The best choice is the one that you, personally, will complete and build a full game in. Presumably OP knew how to use gamemaker studio, and could either choose between:
1) Starting to build the game straight away in Game Maker Studio
and
2) Learning C# / Unity and then starting to build your own engine and plug together tile editors and things which will then let you build the game.
In the case above, Option 1 will probably have more success as you are starting to build the thing rather than defering building the thing.
Agreed, one thing I always reply when people ask about what language, middleware to use, I always give the non-answer of using whatever people are confortable with.
Minecraft would probably never have happened, if Notch started by asking in the forums if he should use Java or not.
Likewise many 8 and 16 bit titles started as BASIC, C, Pascal prototypes before being rewritten into Assembly, after their concept was proven, or thrown away before too much was invested into them.
Discussing what languages, middleware, API, whatever, instead of actually designing the game is what slows development down.
Yup, Toby Fox more or less did this exact sort of research in 2020 while developing Deltarune, and came to similar conclusions:
We had actually attempted to develop the game since the time too. Development started around March 2019 and a 99% work was spent on investigating engines alternate to GameMaker, which I used for Chapter 1.
Without getting into the details, I decided a few months ago to go back to GameMaker after all. It still felt like the best fit for the project. So using Chapter 1 as a base, we've started creating Chapter 2 since May 2020.
A lot of progress has been made since that time. I believe we can complete this chapter, content-wise, before the end of the year (not accounting for translation, bugtesting, and porting).
I feel very confident. And the strange thing is, even though we ended up using the original engine, I don't regret the lost time, either. Not only was I still busy designing the game, but during that long period, I was able to think of many ideas that make the game's story and characters better.
Obvious caveat: He already had development in the game on one engine so that may have swayed factors. And of course, DeltaRune isn't going to be struggling to perform on any reasonable platform. But ultimately, the tool didn't limit his choices enough to make the jump.
At the end of the day, work to your strengths and consider your scope. Many people here are programmers so I understand if they want something more flexible than Gamemaker. But I wouldn't necessarily jump to Unreal Engine 4/5 if I'm making a small-ish scale 2D game.
1) Starting to build the game straight away in Game Maker Studio
and
2) Learning C# / Unity and then starting to build your own engine and plug together tile editors and things which will then let you build the game.
In the case above, Option 1 will probably have more success as you are starting to build the thing rather than defering building the thing.