With a day spend away from the making of the game, feedback, and a clear mind, I can share some thoughts about Cube Infection.
On the one hand, the gameplay is a bit hard to understand. That can make the whole game boring ’cause you don’t “get it”. Rules not clear enough, timing-based gameplay, with flashy lights everywhere, make the game a bit tricky at first.
This is the result of two bad moves :
As I was aware of my lacks in graphics and audio, I may have focus too much on them. I never made a game before, even a tiny one. I was quite disappointed when I saw the theme, thinking -“Oh no, being a villain usually is good for monsters or humans. Meaning I’m gonna have to draw…”. A bit simplistic, but when you only have 48h, I think you need to quickly pick a global guideline. In my case, instead of thinking “I want a cute little entity to eat all its siblings and evolve into a big badaboom badass monster”, I was more like “You know…I can draw squares and circles !”. And we’ll see later what can be done with it. -Meep-, bad idea. Some talented people can combine good visuals and creative mind, not me !
The second follows the first. With no global guidelines in mind, I was switching ideas almost every 15 minutes. Writing some a function for an idea, a class for another, etc..I ended up with a randomly generated grid, some squares, some links between them, money, explosions. Cool stuff for sure, but that does not make a game. The gameplay you have now is a sum of some ideas that came in my mind across the whol week-end, not a real gameplay thought from the beginning, as a whole one.
From the other hand, this leads me to what I consider to be cool visuals and music. And I was pleasantly surprised of that. Because of my gameplay problem, I have not been able to give alphas and betas on IRC. I began only with colored squares, then found a cool Photoshop tuto to make space backgrounds (Here, french). I was then able to give the game an identity. Flashy lights would give a great combo, along with a punchy music. (To be honnest, I now found the music to be a bit too fast !)
To sum up, I’d like to say that I was facing a situation were my strengths and weaknesses were just reverted 
Of course, some little (huge ?) other mistakes were done during these two days, like beginning without testing my environnement (I lost one hour recompiling SFML from scratch…), or not testing the game on other PCs until the very end. But everything went really well, and I’m proud of my first Ludum Dare. I hope to be here next time.
The IRC channel was really motivating, and I want to thanks everyone that was there for the great mood !
See you.