of this game that we wished we’d started with something simpler.

Astroflux is available to play at astroflux.net. The Fula Fisken crew is currently planning their trip to Casual Connect in San Francisco where they will be showcasing the game. You can follow Magnus @magnuslind and Fula Fisken can be found @fulafisken, on Facebook and on fulafisken.se.

ContributionsPostmortem

Indie Showcase: Splitman 2 – In’s and Out’s of Independent Gaming

May 20, 2013 — by Mariia Lototska

feature7.jpg

Bari Silvestre and Gilbert De Vera were both friends and classmates in high school that shared the same passion for arts and gaming. They both dreamed of creating their own games in the future, which led them to where they are now. Splitman is one of the many games they’ve made. For them, Splitman reflects how they both reached their dream….. and it’s by means of cooperation.

Splitman was created back in August 2011, when Bari was so inspired by the game Gravinaytor, a clone of another game called The Gravity Game. Bari wanted to incorporate several key features of the game and the visual of another very popular video game, Megaman. Bari asked Gilbert to learn pixel art to deliver a Megaman-like graphics for their upcoming game. And this is where their journey started.

Meet the 2-Man Team

BariBari Silvestre is a self taught programmer and game designer. Although he always had the passion for drawing and sketching, it wasn’t his first career. His professional career started off making use of his degree in Accounting by doing clerical jobs after graduating from college. But the artist in Bari kept on haunting him. Last 2010, he finally decided to fulfill his lifelong passion of becoming a graphic artist. But as we all know, following your passion isn’t always too easy. Bari has been a freelance game developer for more than three years. Aside from making flash games like Belial, Splitman, Zombies vs Penguins, Pretentious Game and Vanguards, He has also released mobile games like Pinoy Hangman and Pingu’s Quest.

gilbertGilbert De Vera has always thought of creating graphics for games, even as a kid. But, like Bari, it didn’t happen right away. In his case, his professional career started off in the BPO industry after graduating from college. But in 2010, he decided to accomplish his dream of becoming a game artist through the game Belial, which he co-created with Bari. He has also been in the gaming industry for more than three years as an animator and graphic artist. A proud husband and father of two boys, he has been involved in the creation of games like: Villainous, Zombies vs. Penguins, Belial and Nano War – to name a few.

Splitman 1 vs. Splitman 2

Splitman 1 and 2 were both widely accepted by players; this is pretty evident on the numbers of plays these games received.  They were both featured on www.jayisgames.com, front paged on Newgrounds.com and both were featured on Casual Connect 2012-2013 respectively. But what really makes these two versions different from each other? Here are the lists of differences that can be observed:

Splitman 1

Splitman– Pixel art
– Story style game instruction
– No level screen

 

 

 

 

Splitman 2

Splitman 2– Vector graphics
– Simple game instruction
– Added Keycards to collect
– Level  screen added
– 2 different endings
– Added reverse gravity
– Can stand on dead clones

Re-Inventing Splitman 2

The idea of doing some changes on Splitman 2 came up, when Bari and Gilbert were talking about how Megaman evolved from pixel art into a cartoony type of graphics. Since Splitman was based on this popular video game, they thought of doing those changes as well. Aside from the graphics, they’ve also considered adding some features that will make the game slightly different from its former version:

They’ve added the reverse gravity feature that really made the game more appealing and challenging.

Keycards were added for additional objectives players can complete, these keycards are needed if players wanted to see the alternate ending.

Talking about alternate ending, Bari and Gilbert were really used to making alternate endings for their games. They made several versions of Belial, wherein players can choose the ending they want to see, and they thought of incorporating this technique to Splitman 2 as well.

One unique feature of Splitman 2 is that the character can step on to its dead clones. This is pretty useful when traversing a series of spikes hazard.

Obstacles and Trials

There’s never been an easy way to success, as many would say, and this proved to be true when Bari and Gilbert created Splitman 2. Both are the father of two children, and are both working at home, so it only means that there’s a lot of temptation not to be able to work efficiently due to different distractions inside the house. Also a minor setback is that Bari’s location is a few miles away from Gilbert’s place, so they only communicate via skype or via SMS. However, these hindrances didn’t stop them on creating games. They were able to cope up with these obstacles and have become used to working with it.

Sweet Success

So far, both Splitman 1 and Splitman 2 did a fantastic job of satisfying its players, evident with the positive feedback they’re getting and the number of gameplays it has reached. Splitman 2 alone has reached 5 million views from February to March and is continuously growing. The game was featured on many big web portal including Newgrounds, Notdoppler, Mochi and many more. Here are some of the great comments we’ve received from different big portals:

NewGrounds

Kongregate

14

Bari and Gilbert are already planning to create the third installment of Splitman this year. They’ve already showed a glimpse of what’s coming at the alternate ending of Splitman 2, so fans could expect another exciting and challenging edition of Splitman on the coming months ahead.

