On February 28, 2018, I played Shadow Complex Remastered on a whim, during a phase where I was digging into all of those games I had gotten for free and never looked into. This was one of those games, though I don't remember how or when I got it, and after I started streaming it, I liked it enough that I felt the need to make an excuse to keep playing it for more than a day. I thought it would be fun to have an alliterative gimmick stream month, and so the first Metroidvania March started on March 1, 2018. I've done this six times now and only now I've decided to take the time to write up my full thoughts for each game on this website. In the past I wrote reviews on my social medias and expected I would come back later and compile them into one place. This post contains revisions of those reviews. However, not all games were reviewed, so those are discussed only briefly.

1. Metroid

Year: 1986
Developer: Nintendo
Platform: NES

I've played Metroid 1 in several forms before, but it never held my interest for long enough to get to the first boss. As I rapidly approach the mid-30s, I feel that someone of my age with a vested interest in retro console games really should have better proficiency in the classics, so I decided now was a good time to commit to finishing the game that launched one of my favorite genres.

As the chat for my playthrough put it, Metroid 1 is a game that has gotten harder over the years, not easier. Modern video game design sensibilities would not forgive many aspects of its design, the worst of which is how you will always start a life with 30 energy. Never mind the common complaints about the map design, it's excessively difficult to make any headway if you happen to die, because it's very time-consuming to recover no matter where you have to do it. I had to resort to using an anti-grinding hack to finish it.

It's not all bad, though—I didn't have any problems getting around. The game world was small enough that I didn't need to draw out a map, and even when I felt like I was lost, the soft reset was always there to bail me out of a bad spot. I like to think this is due to my own experience with the genre and knowing how the limitations of the NES constrains the size of both the world and individual rooms. Even though a lot of rooms were identical in structure, the regions were at least still color-coded.

More than once I asked out loud why this Metroid game ever got popular, and especially why it got popular outside Japan. With how Japan loved their arcade games with obscure secrets needed to win, I would think Metroid would fit right in, but apparently not. I still don't have a real answer for that, and neither do I really feel a need to beat the game without the assistance of a hack that allows starting with full health instead of at death's door. I finished it, and that's good enough for me.

2. Axiom Verge 2

Year: 2021
Developer: Thomas Happ Games
Platform: PC

I've come to the conclusion that what I enjoyed most of the Axiom Verge games is how streamlined their upgrade paths are. What I like most about Metroidvania as a genre is the exploration and constant, steady character growth, and I think that's the strong point of Axiom Verge.

Axiom Verge 2 is to Axiom Verge 1 what Axiom Verge 1 was to Metroid. The first game kinda looks like Metroid to make you think that Trace will get similar movement capabilities, and then what he gets is nothing like Morph Ball and Space Jump. This game does the same thing to the first, in that when you find a familiar obstacle, you expect to get a similar upgrade, but Indra instead gets either something completely different or something from the first game that works completely differently.

While I didn't get the impression that these changes were done just for the sake of change, I did have an even harder time following its plot, and the first game for its faults at least had a conflict that was easier to get invested in. Once again, the lore is filled out in random order with a bunch of journals in unlikely places. This to me isn't good storytelling, and the story had most of the same themes as the first game anyway, except with names from a different culture. On that note, I guess it's kinda neat that the protagonist is a multi-billionaire CEO Indian woman who must be in her 50s by the time the game starts. I certainly can't think of any other video game protagonist who is anything like that.

As far as gameplay, it focuses much more on exploration over combat. In doing so, it took one step forward and two steps back. While I did enjoy the rapid progress and ability gain, Axiom Verge 2 discarded traditional upgrades for skill tree points. I hate skill trees, and it's not satisfying to kill a boss when all you get out of it is skill points. They're basically gift cards for power. Because all but a few boss encounters are ultimately optional, the final boss was necessarily anticlimactic. Nothing is ever a real threat because death is even less of a punishment than ever before.

Overall I did enjoy the game, even if it was fairly easy for the aforementioned reasons and the storyline wasn't told very well, but I can't decide if it's better than the first game. There's not as much fun stuff as the first game like the silly codes and bonus areas.

3. Castlevania: Aria of Sorrow

Year: 2003
Developer: Konami
Platform: Game Boy Advance

This was a game that I once came close to finishing, until I made the choice to let my sister borrow my DS Lite and a lot of my games, including this one. I never saw any of them again, so like with Metroid, I decided now was a good time to make right on this.

To me, what makes Metroidvanias fun is potential. You're made aware quickly of all these passages you can't cross and won't be able to for a long time, and then later you do just that and put on an excessive display of power as you do it. The best games put you on a steady diet of new ability gain and gradual strength increase, and I believe that Castlevania: Dawn of Sorrow struck this balance perfectly. Aria of Sorrow is the last prototype before perfection.

