Game analysis: Index

I´ll be reviewing some good 2D platformers, as a way to find what makes them great and try to use those concepts in Kitsune. 

Learn from these:

• 1001 Spikes
• Ape Scape
• Angry Video Game Nerd games: They actually have pretty interesting level designs (skewed towards almost unfair challenge in 1, and slightly easy in 2)
• Braid: Focus on puzzles
• Castlevania 1: Levels and enemy patterns designed around the subweapons and player jump arcs
• Castlevania 3: Branching paths and level design
• Castlevania 4: Camera control, player movement and vast design differences from 1-3
• Castlevania: Order of Ecclesia: Nice level design, melee combat
• Cave Story: Pay special attention to the camera
• Contra 3/4: Study how respawning enemies are an integral part of the level design and the movement palette is still built around that
• Donkey Kong '94 (Gameboy): Triple jump mechanic+backflip, level designs... most of it
• Donkey Kong Country 2
• Donkey Kong Country Returns
• Donkey Kong Country Tropical Freeze: Good boss fights
• Duck Tales Remastered
• Freedom Planet: Modern Sonic done right
• Guacamelee: Cohesive graphical style, soundtrack and responsive controls. Great mix of action, platforming and exploration.
• Gunvolt: MMZ style bossfights
• I wanna be the Guy: Absurdly ultra difficult, etc, to check out how they handle difficulty and player anticipation/bait and switch (watch normal/blind playthrough and speedrun playthroughs, interesting stuff to see there regarding player movement, how to lead the player, etc)
• Leap Day (Android): Awesome way to make a responsive platformer for mobile. Great level design, very good challenges for double jumping and wall jumping.
• Mega Man 2
• Mega Man 3
• Megaman 3 (Gameboy): Level design and camera. Screen space is a LOT more limited than on NES
• Megaman 5 (Gameboy): Level design and camera. Screen space is a LOT more limited than on NES
• Mega Man 9
• Mega Man 10
• Megaman X for its everything (it does teaching the mechanics through its introduction stages/menu better than Super Metroid - honest!)
• Megaman Xtreme
• Mega Man Zero 1: Focus on boss fights
• Mega Man Zero 2: Focus on boss fights, excellent level design
• Mega Man Zero 3: Focus on boss fights, excellent level design
• Mega Man Zero 4: Focus on boss fights
• Metroid Zero Mission: Overall movement palette, and level design alternative paths for speedrunning
• New Super Mario Bros U: Best level design in Mario games?
• New Super Luigi U: Small tweak of mechanics to spice things up. Play immediately after Super Mario Bros U.
• Ori and the Blind Forest: Tight controls, original game mechanics, exploration
• Rayman Jungle Run
• Rayman Legends
• Shantae Risky´s Revenge: Polished up version of the original with similar level design considerations and a very interesting take on enemy patterns where they usually throw you off rythm compared to what you´re usually used to. Most enemies will either anticipate or wait a second more than you´d think, making most battles require a slightly different rythm than in similar games.
• Shantae the Pirate´s Curse: More varied traversal palette. more linear LD, but more open ended.
• Shovel Knight: Level design, development of mechanics, alternate paths looping back
• Sonic The Hedgehog (Master System)
• Sonic 1 Genesys: Original platforms & exploration
• Sonic & Knuckles
• Splatoon: Best holystic mechanics since Super Mario Bros.
• Super Mario World : THE GAME
• Super Mario Bros 2: Unconventional jump arcs (play anyone but Mario and relearn everything you think you know about platformers) and linear-yet-exploration-driven LD
• Super Mario Bros 3: Experimental
• Super Meat Boy: Focus on skill, challenge design & difficulty, tight controls
• Steamworld Dig: "Create your own level design" aspect and overall movement feel, one of the better ones in recent memory
• The Smurf's Nightmare (Gameboy): Triple jump mechanic
• The Wonderful 101: Game systems.
• Wario Land 3 (great exploration mechanics)
• Wario Land 4: Tighter Wario Land 3

---

What NOT to do

• Mutant Mudds/Xeodrifter are pretty "by the numbers", but it's interesting to see what that means exactly in their respective subgenres :-D
• Mercenary Kings: Mainly for their delayed jump, to study how terrible it feels (at least, I think it pretty much kills the game)
• Tintin, Spirou and Indiana Jones on Gameboy/SNES to see and understand how and why collision can go horribly wrong :-D (and that's in addition to poor level design and other fun stuff).
• Shantae (original): It´s LD is hampered by its camera (way too close, leading to a LOT of blind/faith jumps).

