A developer postmortem of the good, the ugly and the bad of the development process of HackNBlocks, our latest puzzle game.
This post was originally posted at Ipsilon Development DevLog at http://ipsilondev.tumblr.com/.
Beware, this is gonna be a loooonggg post.
I want to share the experience of developing this game, with insights of what went well, what went wrong, and what we can improve in the process.
So, HackNBlocks was planning to be a simple game, with a common mechanic, and an appealing heroine that would empathize with the user, and motivate him to beat up the levels. That was the original idea, and my vision. The result was pretty different.
In the end, the game is like many of others, with no special attractive nor feature that would make it “shine”. The reasons for that, are below.
My main focus to make this game, special from others, was to connect through the main character and it’s mission. For this, i planned to show the character in special occasions and showing “feelings” (fear, joy, anger, etc..) when you where playing the levels. Unfortunately, the artist i hired to do the task, done a good and bad job at the same time. I asked for frames done in vector format, as i needed to re-size them at will, and i was not sure where i was going to be placed at that time. Even if the final art was good enough, she used a program to convert them to vector. that means that each frame had some inconsistency in size and placement of some elements, and it really looked awful when animated. Plus, the edition was impossible, why it had so many joints, that could not be manually fixed, so that destroyed the possibility of “empathize” with the user.
The mechanic was pretty simple, so i focused in making things run smoothly, faster and clean transitions, menu, GUI etc.. but I lost a lot of time, trying to do a gradient dynamic background that moved at 45 degrees. it looked nice but when i finally got the code done, and tested on all platforms, was a nightmare to align the elements on every platform. I could not give it up, so it took more hours to fix it, and then, saw that on low end devices, was crashing. this was poor choice of my part, why i ended with a nice and simple background that took me 10 hours of art & coding, and wasted around 50hs in the other that never worked. Must keep things simple.
Yes, you are reading right. Between coding, testing, preparing the art (that was done but still, must be tweaked to be used in the code) and posting in all the appstore, it took me that amount of time. I used OpenFL to code it, a framework i’m very familiar with, and i already had a codebase that managed the screen size and elements for all platforms, but it required several tweaks, to run this smooth, like threads. Still, most of the time went in the game itself, doing the levels and integrate them took me around 50hs (for 50 levels). Survival mode took me 80hs. It’s just too much. and i’m not sure i can actually improve it, but i think that any game would require this kind of time and that we could produce something of more quality, more unique, in the same amount of time, with a more proper planning.
I have been thinking a lot of this. I really love to do games, but i love to develop, to create things too. For that reason i do IoT, and other stuff. But game development is not really paying for the effort and most importantly, for the financial costs. i have been also thinking of changing the engine, for 3D support and see if i can improve the production time, but i’m aware that right now i can’t afford the curve of learning and adaptation. So i’m gonna stick with OpenFL. I’m not sure, even if i want, if i’m gonna do another game right away. I need to finish SaleFad (the retail store app) and i have other stuff in mind too. As a company, we need some real funding at this point, to keep doing and creating things.
I was aware that i needed to promote the game very early one, as marketing is one of the biggest problem with the flood of game and apps out there. I made everything i could, from posting on forums, to reddit, to TW, to FB etc.. i know i made wrong at keeping the devlog half empty and should be updated with real content, like all the things i’m writing here, posting it more progressively. But what i’m worried, is how to build an audience, that would be excited for the things we do. That’s my biggest fail. I need to work double at generating content (videos, posts, etc..) that would make the people to remember our company, our work, to follow our progress, why is that what would benefit to spread the word when we launch a new product or service. Plus, i need to build a relationship with the press, that now is nonexistent, but i know that i need to offer something that they would be interested, to have any kind of communication. It’s the chicken and the egg problem basically.
Has been a tough project, but i learned a lot. At the same, i identified the part we need to work on, double in marketing, better planning and code productivity. Even if i fail at code productivity, and have just a slightly better planning, if i fail at building an audience, we are doomed. So, there is where we are heading now.
If someone reads this, i hope it helps to other developers, and i will be posting the financials results of this game in a month aprox.
You can try the game at ipsilondev.com/hacknblocks
Read more about:
BlogsYou May Also Like
