Software Project Search and Rescue

This article is adapted from the original article on the author’s site at AndrewComeau.com.

In 2023, I took on an epic project to create a Roguelike game in the C# language. I’ve been a fan of Rogue, the classic dungeon crawling game, since my college days and I saw it as the ultimate challenge after all these years of programming experience.

In addition to designing the game, I kept a programmer’s journal of the process on my site at AndrewComeau.com and posted the code to Github. I added a new chapter roughly once a week and showing many of the techniques I used for such things as map generation, drawing hallways, managing player inventory, calculating attacks and damage during fights with monsters, etc.. Aside from the programming challenge, I thought it would make a good tutorial for others who wanted to learn the language.

I started in February. Around July, after I’d written about 30 chapters, other life business started to intervene and the project had to go on hold. Coming back to an “epic” project is kind of like trying to come back to a relationship after a long period of silence; the silence ends up justifying itself and the effort it takes to break it seems to grow with every day.

Earlier this year, the project started calling to me again. I decided to compile the online version into a Kindle book and found I’d written over 58,000 words. Sunk cost fallacy aside, that in itself demands that this project not be forgotten. Finally, this month, I realized that I needed a project to focus on and I hearld the creak of the dungeon door opening.

Courtesy of Google Gemini —” The stone entrance to a dark and ominous dungeon hangs lightly open with a pair of glowing eyes peeking out. Torches adorn the wall on either side of the door.”

From the introduction to the 2026 project retrieval:

The last report from the Rogue C# expedition in 2023 ended with some notes about dubious experiments with scrolls and potions.

“There are some other potions that are a little more involved in their effects and might require more coding but I’ll detail those as I come to them.”

In light of this last, brief, report, it’s possible that our explorer might have gotten a little careless while testing some of the more exotic items — caution has always been advisable with anything found in these dark passages. It’s also possible that, like many others, he got a bit overconfident while exploring the depths of these caverns or became lost in one of the many twisting mazes filled with dead-ends, traps and the occasional hungry predator.

Three years later, many levels of the dungeon remain unexplored as the Amulet of Yendor remains down there, somewhere, waiting to be found along with the stores of gold and other treasures. Once again, we approach the entrance to the dungeon, equipped with advanced tools and even a couple of A.I. assistants that have evolved since the last expedition. Whatever the previous explorer’s fate, we also have his complete notes, wrestled away from one of the creatures during a brief foray into the dungeon earlier this year. Hopefully, they will serve as a guide.

The dungeon, its mysteries and its inhabitants await ….

To say that I’m completing the project is actually the wrong word because this series is just as much about demonstrating the C# language and solving the challenges that come with a large software project as it is about the game itself. The challenge now is to rescue that stalled project and carry it forward again wherever it might go from here while still finding the best ways to do that with C#. It’s hard to really say when a game like this is really “finished” but we can review what’s been done and what remains and bring the project to something that is worthy of being called Version 1.0.

The first step is to rediscover the project itself and take a fresh look at the requirements and the code that’s been written from them. Three years is a long time and that time brings a fresh perspective. In the upcoming chapters, I’ll be looking at where the project is at this point and some of the design choices that brought it here.

While reviewing the project for continuation, I ran into the question of which version of Microsoft Visual Studio I was going to use. The project was originally developed in Visual Studio 2022 a few years ago but Visual Studio 2026 Community is now available.

Then I noticed the memory recommendations. I don’t have a 64 GB machine and don’t plan on getting one but there didn’t seem to be much point in carrying a C# project forward if I couldn’t support the current version so I decided to see how concrete those requirements were by installing Visual Studio 2026 on both my 8 GB Lenovo Idea Pad 5 and my **Kamrui 16 GB mini PC** to see how it would work.

You can see how VS 2026 performed in my video on YouTube.

[embed]

That’s one technical challenge out of the way. There will be many more. If you’re interested in learning something about C# game programming or simply seeing how many new levels of this dungeon I can discover before getting lost, I’ll be sharing some of the chapters here on Medium and you can see the entire series on my site.