Pre-alpha PROJECT PEGAS (RUS) GODOT

Hello to everyone who cares about development! 👾

Today - a frank conversation about why you haven't seen news on Project Pegasus in recent months. No, I haven't abandoned game development, as it might seem. I just realized that I was moving in the wrong direction in my projects. Of course, the limitations of development using neural networks also played a role. But this is not the main thing and here's why.

🌌 Dreamed of the stars, but stumbled at the start

When I was conceiving Pegasus, I wanted to make not just "another shooter", but a game where you can feel the control of the ship with every cell of your brain. But instead of focusing on mechanics... once again I delved into the depths of visuals. Detailing models, constantly redrawing interfaces and other things - all this took up 90% of the time. And then I opened the code and realized: the ships fly like irons, and the battle resembles tag rather than space battles.

Yes, the enemies now shoot accurately, and meteorites give a reward (thanks to the December improvements!). But without a polished "skeleton", even a cool picture will not save.

🛠 Code: a battle with neural networks and yourself

From the very beginning, I relied on the help of AI: Copilot, OpenAI, guides from YouTube. But when Microsoft "improved" Copilot, it began to praise me for the code instead of correcting it 😅. The transition to OpenAI saved, but even with it you have to wade through tons of meaningless answers.

I depend too much on neural networks instead of understanding the code myself. It's like building a house, renting the hands of builders, but not knowing where the load-bearing walls are. Result: bugs with flight physics, lags in the recharge system, and the promised "fueling stations" remained a blueprint. And that's not even scary, since this is a non-commercial story, I will still continue to do this with the help of neural networks. Games that are simple to implement can captivate the player.

🔥 What went wrong?

1. Perfectionism in visuals. Spent weeks on ship models, although the player is more interested in how they are controlled.

2. Blind faith in AI assistants. Without understanding the code, even neural networks will not save.

3. The desire to add EVERYTHING at once. Resources, trade, bosses - but without a stable foundation, this is just a pile of ideas.

4. Core mechanics are important to a beginner. This applies to both the player and the developer. The game should be easy to learn and catchy even if you just told its essence in words!

5. Even if the game is interesting to play ONLY for me, it is already worthy of seeing the light of day!

🛑 Pause? This is not the end

I am not abandoning Pegasus. I just decided to post what I have on itch.io for now. All the games I made are saved as projects and I can always continue them. Pegasus in the form in which you can play it here (link) both from a phone and from a computer will be changed later.

❓And what now?

I am developing a platformer-jumper, or rather I have an abandoned one, but now with a fresh look I will bring it to a playable stage, so that eventually I can also show it to you and let you play.

Mini-spoiler - this jumper is already very addictive and I often just get stuck in the gameplay, forgetting that in fact I should not play but check the mechanics and fix bugs.