ContributionsPostmortem

Indie Showcase: Critical Force Entertainment’s Critical Missions: SWAT (iOS, Android and Web)

April 8, 2013 — by Martijn van Dijk

critical_missions_featured.jpg

Critical Force Entertainment Ltd is a new game development studio founded in Kajaani, Finland. The studio created Critical Missions: SWAT, a first-person shooter available for iOS, Andriod (released under Studio OnMars) and playable on Kongregate. The company focuses on developing premium and free-to-play crossplatform games with a special focus on the Asian market. So far, the company is self-funded, but investors are welcome. 

Veli-Pekka Piirainen is CEO and founder of Critical Force Entertainment Ltd. He is a former studio manager of Supercell North as well as a lecturer and head of Kajak Game Development Lab. Piirainen is also co-founder of NMP Games Ltd.

A student’s hobby project

Veli-Pekka Piirainen
Veli-Pekka Piirainen

In December 2011, I hired Igor Levochkin – one of the students at a school I taught at – as a programmer in my new startup company after following his work for the past two years. Igor and I would make games for the Apple AppStore, and we started making a prototype of a game called BomberBall. At the same time, Igor put his hobby game project in Kongregate. Early January 2012, Igor showed me that there were hundreds of players playing his hobby project game, but I didn’t pay much attention to it. I just thought it could be a good marketing channel for our iOS game.

However, at the end of January 2012, there were a couple of thousand players playing it and I started to get more interested in it. I gave Igor a Sony Xperia Play phone and told him to port the game to that device. Igor managed to have it up and running in a matter of days. Next, I told Igor to port the game to iOS; this was bit more difficult since he was not familiar with Mac and Xcode. After a week, the game was also running on iOS. Now I really started to see some potential in the game. Despite all this work on Igor’s project, we also continued to develop BomberBall because I wanted to have a good prototype for the GDC in San Francisco. I demonstrated both prototypes at the GDC and Igor’s project, Critical Strike Portable, gained more interest from the public. After that trip, we decided to concentrate fully on Critical Strike Portable.

Keeping up with high popularity

Igor started fulltime development on Critical Strike Portable by adding new weapons and features. I still worked part time at the university and couldn’t fully concentrate on the game development. I trusted Igor and also a team of Russian volunteers, who supported us in the growth of the user community as well as map creation. Another important task was to make a proper and more user friendly User Interface (UI) for the game. Unfortunately, Unity 3D’s tools for this job were pretty limited and we didn’t have any artist or UI specialist in our team to design a nice, good-looking and functional UI. So Igor made a “coder-style” UI with many different settings and options inspired by Counter Strike style menus. That UI was easy to use with a mouse, but for mobile phones with touch screens, we needed a different kind of UI.

The user interface of the mobile version.
The user interface of the mobile version.

Because I was inexperienced in game marketing, I hired Teemu Riikonen in April 2012 to lead the studio as well as take care of publishing and marketing of the game. Our next employee was Thanabodi Thongchat, a 2D artist from Thailand. She started designing backgrounds and UI graphics for the game in June 2012. Igor implemented more and more features to the game like new game modes, zombies, graphical effects, as well as fixing bugs. We released new versions on Kongregate weekly and got feedback from players on how to improve the game. At the end of June 2012, we had nearly 30,000 daily average users playing the web version of our game, but we were still growing.

We got over 1 million downloads in one month.

On June 26th, we released a free Android version of our game with exactly the same UI and almost the same features as the web version. Even though it was not so easy to use and the menu elements were pretty small on a phone screen, its popularity surprised us. We got over 1 million downloads in one month.
But the problem was that many players didn’t continue the game after their first try. Only hardcore players did so. We decided to create a totally different and simpler UI for mobile devices, because the current quality was not good enough for Apple’s AppStore to sell it as a premium game.

At the end of August 2012, two game development students, Olli Lahtinen and Aapo Lehikoinen, started their internship in my company. They started to build a totally new UI, added new controls for the iOS version of the game with a new NGUI toolkit we bought from the Unity Asset Store and started to design new maps for the game with Hammer editor. We also needed new character models, guns and animations for the iOS version. Modeling and animations were outsourced to freelancers in Thailand and our Thai artist was leading that work. Unfortunately, the quality was poor and delivery was very late. After that, all animations were outsourced to two Finnish startup game studios and for the modeling of guns, I hired another student.

A screenshot of the zombiemode of Critical Missions: SWAT.
A screenshot of the zombiemode in Critical Missions: SWAT.

