The frustration that inevitably follows the initial euphoria

So I had this 10-15 year old open source game projects written in Java (AntiChess, SpiceTrade, Domino On Acid, 4DMaze) and I could quickly get them to run again and build with Gradle and I could do easily some static code analysis and throw out redundant stuff, fix deprecated code usage and reduce the overall amount of code, all without changing anything how these games looked or worked.

But now I have spent something like 20-30 hours (not really counted, could even be more) and frustration sets in. Mostly because

• There is no visible progress. I simplify the code base and write documentation and tests, but the resulting game itself does not change even a single bit. I realize how far I'm from making substantial edits.
• Real understanding of the code takes so much time. I have only gained superficial understanding of 4DMaze and Domino On Acid, no understanding of SpiceTrade and half-understanding of AntiChess so far. It really takes time to read the code, to understand possible side effects and to plan and conduct meaningful changes.
• I realize how much work it would be to make substantial changes (not only minor fixes). I do not plan substantial changes for 4DMaze and Domino On Acid, so these two might see an official new release sooner. AntiChess is more work than anticipated, the 4 player mode is only doing random moves and the 2 and 4 player modes do not harmonize very well currently. SpiceTrade I would like to do something, but it has this inbuilt scripting engine that is outdated.
• With this speed (which is probably a much more realistic estimate of how much time it takes to "improve" old games) I will never finish the hundreds of games I would like to engage with. I might need to give up on that dream and do some serious prioritization.
• It's easy to break things and the test coverage is far below 100%. More often than I would have liked, I need to go back and revert changes in order to not break the apps badly. Generification is a potential source, adding final is another one (when reflexion is used somewhere else in the game).
  

 That's all not really a reason to give up and there is still hope that I will get faster over time and can use synergies. I'll surely become a better Java programmer in this phase.