• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

A visual appreciation of Shovel Knight's first level, and of tutorials done RIGHT

Neiteio

Member
Dragon's Dogma is similar to this, but with a more complex system. It's an action RPG with a ton of elements, but it's never really that intrusive (well, it can be, but only if you decent that it should be).

You start off in an area that takes less than 10 minutes to beat. Here's how it goes down.

Th game immediately sets you off in a pre-determined Fighter class, which is the most basic and the easiest way to guide people into the game. That character is joined by a Pawn, which is an AI controlled character that comments on the surroundings. It's kinda constant, but it's done in such a manner that you get tips depending on the area you are in, or what you are fighting - goblins hate fire and such.

So anyway, the game leads you through a path where you learn how to fight different enemies. Ground combat is covered with the goblin fights, alongside how to utilize items (there's a barrel you can pick up and throw during the second fight, right beside the larger/greater goblin dude). Sky battles are covered through a fight with the harpies, and then they combine together, making you use the terrain to get at the harpies while managing the goblins below.

Then you fight a boss, a chimera, where your pawn tells you to climb away and hack. You have the support of other AI characters and more pawns, so while it looks daunting, it's actually not so bad, and your characters play off you and work in tandem. Even the boss teaches you to aim for weakpoints on monsters by stopping certain attacks once a head is cutoff.

There's a few problems here and there, but there's a lot more to that, and the beginning of the game that makes Dragon's Dogma a bar for good game design in the modern days.

Edit: I forgot to note the items system. See, you can combine shit to make better, laster longing shits. At the very start, you see some boxes, and because you are programmed as a gamer to smash everything and anything, they give you the reward of a herb. You look at your options menu and are about to eat it when, what's that? A combine option? Can't use that now, but hey, I know it's there so I'll come back to it later. I believe the Pawn even says something about that as well.

Also, there's a very obvious ledge you can grab, and if you make the jump, you get a reward for exploring. If you don't, you don't even take damage from the fall, you just have to walk up the way again and attempt a second time. I didn't know this until my second playthrough, but if you pay attention to the game, it will reveal many secrets without having a blinking arrow on screen.
I wanted to quote this at the top of the new page (for you 50-posts-per-page people).

Excellent analysis of the tutorial level from Dragon's Dogma. This is the kind of response I was looking for in this thread -- not just commentary on how Shovel Knight handles tutorials, but also commentary on how other games handle their own tutorials. :)

Also, I keep hearing great things about Dragon's Dogma. Maybe I'll check that out soon.
 
I'm not sure if this one is intentional or not, but SMT Nocturne has the Matador, essentially a get good or die boss. He shows the player how effective buffs, especially agility ones are and how devastating the enemy exploiting weaknesses can be. In order to beat the guy, you need to use debuffs, cover your weaknesses and form a decent party, and the lessons learned beating him can carry you through the rest of the game.

For something slightly different: A time when a game really needed to do this was Metal Gear Rising: Revengeance. During the final boss the player is required to do a fairly precise cut in blade mode. A lot of players seem to have trouble with this since the game never really necessitates a similar cut before that point. If the game forced the mechanic on the player a bit more earlier on it might have reduced the number of complaints at that point.
 

ghibli99

Member
Great OP, Neiteio! I felt many of the same things playing the first level my first time through. At first I didn't even realize that there wasn't any text popping up, and much of my enjoyment came from muscle memory and other traditions of old and modern platformers guiding my way. Finding the hidden areas is one of the best parts of Shovel Knight, and many areas are designed in such a way that if you flail wildly or wait too long, you or the enemies will destroy your one way to get there... and going off-screen and coming back doesn't regenerate certain blocks, so you either have to die and continue or replay later. Sucks if you're going for the no-checkpoint feat. :)
 

ZeroGravity

Member
These types of tutorials are great and all, but I don't think you can apply them to any old game.

Platformers, or games with rather simple mechanics are great for it but I fail to see how they can be applied to something more complex.
Yup, my feeling as well.