Shantae Risky´s Revenge (feat. Pehesse)

Shantae: Risky´s Revenge

Beatiful pixel art graphics and humor.

I had the great luck of being able to discuss this game with Honey Rose creator Pehesse:

• correojon: About backtracking: I´ve been playing Shantae Risky´s Revenge this week on the 3DS and well, the game´s beautiful and fun, the dungeons are great...but the constant backtracking really hurts it. It doesn´t use the upgrades you get to facilitate traversal in any way, in fact the game adds more and more time consuming enemies (the strong mummies, the mermaids which can hide in ponds of water to become invincible...) in the parts you have already traversed many times, placing them in locations that stop your momentum (like the fucking Imps in the trajectory of jumps over bottomless pits), making them respawn infinitely or without sense (enemies spawning just behind you, or right before you and instantly attacking). This hurts the game flow tremendously and even permeates into some dungeons: it seems the devs said "hey, a Metroidvania must have backtracking" but they didn´t stop to think about how to make it fun or even include it in the flow of the game. Instead of using backtracking to show the player how powerful he´s become it achieves the opposite effect and makes him feel like the new abbilities are extremely circunstancial and have only been designed for gating and to have an excuse to backtrack. Also, warp point locations make no sense, I can´t understand why there isn´t a general warp point in the city. The controls (turning and attacking always fails) and the hitboxes are a mess. I´m right before the last dungeon, but I just don´t want to traverse again the same locations to get to the door or to find more magic jams (which is a big pity, as the reward system for finding jam is awesome!). Is Pirate´s Curse better in these regards?
• Pehesse: Yes, I can see how you'd take things that way (I found your thoughts on Duck Tales Remastered quite interesting in the same regard, though I have a much more forgiving outlook on the game overall - I still prefer the original, and as you found out, find the experience on easy quite enjoyable). I agree about some of your points on enemy placement and rhythm on Risky's Revenge, though don't necessarily agree with your conclusion - I think it's actually part of the fun that they consciously throw you for a loop and off the common "flow". I believe it's a conscious design decision to have them work off a different rhythm than what you'd commonly expect, making the game hard to play when you rely on usual platformers rhythm and requiring to adapt to slightly different patterns. However, the backtracking overall is very poorly done, that I completely agree with, and the warp squids system especially is almost inexcusable - I hope it is a relic from the original design of the game which was butchered when the game was segmented into smaller parts, and not something that was planned at the later stages of development, because it demonstrates pretty poor judgement. Shantae games overall I don't find have the greatest level design (I'm surprised you criticize their controls, movement and hitboxes, though, since I think they're fairly precise and faithful to what's going on screen) - their use of backtracking is spotty, forward prevision isn't always the best either with blind jumps galore in the original and a general lack of sense of direction in RR. Pirate's Curse I think is better in all respects, though what it learnt in the process, it lost in personality - it feels a bit like a collection of design tricks used and recognized in many games since, which evidently work, but can't really claim originality or impact anymore. The heart of those games lie elsewhere for me (slightly off rhythm, as I said, and general presentation), so I can understand how viewing them strictly from a platforming study standpoint would make them a disappointing experience. Though that's why the adage exists: a game is sometimes more than the sum of its individual parts :-D
• correojon: I loved the Metroidvania approach at the beginning, but after revisiting the same location for the 4th time it stopped being fun. Besides, the upgrades you find for the monkey, elephant and mermaid seem almost one-time-use: you just use them for some very specific obstacle and may as well loose them afterwards; they are effectively keys. The controls worked pretty well most of the time, but turning and attacking would fail more times than work. Also, when walking next to a wall you had to release all buttons for Shantae to respond to any new commands. The hitbox of Shantae is fixed to the sprite instead of having soem safety zone, which would make you take some hits when only the tip of the sprite was colliding with the enemy, this is a big no-no and something games have been doing correctly by separating the sprite and hitbox since 30 years ago. When crouching I don´t know if it was because the controls took some time to react or the hitbox didn´t update in time, but I could never evade any attack at the last moment this way. The opposite happens when you are in monkey form: your hitbox is so small it was very difficult to grab chains or other things to climb in the middle of a jump. The backwards dash...it looks nice but I think I haven´t needed it once during the game. Maybe I used it in one bossfight, but apart from that it´s a move that makes no sense. It seems the devs put it in because Castlevanias usually have it, but didn´t take it into account when designing anything else in the game. The game has some platforming challenges, but I wouldn´t call this a platformer either. Anyway, it looks gorgeous and I´m studying it´s pixelart with detail and the music is great too, though not as good as Shovel Knight´s for my taste (after playing SK I´m a fan of Jake Kaufmann), though it may be more because of the style than anything else.

