Onion Assault is a 2.5D platformer heavily inspired by Super Mario Bros 2/Doki Doki Panic, in that you can pluck vegetables from the ground and throw them at enemies, in addition to lifting and throwing enemies you've jumped on. There are also three Onion Coins in each stage for completionists to collect, which is a feature brought over from more modern 2D platformers like NSMB and the modern Kirby games. It features a simple yet brilliant impressionistic art direction with vibrant primary colours and outlines around every object, some clever use of the grab mechanic (especially during bossfights and the final stage), and a mostly rewarding challenge in collecting all the Onion coins, in spite of the flaws holding it back. The game has a lot of really great ideas throughout. Unfortunately, most of those ideas fail to live up to their potential due to poor execution, questionable design choices and a lack of focus-testing that severely impacts the overall quality of the product. The main issues affecting this game stem from one source; the controls. Something so basic and fundamental for a platformer to get right. Yet the player character in this game is one of the most unreliable I've controlled in a while. Whereas platformers like Mario give you a good degree of control over where you're going to jump, and reliably convey information to the player such as jump height and distance when performing certain inputs, this game does the exact opposite. The character will often under shoot or overshoot a jump by default, forcing the player to make constant micro-adjustments mid-air if they're not already plummeting to their deaths after missing a platform. This issue didn't change the more time I spent with the game, and it's not a simple case of "getting used to the controls" or "git gud". The majority of deaths I experienced didn't feel like my own fault at all. Then there's the run button, which is standard fare for any platformers inspired by Mario. The issues I mentioned before are only further exacerbated with this mechanic, and worse yet, running is mapped to the same button as grabbing. So if you're already on an enemy or object, and want to do a running-jump off it to clear a gap, you're out of luck unless you already jumped on said enemy with the Y button already held down. I know SMB2 did this, but in that game's defense, the NES controller only had 2 face buttons; A and B. The Switch controllers have four face buttons and two shoulder and trigger buttons. There was no excuse not to take advantage of the real-estate on the Switch controllers, and map grabbing to a separate button from running, especially when it could've mitigated a lot of accidental deaths from the player. As it stands, this is an archaic and outdated design mentality. Speaking of outdated elements in the game, the player can take damage just from bumping into enemies from the side, even when they're not actively attacking or flaunting a weapon. When you're trying to jump on top of an enemy (some of which make a surprisingly small target) with the game's lethargic, unreliable controls, and repeatedly getting hit , this becomes a major problem. But even if it weren't, I don't understand why this particular hazard was put in place when games like Wario Land and recently Pizza Tower already fixed this issue. It'd make sense if the enemy's sides were covered in sharp spikes or if they brandished a spear, but when they haven't even noticed me before I bump into them from behind, I shouldn't be the one taking damage. Not to pick on the devs too much, but I get the impression that Horberg Productions got so used to their borked controls and levels that don't accommodate them very well, that they couldn't see the forest from the trees. The game was clearly in desperate need of focus-testing, even just with a couple of other people who enjoy these types of platformers, to course correct it early on and fix the fundamental issues permeating the experience. It would've humbled the devs and helped them identify these issues before they started charging money for it. I know my review has been scathing so far, and I do want to stress that this isn't a bad game, but it falls just short of being good despite its many worthwhile ideas that I can't help but wonder how and why we ended up with the mediocre product we got. I'm not going to give the devs a free pass, regardless of how small the team was, because Pizza Tower was released on the same day as this, made largely by one guy with less experience than any one member of Horburg Productions, and shows an abundance of competence, polish, detail and thought rivalling the best triple-A platformers, while providing a brutal yet mostly fair challenge for those trying to fully complete the game. If Bertil Horberg would spend less time isolated in his small bubble while virtue-signalling to idiots on twitter, maybe he could make something worth the asking price. There's no excuse here.

The game honestly plays more like a run and gun than a platformer. It does have a lot in common with old style platformers but not the tightness of the controls you would expect from SNES games. Enemies seem to respawn as soon as you move offscreen for a second, which is not a very fun mechanic, and really sets it apart, in a bad way, from some of it's influences. I died a lot more on early levels than I was expecting, but it was mostly because of what I consider to be fiddly controls than because the game is "hard for the right reasons". It is very fun picking up a person and throwing them off a cliff though. It's an alright game, and I don't mind supporting the dev who always puts his games on sale for dirt cheap, but I think it's more of a five dollar game. I like Polyroll, Juiced, and Splasher better, off the top of my head, not to mention the less budget titles which you already know.