It's the same kind of response to people who rush to post Egoraptor's Mega Man sequelitis video in response to any discussion on tutorials. What works for one game isn't going to work for every game. There are many ways to teach players how to approach a game.
 

Effect

Member
I would like to think that they looked at Super Mario Bros 1-1 and basically applied it here but more fleshed out.



The only one I can think of at the moment is Wario Land the shake dimension and Donkey Kong Country Tropical Freeze.

This. It does look like they looked at Super Mario Bros 1-1 and how it was designed and the talk from developers (Miyamoto) surrounding it.
 

Neiteio

Member
One detail I neglected to mention in the OP is the first level also teaches you about dirt piles embedded in the walls. When you encounter them (on at least two separate occasions), you're immediately tempted to hit them with your shovel since they look the same as the dirt piles you found on the ground at the start of the level. But when you hit the ones on the wall, you dislodge a bouncing ball of rock instead. Hit it again for treasure.
 

Mistouze

user-friendly man-cashews
Damn you Neiteio, you're making the wait for the game on the european eshop so much harder.

Thanks for thread, really enjoyed reading that screen-by-screen analysis.
 

GG-Duo

Member
I've just started Shovel Knight, and along these lines I have a complaint:

I think it's lame the way they introduced the
magic platform books
in King Knight's stage.

It's not until the last third of the stage that you really encounter them - and you don't really get
enough time to learn all of their properties, such as how long before platforms disappear, and
that you can keep pogoing on an open book to ready yourself before jumping on the magic platforms.
As a result, I felt the magic platforms + pouring lava screen was much harder than the ones before.

I don’t know if they will appear throughout the game, but they probably should have introduced them in the tutorial stage too.
But I dunno, it may be just me…
 

Neiteio

Member
I've just started Shovel Knight, and along these lines I have a complaint:

I think it's lame the way they introduced the
magic platform books
in King Knight's stage.

It's not until the last third of the stage that you really encounter them - and you don't really get
enough time to learn all of their properties, such as how long before platforms disappear, and
that you can keep pogoing on an open book to ready yourself before jumping on the magic platforms.
As a result, I felt the magic platforms + pouring lava screen was much harder than the ones before.

I don’t know if they will appear throughout the game, but they probably should have introduced them in the tutorial stage too.
But I dunno, it may be just me…
If this is the first time you encounter them (~ 9:26 mark), they introduce them over solid ground so there's no penalty from falling. This allows you to learn their properties before they test you for real. :)
 
These types of tutorials are great and all, but I don't think you can apply them to any old game.

Platformers, or games with rather simple mechanics are great for it but I fail to see how they can be applied to something more complex.

What are some examples of games where this wouldn't work?

If Portal can take such a complex mechanic and perfectly ease the player into it, I find it difficult to think of a mechanic that couldn't be done this way.
 

Goldmund

Member
Nice breakdown. What's problematic is that this "training wheel" procedure (introduce element in a safe environment, vary/mix with other elements in a safe environment, remove other elements and add threat, mix with other already established ones and a threat) continues through what feels like the entire game, making the levels way longer than they need to be. Since the types of actions the game can safely ask any player to perform (due to upgrades being optional) are incredibly limited, this type of design should have stopped right there, after the Plains; the player won't have to rely on trial & error, he can basically only shovel/hit, jump and pogo; he can be trusted to make smart decisions when he encounters something new. Instead they tacked on a weaning period for almost anything (be it an enemy type, a type of platforming sequence, an environmental hazard etc.). It's super annoying and inelegant.
 

