I struggled with Samus Returns and the Prime games weren't my vibe, so I have to go back a long way for great Metroid. After seeing good reviews and a lot of positive word-of-mouth for Metroid Dread, it was near the top of my list of games to play when I'd finally pick up a Switch.
I was finally able to get my hands on my own copy of the game recently, and it started out promising, tapping into some of my Super Metroid memories. Unfortunately, I never really did get on with the game's stealth sections--if you want to call it that--with the EMMIs. Frustrating doesn't even begin to describe it. I hit game over more in this game than in every other Metroid game I've played, combined.
The game does eventually open up and it quickly grabbed me when it did. There is some incredible boss design here and (mostly) good map design. Locking map connection shortcuts behind power bombs wasn't a good move considering when you get them, but that faded to a minor annoyance by the end. After Samus Returns, it's easy to forget that you're supposed to get a bit lost.
What may be the highlight for me was, surprisingly, an interesting story. It's mostly told through atmosphere and subtle storytelling, like Hollow Knight, but there are some well-crafted cinematics sprinkled in. Samus is as expressive here as I've ever seen her. A really satisfying ending caps it off.
I'm excited to see what's next for the series. It's just a shame Dread has the EMMI sections because it just about guarantees I'll never replay it.
The thing is, different people have different tastes, and adding depth and complexity in a way that isn't overwhelming to the majority of customers takes time.
You can't just drop everything in from the start. You have to build in a sense of progression, and games with a certain level of mechanical depth are always going to shine most once you have expanded out into seeing the possibilities of what they can do. If you drop it all in at once, many of the "lower level" mechanics get trivialized and never explored properly by most players, removing a lot of the subtle details of what they allow.
Even going way back to something like Mario without a bunch of abilities, level 1-1 starts you out with a simple map and it slowly builds out. It needs to for you to learn it well enough to succeed at later levels (without prior platformer experience, which you can't assume).
I think that we mostly agree. My contention is that pretty much the entire game should still be engaging to play; having a long total play time shouldn't excuse that, and a shorter play time simply doesn't allow for it. Plenty of games have shown that it's possible to gradually layer mechanics one or two at a time, creating experiences around those smaller subsets of abilities that are still entertaining. I work in education and this idea is vital to what I do. Asking students to sit down and listen quietly as I feed them a mountain of boring details while promising, "Soon you'll know enough to do something interesting, just a little longer," is a sure-fire recipe for losing my audience.
And as I think you may have intimated, creating environments that require the use of only one ability at a time reduces those abilities to a boring list. When you've finally taught the player each ability in isolation, and suddenly start mixing everything up once they get to the "good part" of the game, they'll virtually have to "relearn" everything anyway.
We don't need to give the player everything at once to make our games interesting, but we do need to make sure that what we're giving them piecemeal is interesting in the moment.