Ducktales Remastered

http://www.neogaf.com/forum/showthread.php?t=1231675

First impressions

I´ve been playing a lot of well regarded 2D platformers recently (Mario 3, SMW, Megaman 2&3, Super Meat Boy, Shovel Knight...) and even if I found some flaws in each of them, it was clear on every case why they were regarded as great games. But then I reached the next item in my list: Duck Tales Remastered.

The pogo stick mechanic is good, but I´m not capable of finding anything that makes me want to play the game. In fact I´m feeling like I constantly want to drop it and that it is wasting my time and doesn´t properly value it. Apart from the problems not related to gameplay (constant cutscenes totally breaking the pace of the game), these have been the worst offenders to me: Bad hitboxes on hazards: This is specially obvious in the Amazon stage, where one stray pixel of the spikes is enough to hit you. Obstacles that force you to wait: The mummies in Transilvania or the spiders over spikes in the Amazon force you to wait until you can kill them/pass through. This is specially atrociuos as if you loose all lives you have to repeat the whole level from the beginning, so it became really frustrating. Which brings up the next point: Why design the obstacles based on timing if you´re going to make the player repeat them over and over again when he dies (and he will die)? This game would have benefitted a lot from a design more oriented towards speed running, with obstacles thought in a way that a player that has already beat them once could easily find another way to get over them quickly. I haven´t found any mechanic progression or interesting obstacles thus far. Everything seems to derive from the same basic mechanics but doesn´t do anything interesting with them, it doesn´t twist them or try to add something new. All obstacles are either based on waiting for the correct time to push forward, jump or attack without going any deeper or having any other consequences. It feels like the intro stage shows everything there is to the game and the rest is just repetition under a different coating. The diamonds are infuriating: They spawn after you go through their location, so you have to constantly take 2 steps forward and 1 backwards to collect them. I started caring about them before I finished the intro level.

The game is difficult, I´m dying a lot, but I have no problem enduring death and retrying until I get better. I´ve finished many games more difficult than this one, if they ever gave me reasons to keep on trying and I´m not finding any of those reason in Duck Tales.

I´ve only played 4 levels (intro, Amazon, Transilvania and a bit of The Moon) but I already feel like I´ve had my share of this game. What am I missing? Why is this game so well regarded? I really can´t think of any reasons why anyone would willingly choose to play this over any other good 2D platformer.

Playing on easy

OK, so I tried the game in Easy and wow it really is much better now. The only differences are that enemies now deal half damage and that you have unlimited lives, so you no longer need to replay the whole level when you loose all lives. This makes the bad hitboxes much less frustrating. Also, instant deaths are not so bad now. IMO it´s a bad idea to put a limit in lives in a game focused on exploration, before I was too afraid to loose one of my precious lives trying to go after secrets, it really changes the way you play the game. Specially bad idea if there are instant death obstacles.