Unfortunately, we had to remake all maps done with the Hammer editor (16 total), because our lawyer said we probably weren’t allowed to use that tool, since it’s licensing agreement is not clear enough. Our lawyer also recommended us to change the name of the game from Critical Strike Portable to something else, because that name reminds too much of Valve’s Counter Strike (Critical Missions: SWAT was born then). Our original plan was to release the iOS version in the end of September, but it was released in the end of November due to these difficulties. A new Android version was released just before Christmas, a Lite version in the beginning of January 2013 and the Mac version is in the review process as of this writing.

The iOS market is very competitive

At the end of the year, the amount of our players had increased dramatically. We had almost 200,000 daily players on the web and over 100,000 daily players on mobile devices, but all were playing our free versions. Monetizing with premium version seemed to be much more difficult than we thought it would be. The iOS market is very competitive and full of games, so getting visibility is very hard. We also had bad luck with a very important review, because the reviewer didn’t like our controls at all (many other not so significant reviewers did like them, however). Because of this, we didn’t start to get income fast but our server costs rose dramatically due to the massive amount of users. We also had some trouble with one specific server provider, who just calmly cut off the lines to our map server without any warning due to dramatically risen network traffic.

Looking back

Our biggest mistake was to save money in wrong places and get low quality from our international freelancers. We trusted our own artist’s capabilities to handle leading of the outsourcing, but she was too inexperienced for that. Of course, rates a quarter of the price compared to local studios were very attractive, but then the harsh reality revealed we had to do everything over again after that miserable trial period. It would have been wiser to use more professional outsourcing studios in the very beginning.

Our second mistake was not to solely focus on Critical Strike in the very beginning, but to also make the BomberBall prototype. Something else I would change was not to have a tighter management; everything went forward more or less without proper planning and scheduling. A fourth mistake was not to take a professional publisher to publish the premium iOS version. We thought it would be easy to self publish, because we had such great success with the free Android version, but we were wrong. A last mistake was not to pay enough attention to the server capacity, but that was more or less because of our inexperience with servers and also our idea to save money.

ContributionsPostmortem

Indie Showcase: Pixowl’s The Sandbox (iOS & Android)

March 5, 2013 — by Bart Eijk

featured.jpg

Headquartered in San Francisco, CA, Pixowl, Inc. is a multi-platform mobile game development company. Pixowl’s experienced international team strives to meet and exceed the desires of gamers worldwide by combining cutting-edge design with unique IPs, while making cost-effective production a priority. With the success of its two first iOS games, Doodle Grub & Safari Party, Pixowl is poised for a bright future in multi-platform games development.

Hello, my name is Sebastien Borget and I am co-founder and COO at Pixowl. I have managed the production and marketing of The Sandbox, a pixel art world-creating simulation game for iPhone and iPad (and as of recently, Android). The game was released in May 2012 and turned into a success with 2 million players up to now. Apple also nominated it for the iTunes Best Of Games 2012.

In the beginning, there was…

Their complementary talent could sparkle great things, once headed in the right direction

The project all started by hiring a developer Onimatrix (“Oni”), whose real name is Pablo Iglesias. We offered him to work on porting his personal project The Sandbox on Kongregate to mobile and tablet devices. By that time, his project already garnered some traction just on word of mouth, with 800.000 plays and a few thousands of worlds shared. It was targeted at a very niche category of players, with a pretty basic interface, no tutorial, instructions, levels or flow that you would typically enjoy on mobile games. It was, so to say, a pure physics sandbox: dropping elements over a black screen and checking their interactions – that’s it!

team

In the following weeks, we started assembling a team of seniors from the console game industry (more than 6 years of experience each) around Oni: Guillermo Averbuj as game designer, Sebastian Koziner as lead artist and project manager and Gerardo Heidel as a second developer. The team was working in perfect symbiosis and each developer knew each other from previous work experiences, thus facilitating the friendly relationship between them. Their complementary talent could sparkle great things, once headed in the right direction. But we still had to find it…

ContributionsPostmortem

Indie Showcase: Dream Symphony (Flash)

March 4, 2013 — by Martijn van Dijk

Dream-Symphony.jpg

George Zarkua is an indie game developer with four years of experience. He is the artist and game designer of Nude Hunter, Ragdoll Ball, Dream Symphony, and Spicy Story. He recently started his own studio, called Zarkua Studio.

A Vague Start

George Zarkua
George Zarkua

When the idea of this game appeared in my head, I had already done four projects in a rush mode. I enjoyed making games quickly, not only because it really saved valuable time, but also because this way the projects didn’t last long enough to bore me. The main idea for my new game was a bit blurry and existed only as an idea. The only clear thing was the concept, so I took it as a constant foundation that has been preserved from the beginning to the end of the development process.

I use tricks to escape the routine

When developing a game I try not to work in the same pattern over and over. I use tricks to escape the routine. On Dream Symphony, I tried to leave my comfort zone and tread into an area I had no experience with – music.