Neiteio

Member
Nice breakdown. What's problematic is that this "training wheel" procedure (introduce element in a safe environment, vary/mix with other elements in a safe environment, remove other elements and add threat, mix with other already established ones and a threat) continues through what feels like the entire game, making the levels way longer than they need to be. Since the types of actions the game can safely ask any player to perform (due to upgrades being optional) are incredibly limited, this type of design should have stopped right there, after the Plains; the player won't have to rely on trial & error, he can basically only shovel/hit, jump and pogo; he can be trusted to make smart decisions when he encounters something new. Instead they tacked on a weaning period for almost anything (be it an enemy type, a type of platforming sequence, an environmental hazard etc.). It's super annoying and inelegant.
I'll have to respectfully disagree on that last point. :)

In the previous example of Pridemoore Keep, your weaning period with the magic books lasts for one quick screen -- one quick jog up the magic books. A mere three seconds of training -- the equivalent of a "sentence" in the parlance of gameplay. What immediately follows on the next screen is a gauntlet featuring multiple such platforms over a bottomless pit with lava streams in-between that can knock you down if you mistime your jumps.

Give the 3-5 seconds reaching that point, the setup is hardly annoying or inelegant. It doesn't even stand out as a tutorial; it falls right in line with the game's natural rhythm and what would've just been "connective tissue" for pacing purposes in a lesser game.

The game trusts you to learn at a rapid rate at this point (the first "true" level) precisely because the tutorial was so thorough teaching you the core cadence of gameplay, and the mechanics that form its basic vocabulary. :)
 
Shovel Knight does a good job of introducing new mechanics throughout the game. Every level has some kind of twist that is introduced elegantly.

That said, it's not a particularly rare feat for this style of game, at least in this day and age. The opening level is good, for sure, but it's not god tier like 1-1 or the intro stage in Mega Man X. Off the top of my head, I'd put the first few levels of A Boy And His Blob (Wii) and Super Meat Boy above Shovel Knight.
 

Goldmund

Member
I'll have to respectfully disagree. In the previous example of Pridemoore Keep, your weaning period with the magics books lasts for one quick screen -- one quick jog up the magic books. A mere three seconds of training -- the equivalent of a "sentence" in the parlance of gameplay. What immediately follows on the next screen is a gauntlet featuring multiple such platforms over a bottomless pit with lava streams in-between that can knock you down if you mistime your jumps.

Give the 3-5 seconds reaching that point, the setup is hardly annoying or inelegant. It doesn't even stand out as a tutorial; it falls right in line with the game's natural rhythm and what would've just been "connective tissue" for pacing in lesser games.

The game trusts you to learn at a rapid rate at this point (the first "true" level) precisely because the tutorial was so thorough teaching you the core cadence of gameplay, and the mechanics that form its basic vocabulary. :)
Yeah, the magic books are a nice exception, and by no means the only one. Look at the stretch leading up to them, though, enemies like the jousting horse, the flying rats, or hazards like the cauldrons and the chandeliers -- same pattern. Another good example are the jellyfish in the Flying Machine. While they are introduced right away hovering over a pit, they're positioned in a way that it's very unlikely you'll not reach the ladder. Two screens further you have the variation: now they can electrocute you; if you fall, though, you merely are sent back to the previous screen. Two screens further you have both pits that lead to your death and electrocution alongside the introduction of two new elements. This pattern continues (the fans, the wind, the propeller guys, obscuring parts of the level); almost always in this way. This is the final regular level! Yeah, the occurrence of new elements is often cleverly interwoven with older elements at varying states of complexity (say one's two steps ahead, another one three), -- but it still strains the level design. Almost every level is too long. It's obviously not the only reason, but it's one.

EDIT: You edited your post. It sounded like your disagreed with the general critique. I guess you can ignore this post then.
 

Neiteio

Member
the top of my head, I'd put the first few levels of A Boy And His Blob (Wii) and Super Meat Boy above Shovel Knight.
But then you're comparing multiple levels from other games against one level from Shovel Knight. I know that's especially the case in Super Meat Boy, which has a very piecemeal approach of one (maybe two) mechanics per stage.

Shovel Knight's first level is a textbook example of the ideal way to teach, when possible, and what games should strive to emulate as closely as possible. Depending on the game, there may be a point where some systems require a more overt explanation -- but too many games take the easy way out well before that point.
 