Aria of Sorrow is, compared to Dawn of Sorrow, unrefined at worst. Even if I did somehow go through much of the game without the slide ability (and make the Death fight much harder because of it), I was so into the game that I ended up streaming it in full in one sitting. You're always gaining new powers thanks to the soul system, and up to the very end of the game you're still getting new ones: free infinite super jumps and virtually-free super dashes. There aren't many better ways to put on a excess display of power in a Metroidvania. It took only one game later to get everything right. Even though Dawn of Sorrow has its own flaws, the gameplay papers over them. If nothing else, finishing Aria of Sorrow after all these years only confirmed that Dawn of Sorrow is still my favorite Metroidvania of all time.

One of the goals of Metroidvania March is to see if any newer games would be able to claim this title from Konami's 2005 masterpiece. The search continues...

4. Astalon: Tears of the Earth

Year: 2021
Developer: LABS Works
Publisher: Dangen Entertainment
Platform: PC

I made my skepticism of Astalon plain in my stream promo and my on-screen notes, because from the pixel art and grimdark setting, I expected it to echo games like Blasphemous and Hollow Knight, which incorporate Dark Souls mechanics and revel in making you not want to play them. I am happy to say that I was completely wrong to doubt it, because I liked it enough to get 100% in the main game on stream to see the true ending. Despite the problems I had with it, I was actually excited to play it through to the end. It got the scale and pacing right.

Even though I was dreading how it padded the upgrade path with features taken for granted by charging for them just like Hollow Knight did, it didn't amount to much. I suppose the purpose of that is to allow the player to not have those niceties if they don't want them. Sure, having to buy the ability to see where on the map locked doors are after seeing them was stupid, but you can get it early enough that it's not a big deal in practice, and there are easy tricks to vastly accelerate orb income. I always make a point of saying it's important to give the player something to do with in-game currency as soon as possible, and I guess that's one way to do it. What surprised me is how generous it is with it. The shop can only be accessed upon death, like Rogue Legacy. Unlike Rogue Legacy, or certain other games that build death into the game loop for that matter, it's not a use-it-or-lose-it affair. You don't lose any money on death or respawn. In fact, you lose nothing at all except your position on the map.

I fully expected to get screwed out of large gains from an unfair death, but that was impossible. Once I realized that, I recognized how accessible Astalon wanted to be, and I was all too thrilled to see it to completion with an offer like that. Going in, I was expecting the biggest holdup in the game would be the death mechanics, or possibly the bosses. The actual biggest enemy in the game is the ladders. Two years after release, they're so glitchy and unreliable that I have to wonder if they'll ever be fixed.

Speaking of the bosses, the hardest bosses in the game are the first two. The first because you have no powers, and the second because it takes too long to be able to damage it. Then many hours pass before the third and by then you're so strong that nothing can threaten you. The multiple playable characters are pretty unbalanced too, in that I was using two of them much more than the rest even after upgrading them all. The whole game is kind of uneven that way, but that didn't bother me with how precisely it controlled (ladders notwithstanding).

Astalon understands the thrill of seeing a bunch of restrictions and barricades for many hours and then, with a single upgrade, suddenly being able to steamroll right past them at just the right pacing. There will definitely be further playthroughs in future Metroidvania Marches. Part of the thrill, I think, comes from the structure. It's a single-screen game similar to La-Mulana, and there are dozens of screens with unique decorations of monster statues and grotesques used nowhere else. Without scrolling, it's always exciting to see the next screen. Also like La-Mulana, a few of the puzzles are mildly cryptic, but never ineffable, and some of the ladders are completely obscured unless you know to look for them, either by coming in from the other side or reading a clue on a faraway signpost. Either that, or you fall straight through the floor because it's hiding a ladder whose physics don't work right and fail to keep you on top for some reason. This could potentially happen in the sixth or so screen of the game.

5. Castlevania: Order of Ecclesia

Year: 2008
Developer: Konami
Platform: Nintendo DS

One of the problems I have with the name "Metroidvania" as a genre is that the genre as a whole has long advanced past both Metroid and Castlevania in every way. I gave some thought about why I enjoyed Order of Ecclesia in 2008 and why I didn't as like it as much on second playthrough. I've concluded that both of these were due to the game's desire to experiment with the mechanics and structure, being the 7th Castlevania of its kind.

On the surface, it tried to combine the gimmicks of Dawn of Sorrow and Portrait of Ruin with the Glyphs and the fractured world structure, but did a poor job of both. Glyphs replaced weapons, but the dependence on MP and the two-button controls that demand specific mash timing are stifling. The desire to break up the world into discrete sectors with a loose map connecting them resulted in a lot of recycled graphic sets, rooms, music, and Dracula's castle being about 25% the size of a standard castle, also with a lot of recycled rooms and music.