I played the African and the Hymalayan levels and I actually found them fun. I specially liked the mix of patterns of the rabbits and goats in the snow level, along with the restriction of getting stuck in snow if you pogo on it. I replayed the Amazon level and I enjoyed it a lot more this time. Also, I liked a lot the boss there. On the bad side, the game seems too easy now, there´s no real danger of dying apart from instant-death obstacles, There are no real challenges that require skill, enemies feel more like a small nuance than a real danger, specially if you find some of the heart upgrades. An intermediate setting between easy and medium with infinite lives but less health (and better hitboxes) could´ve hit the sweet spot.

The levels are a bit repetitive and the challenges don´t seem to follow any kind of progression: in the Amazon level you can find a secret room almost at the start where you have to bounce on 4 of those green creatures that jump from the water. They didn´t come back until much later in the level and the challenges they were used in were nowhere near the complexity of that first challenge, it felt a bit weird. Also I missed one piece of the engine in the snow level and had to retrace a big part of the level, when with a small modification of the level layout I could have gotten back in seconds...not something I liked, the layout felt a bit lazy.

I set hard pogo on but nothing changed and the option appears disabled in the settings now, maybe it´s not available in easy difficulty?

I´m not able to make small bounces, the controls feel really unresponsive when trying to do it. Big flaw as this limits severly the movement palette. Some challenges doesn´t seem to take this into account. The pogo bounce also failed sometimes if landing on the border of a platform.

This game is nowhere near Mario, Meat Boy or Mega Man, but at least I´m enjoying it now. It shows some bad design decissions here and there though.

Wrapping up

So I finished the game, I liked the last level a lot! Specially the last bossfight and the run to the top of the volcano. I don´t know if the complaints about this last run only apply to the original, but in this version I had no problem doing it on my second try.

One thing I liked about this game were the bosfights: the lunar rat was a bit simple, but the rest, with special mention to the Africa and final bosses were very good.

Not much to add to my last post, I ended enjoying the game :)

PS: The credits were awesome!

Super Meat Boy

Super Meat Boy (in progress)

31/05/2016

• Right trigger to run (optional): Apart from the standard face button, the RT can be used to run, just as I had programmed. I played using it and it further convinced me that this is a much better choice than using the face button, so I must encourage it in the game. Dropping the face button won´t be a good idea as many people are accustomed to it, so both options will be available at all times.
• Jump height control is really easy: Meat Boy reacts instantly to releasing the jump button with no accumulated momentum.
• Left and right movement: Meat Boy reacts instantly to left and right inputs, specially if the RUN button is held down. Maybe too much for my taste, this was clearly done to favor speed running. Direction correction in the air is not as responsive and there seems to be some momentum there. No overshoot in jumps, except maybe if the RUN key is held after jumping and before correcting.
• All momentum is killed when landing if no key is held: This gets some time to get used to, specially because left/right movement is so responsive, but it´s a great tool once you get the hang of it. Sonic in MS did the same and it helped a lot with platforming. Without this, the tight platforming in Meat Boy wouldn´t be possible. However, it feels weird when mixed with the high sideways acceleration when jumping.
• Level design is awesome: It teaches the player step by step all the necessary mechanics and evolves them gradually through the different levels.
• Level and game design are clearly focused towards speed running: The quick loop of dying->respawning or beating a level->going to the next one favor a fast rythm, as does the time counter at top and the result at the end of the level. There are alternate routes to try to get shorter times in every level and levels are short.
• There is also exploration: There are entries to a secret world and hidden bandages to get which favor exploration. The bandages also require tight platforming. However, with the fast rythm of the game exploration seems to be discouraged and feels out of place: when you explore you feel like you are wasting time because the rest of the game is telling you to hurry up.
• Graphics are a bit...meh: The game clearly shows it´s age and the small team that made it in this regard.
• Humor is awesome: Constant references to other games (Street Fighter, Castlevania, Mortal Kombat...). Dark humor (poor squirrels!) made me laugh more than once.

