Play my game GENTRIFRY in browser/Win/Mac

I was very happy that I could attend Ludum Dare thanks to my family support (my girlfriend and her mum took care of our 18-months old daughter the whole weekend). I knew before the jam that I wanted to do a game with a procedurally generated city (which already covered several possible themes, like "you control the game, not the player") presented in isometric perspective, like my beloved SimCity 2000.

Plus, I already thought about a "story theme", gentrification, because there are few games about it, and I live in an area of Berlin which is about to get gentrified, so I had some expertise about it ;-)

When I read the chosen theme in the morning, I eventually came to the following scope:

Scope

• a game with a procedurally generated city
• a classic isometric camera
• done with Unity 3D, in 3D
• hence quick&dirty production of a lot of low-poly models in Blender and Pro Builder
• genre: real-time strategy, a very complex, thus time-consuming genre
• i.e. keeping the scope as small as possible if I want to cope with the Compo deadline!
• gameplay:
  • streets are haunted by enemies: they raise gentrification in each building within their 3x3 field surrounding per round.
  • player places "anti-forces" which do the opposite
  • enemies and heroes move along the streets, if they got stuck at the end or reach intersection, rotate randomly
  • game over when 50% of the city is gentrified, you win if you stay below for X rounds
• if enough time:
  • add more heroes and enemies
  • add UI (toggle group) for hero select, which would become necessary then
  • add a log output that explains a little bit
• deployment for browser to comfort the players who are too lazy to download all games - meaning that I had to stay on the ground regarding the technical requirements to have a performant WebGL built

I knew it could be dangerous to try too much, and real-time strategy is complicated – that's why I planned to do what I solemnly call modular development. Goal 1 was to finish a version which resembles the whole concept as a working game, Goal 2 was to add assets if there was some time left.

What went well

• modular development worked out very well! I had a playable game when approx. 60% of the game jam's time was over, did a long break and than modeled and included more heroes and enemies, recorded some sounds and did some UI design
• 3D modeling workflow got reasonably quick and effective: after a ca. 6-months Blender break, I turned it on again, had forgotten most of the shortcuts, got cold sweat because I thought I couldn't do any models and quit the jam immediately, then calmed down, watched 2 youtube tutorials and got into the first model, the filthy dog.
• this was just accomplished by Pro Builder: I "probuilderized" my Blender models and UV mapping with PB worked like a charm, I could even build up a simple 16-color texture atlas from scratch (don't know if it helps performance)

What went wrong

• I could only do basic balancing, which is deadly for a real-time strategy game. I realized too late that the first levels are actually harder than the final stage, because the influence of the enemies on the city is proportionally higher, i.e. gentrification advances much quicker
• like often, I got stuck in small game dev detail problems which turned out to be time-consuming and would have been simple to get rid of with a clear mind. This time, I struggled with the rotation of the figures when they change their direction… it took 1 hour at least and the solution was awfully easy.
• around 6 hours before deadline, I knew there wasn't enough time to do music
• the sound effects are very rough... some city atmosphere would have been nice

If you’d like to have a look at Gentrifry, play it in browser/Win/Mac and rate, please