I think that 15 years ago, I didn't really know what I wanted in Metroidvanias yet, so the flaws weren't obvious to me. It felt familiar yet different, and so that was enough for me. Now that I'm paying more attention to pacing, I can't overlook the structural issues anymore. Even though there were multiple blatant filler sectors consisting of a single flat horizontal hall with no reason not to skip over them on the map once they're completed, I could probably forgive the repetitive nature of the game if enemies didn't deal and take so much damage.

I felt the design ambitions likely caused them to crash headfirst into system and budget limitations, so they padded game length by making enemies stronger than ever before and giving Shanoa the constitution of wax paper. At least mashing from game over to save point is quick. Even worse is that the extra modes aren't as interesting as the previous DS games. Albus isn't very fun to use even though he has a gun, because he can't run and shoot at the same time. Coming from Konami, a company famous for run and gun games, this didn't feel right at all, even if the inability to run while attacking is consistent with how Castlevania characters operate.

In the end I still had some enjoyment with Order of Ecclesia, but while I was disappointed with it by the end, it's still not the worst Castlevania directed by the IGA. But it is weird that Dracula's second form is just him going polygonal and becoming Rugal.

6. Shin Megami Tensei: Synchronicity Prologue

Year: 2017
Developer: Team Ladybug
Publisher: Atlus
Platform: PC

This is one of two unannounced bonus games for the month. I ended up with more spare time than I expected in the last week of March, so I decided to revisit a game I played only once, much like everyone else, due to its nature of being a marketing stunt for Shin Megami Tensei: Strange Journey Redux.

This game got an English fan translation but crediting it has proven unusually difficult. From what I could find, it was created by "Brento_Bento", but they have no recent online presence and download links for it don't credit them or include a readme or anything like that. I get the impression that they no longer want to be found if it's this hard to even get a name for a translation patch made so recently.

7. Castlevania: Dawn of Sorrow

Year: 2005
Developer: Konami
Platform: Nintendo DS

This game was played two times on a whim on consecutive streams, first with the original game, and then with the DSVania Randomizer.

8. Metroid: Zero Mission

Year: 2004
Developer: Nintendo
Platform: Game Boy Advance

Metroid: Zero Mission is a remake of Metroid 1 in ways I didn't expect. Rather than just kinda following the general outline of the original, it takes the same approach to storytelling, in that's it's minimal. A lot of people who played Metroid Fusion remember it for all of the dialogue it had, and Zero Mission, being on the same "engine" in terms of gameplay, does not. It has a few cutscenes but there's no speech, and the few times it does show text is Samus monologuing. The biggest problems I had with Metroid Fusion were the controls and the enemy damage output, and somehow neither of those were an issue in Zero Mission despite feeling exactly the same. I think this is due to the world layout having to follow an 18-year-old template.

I was not looking forward to having to wrestle with the Space Jump again, but it's one of the very last things earned in the game, and at no point is it the linchpin of any life-or-death situations. It somehow felt better to use even though it hasn't actually changed. Another point I noticed is the map. In Metroid 1, the game world was small enough that I never needed it, but here the world has no recycled rooms and differing backdrops that I found myself constantly pausing to see the map. It's a habit of mine to repeatedly glance at the map when it's available. I want to boast that I'm experienced enough in Metroidvanias to not need a map for Metroid 1 because I can mentally keep track of my position, yet when I do have a map I depend on it heavily. Maybe not having a map forced my brain to be more vigilant in paying attention to my surroundings in Metroid 1, which was harder than normal because the game necessarily reused so many screens and had a low color depth. When I don't need to do this, I just don't.

Zero Mission adds a lot of material to the Metroid 1 template, but everything before the first escape doesn't amount to much. The new midbosses feel like rejects from Fusion, not very challenging and there so there's more than three bosses in an already very short game. I really enjoyed the stealth action gameplay. Having only the bare minimum of defenses and outrunning the enemy because you're unable to kill them was actually thrilling, and the frequent save points that now refill health made death less frustrating than they could have been. This, of course, may also describe the EMMI encounters in Metroid Dread, but those went too far in punishing the player for failure and taking too long for giving the player the means to fight back. Zero Mission did just enough in both respects. I'm also just a sucker for the scenario where the powerful hero loses all of their resources and has to handle their problems like a civilian would until they can get them back. And here, you get that and more, with the three "Unknown Items" being revealed simultaneously.

Dread makes these segments a recurring feature, in a bleak whiteout environment, and tries too hard to be survival horror when you're still in armor. In Zero Mission, having no armor makes everything do up to 100 damage, which makes your earlier preparation matter that much more. Having more forgiveness for getting hit, rather than instant death and your only reprieve being an intentionally hard QTE, still works as survival horror. And you can still get jump-scared when a Space Pirate is hanging out in an unlikely spot.

On reflection, I think I might agree with contemporary reviews for Zero Mission in that the game could have stood to be longer, but a game being too short is the best problem it could have. I still don't have an answer for my questions about Metroid 1, but I had fun.