Sonic The Hedgehog (Master System)

Sonic (Master System)

Level design is simple at first, but all levels hide a fair amount of secrets (lives, chaos emeralds, ring monitors, barriers, invulnerabilty monitors...). Looking around (specially downwards) can be used fairly often to find secrets. Sonic is very heavy on the ground, but he has great air mobility, being able to accelerate almost to full speed in the air from a still jump. The thing that makes platforming work is that it´s very easy to kill momentum in the air and specially that if you aren´t pressing any key when you land Sonic will "stick" to the platform he lands on. Overshooting jumps, positioning in the air and falling directly downwards is the way to go, though momentum can be killed even in straight forward jumps. This makes platforming feel awesome and with very little practice makes you feel like you can nail almost any kind of jump. I´m getting the impression Kitsune slides too much after landing from a jump (Mario in SMW doesn´t and as I´ve seen, Sonic either) so I should tweak momentum cancelling when landing. In fact, this approach of killing momentum is constantly enforced by the level design: the player must carefully stop and look around to search for secrets and if he rushes forward he´ll surely crash into an enemy or a spring that will launch him towards spikes or some other hazard. This made that those times where the player could rush forward feel a bit out of place at first, mainly because you could blaze to a hazard or just miss many secrets in a long jump after a descending slope. However at the same time it feels very liberating at times (is specially cool in the long ramps in Scrap Brain, where enemies are even placed to be killed by rolling off slopes) and can be great for speed running.

Low jumps are a bit trickier to perform well, but they still work acceptably.

There are almost no "extreme" jumps required in the whole game (last stage still pending), the player can usually overshoot all jumps and use airtime to correct landing.

The Jungle Boss was great (airborne boss throwing cannonballs on a half-pipe). I also liked a lot the Labyrinth boss, appearing both at the bottom to shoot to the side the player´s on (must wait until the last moment and jump upwards to avoid) and at the top (alternating sides, which forces the player to follow/foresee that). He shoots a slow missile downwards which, when at the height of the player will launch forward with a bit of tracking on the Y-axis. A simple attack, but that worked very well: it´s easy to implement and understand for the player, but has a behaviour that can be menacing in some setups. Pity is it could be cheated by standing at the far side of the screen, so the boss will shoot the missiles facing outwards and they will disappear offscreen without posing a real challenge, but this is also a consequence of following the boss from side to side of the arena, so it could be intended and is a nice reward for the player being proactive.

Special mention to the Bridge background (specially the clouds and somewhat the mountains) but most of all, to the AMAZING Jungle background. Very simple, but it really gave out a kind of "hand-drawn" vibe, similar to Yoshi´s Island but on a lesser level due to tech constraints. I must try making something similar for the Flower Beds in Sakura Field or for settings with more vegetation.

I didn´t enjoy much the Labyrinth stage, mainly because moving underwater was veeery slow and non-responsive, some and sometimes water bubbles wouldn´t spawn.

I understand how I could sink so many hours in this game back in the day, it´s really good :)

Sometimes I would know an enemy, secret or hazard was there before they could appear. I don´t know if it´s because I still somehow remembered it (doubt it, it´s been +20 years!) or because the level design and object placement made sense.

Braid

Braid highlights

Started replaying Braid yesterday with a more analytic mindset trying to get all the puzzle pieces. It surprised me a little that the game is more about puzzles than platformingm even to the point of making the platforming almost trivial. Levels are just small sets of challenges (sometimes even 1 single challenge), there´s no jumping or anything outside the proper challenges, to the point of having to travel walking long distances. In this regard it feels a bit underwhelming, as I was expecting more complex designs, but the levels are just cut down to the bare minimum, while at the same time adding long walking sections here and there.

The puzzles are very clever, but sometimes there´s no way to continue but to exit the level and restart it. Also, sometimes you have to rewind A LOT. Many puzzles are not doable the first time as you´re on a tight timer, so by the time you see everything you need you have to rewind and some elements may be immune to time. BTW, having to rewind long periods of time is boring.

I´m liking the game and I´m finding the puzzles very challenging, but for the moment I´m not convinced this is the masterpiece everyone says, at least from a gameplay and level design view. Still, it´s a huge accomplishment for one single dev and I hope I can come up with something half as unique and great as this.