Despite the fact that indie games are mostly made without reliance on colorful graphics and effects, music always substantiates or even creates the atmosphere in those games.
Among flash games, there are outstanding representatives of the music games genre that are all based on the mechanics where static objects in the scene (such as obstacles or changes in the landscape) occur after the composition reaches a certain time or a certain pitch (Music in Motion and Take a Walk).

I wanted to create a musical flash game, but with rather unusual mechanics. The idea was simple: apart from the normal sound in the game, there were more sounds that were played in the interaction with surrounding objects. I.e. no objects were created depending on the music, but music was created by the objects.

The idea was very vague, and I could not explain what I wanted. So when I offered it to my partner programmers; four out of four said no. I created a thread in a forum, without a precise description of the concept. Before my idea gained any reputation, I got several messages from programmers who offered different kinds of partnership. There were about eight people and to each of them I described the idea. All of them were willing to work on the project., I couldn’t assess their levels, but one of them wrote GDD and showed his previous game with the mechanics of a platformer. This made for an easy choice. I chose Igor Kulakov. The last problem was me. I was tired after two years of working, so we agreed to start the game over time and I left for some relaxations.

At that time, I didn’t think that after the release of the game it would be featured on Newgrounds, Kongregate, NotDoppler, Bored, that we would win three cash prizes, including best game of Maypleyard, that I would read news about my game on leading indie news sites, including JayIsGames and that I would write this postmortem.

Character progress
Character progress

Before leaving, I made a couple of sketches of the main character (a huge goof pumped in a tracksuit, which jumped from cloud to cloud, and bursting bubbles with music) and a couple of backgrounds. It was cute, but not particularly interesting.

It was in a village deep in the heart of Russia where I decided to change the setting of the game. Originally, I had planned to make a quick jumper, with an active music. There I met a creature that changed my view of the future development of the plot. It was a sheep. I really wanted to see it in the game. I had only a pen and a notebook so all that I could do was make a few sketches. The body of the sheep looked like a cloud. In my head I animated the sheep slipping and awakening when you jump on it. This helped me to revive the game. However, the game still was in the same state, without any fundamental differences comparing to the flash jumper games, so I decided to add one more feature. The idea was that at the very beginning of the game the level was grey and while ascending in height, the game gradually became colored. This idea has also formed the basis for further work.

When I got home, the first thing to do for me was beating similar games. Meanwhile, my partner had created a working prototype. It was a very important step. After that, we coordinated our work through Dropbox and thanks to his prototype, we could work separately. He worked evenings and I started my work in the mornings.

Rush Mode

You just need to turn off the internet. Switch it off. You will reach zen

We worked in a rush mode, so the development of the game was enjoyable. Rush mode is the apotheosis of self-discipline, self-control and determination. In the morning, after you get up, cook all the necessary food for the day. Work should ideally take about 10-15 hours a day plus three hours for breaks. You should consider disabling all things that can give a signal. The most important discovery that I made for myself and increased my productivity 5-6 times is that you just need to turn off the internet. Switch it off. You will reach zen. The first time I came across this, lightning hit the transformer vault in my house. That day, I finished a big to-do list for the entire week and even cleaned the room, paid the bills and went for a walk. The second time, I fully encoded and made all the graphics to the simple little match-3 game, which I later sold it for 4k.

You must be completely off-line. And if suddenly you need the internet write down what you need on a piece of paper. In the evening spend an hour online and go to bed cheerfully. Of course, working like this for a long time might not be healthy, but you should try it.

Working on the Sound

The effect created a sense of passage and epicness

We needed a specific type of music with a perfect tempo and a certain feel to it. We hired a professional musician who had to rewrite the main track a few times because it didn’t quite fit the gameplay. When objects exploded, the sounds did not fit with the overall tone of the music. In addition, the levels in the game seamlessly switched, and the track for the next level was a copy of the previous one with the addition of one more instrument. This effect, coupled with the effect of saturation rising, created a sense of passage and epicness. By the end of the level, you could see a completely colored game, with a soundtrack that also felt complete.

The sheep in Dream Symphony
The sheep in Dream Symphony

When all the music was ready, it had to fit the levels. Connected tracks should feel solid. As an artist of this project, I needed a simple program for sound processing. I was looking for a free, easy program with minimum functionality and intuitive function names. I only needed to be able to change the tempo, the pitch, and the volume. By changing these aspects, the music comes to the foreground, and the sound echoed in the background.

The most important part was yet to come. We had to place the sounds in a way that the track seemed to be solid. To do this, sounds had to fit into the gameplay music. The player should feel that he was involved in the creation of music. It should not distract the player from the process. So some sounds had to be neutral and others had to be more in tune with the rest of the audio. The musical instruments only appeared in intervals where there was a deliberate pause. Needless to say, because of these actions the game was really hard to balance. In the end, the balancing of the game took about 30% of the work.

logo
SUPPORTED BY