Edit (12/05/2016):

I reached the final world, some puzzles were really tricky, while others required too much precission which felt akward in a game that hasn´t required such thing anywhere. Still, it´s genious how the game bends it´s own mechanics and even reuses exactly the same levels with these new twists, changing them completely. The revisit to the Hunt level or the boss one with the new mechanics of worlds 4 and 5 were awesome: you instantly recognize them, you know exactly what objective you have but the mechanics transform how you interact with these levels.

I specially liked World 4´s mechanic: moving to the right advances time and moving to the left rewinds it (it seems it´s more like the position in the timeline is defined directly by the position of the player in the x-axis).

Note: This is an awesome idea and I think it could work really well in one of Ookami´s levels. It can be really tricky to setup though, so for this level I could start by dividing the level into smaller rooms. Also, verticality doesn´t seem to be affected by the mechanic, so we could "easily" set up a vertical level, climbing a wal with time control would look really awesome.

Note: I should ask Jarosh if he got his inspiration from this level for the x-sensitive platforms in Octahedron.

An aspect I haven´t discussed yet about the game are the graphics: it´s really high quality, very artistic, like you are in a dream world or in a painting (everything indicates the author went for teh second one). The references to Mario games are very direct and I loved them. Music also sets this oniric and relflexive mood, so it matches really well the general tone of the game.

What I didn´t like were all the long texts in the books...I mean, maybe the story is great, but so much text really put me off when I just wanted to play and once I skipped one book, I just couldn´t bother to read the rest.

The more I play this game the less I think it´s a platformer, it´s a puzzle game where platforming is really not more important than the time mechanics, so even calling it a puzzle platformer doesn´t seem right. This is not something bad, it´s just a personal appreciation meaning that I think you really shouldn´t compare this game to SMW, just like you wouldn´t compare SMW and ALttP. I won´t be able to take anything regarding movement or platforming...however the time mechanics and puzzle settings (specially the revisiting) are top tier concepts I can see working in Kitsune.

Only one world remaining, tonight Braid finally falls!

Edit (16/05/2016): Finished the game, the last world presented a new mechanic where you could create a bubble where time will be slowed down, it was nice. I didn´t like (and this is not only a problem in World 6) that some puzzles required very exact positioning or fiddling of the rewind-forward mechanic. An example is when you have to climb the falling ladder while rewinding-advancing, I just don´t understand how I finally solved that puzzle when I did...it didn´t feel very rewarding.

The last level was awesome, it´s the only instance in the whole game where some story telling is done through gameplay, not counting any obscure metaphors the dev used throughout the game. And it was awesome, totally changing the meaning of the scene to the opposite of the original meaning.

So overall, I enjoyed the game, there were some very clever puzzles but I didn´t like the walls of text for story telling, some times the mechanics felt inconsistent (or rather, their required use) and the levels were short and simple. I loved how it revisited old levels with new mechanics and how it introduced and developed the mechanics through each level. It´s definitely a puzzle game, not a platformer, though there are some platforming components.

Cave Story quick review

Cave Story highlights (in progress)

The camera is indeed very good, but I´m not sure if I can take it into Kitsune. The game is a Metroidvania and there´s not as much platforming as I expected (shooting takes precedence, even more than in Mega Man games I´d say...) and there is lots of vertical movement, but the player usually moves, then stops to shoot, then moves again. The camera drags a bit behind (sometimes it even does a bit of leading) and seems to have a lot of momentum, but the gameplay rythm with the player stopping so frequently always allows it to catch up in a nice way, which is something that won´t be happening in Kitsune, specially in fast platforming sections.

I´ll continue playing it (mostly because it´s an awesome game, much better than I remembered) and pay more attention to camera behaviour, I´m sure that even though I can´t implement this very same camera 1:1 in Kitsune, there are some very nice tricks I can take home.

One thing this game does exceptionally well is to design bidirectional challenges: often the player has to of through one area and later go back. The challenges are carefully designed so they work great in both directions, often presenting a similar but more complex challenge in the way back.

It also does really great gameplay tutorials, there´s a Youtube video about it I should link here.