Alo81

Low Poly Gynecologist
I recaptured your OP screenshots on PC if you'd like to use em. I set it up so they have pixel perfect edges, as opposed to the usual soft blurring that you see in stuff like the Wii version. It looks super crisp and good.

Here's the full album

Here's an example with the pixel perfect edges

shovelknight_2014_07_07_17_04_14_87_by_aloo81-d7pnu5h.png


I've also got them at their original resolution too

shovelknight_2014_07_07_17_04_14_87_by_aloo81-d7pnvpn.png
 

Neiteio

Member
It sounded like your disagreed with the general critique. I guess you can ignore this post then.
I'm not ignoring you. I appreciate your perspective. I'm just saying that for me, the weaning period for each new element was brief enough that it didn't burden the pacing or overall flow of each level. For me, it felt like the game was introducing new elements while more rigorously testing established ones, and layering them together before repeating the cycle anew... but the game was always clever about it, providing new arrangements that kept the game interesting.... such that I never noticed a level dragging. So I guess this is a case where my mileage just varied from yours. :)
 
But then you're comparing multiple levels from other games against one level from Shovel Knight. I know that's especially the case in Super Meat Boy, which has a very piecemeal approach of one (maybe two) mechanics per stage.

Shovel Knight's first level is a textbook example of the ideal way to teach, when possible, and what games should strive to emulate as closely as possible. Depending on the game, there may be a point where some systems require a more overt explanation -- but too many games take the easy way out well before that point.

That's true in a sense, but where levels begin and end is sort of arbitrary, plus I think a lot of people would agree that the levels in Shovel Knight are on the long side for this style of game. If we make the comparison in terms of content, I think the comparison is still apt.

I'm not disagreeing with you in any way, mind.
 

Neiteio

Member
That's true in a sense, but where levels begin and end is sort of arbitrary, plus I think a lot of people would agree that the levels in Shovel Knight are on the long side for this style of game. If we make the comparison in terms of content, I think the comparison is still apt.

I'm not disagreeing with you in any way, mind.
Sure, but what I'm trying to say is that it's not just a matter of how things are taught, but the context in which they are taught. That's why the "full level" distinction is key, IMO -- the mechanics in Shovel Knight are taught across a full level that feels like a full level, rather than a series of exhibits presented piecemeal. This masks the fact you're being taught anything in the first place, teaching players without most of them realizing they're being taught. When a new mechanic is bookended by the start and end of a stage, it's a bit too transparent. :)
 
I see your point.

BTW, if you haven't played A Boy And His Blob, you should check it out. Or at least watch a let's play of the beginning of the game.

Sean Velasco directed both Shovel Knight, and A Boy and His Blob.

Huh. I knew Yacht Club was ex-WayForward, but I never made the connection beyond that. Awesome. The guy definitely knows his way around tutorials.
 

CDX

Member
That said, it's not a particularly rare feat for this style of game, at least in this day and age. The opening level is good, for sure, but it's not god tier like 1-1 or the intro stage in Mega Man X. Off the top of my head, I'd put the first few levels of A Boy And His Blob (Wii) and Super Meat Boy above Shovel Knight.

Sean Velasco directed both Shovel Knight, and A Boy and His Blob.
 
At work, so I can't contribute anything substantive to the thread, but I just wanted to chime in re how much I appreciate level design breakdowns like these. Great job OP!

I will say that while it's more difficult to pull off this completely organic style in games with more complicated control schemes, I think there are principles that still can and should apply. One is that developers should create scenarios that help train players in the mechanics of their game, even if the mechanics themselves are introduced through a text-box or something. Also, I second the posters who cited Metroid Prime and Portal as examples of 3D games that get this right.
 

Elija2

Member
I don't see how this is an example of masterful game design. A bunch of games do this kind of stuff. All the game is doing is putting an obstacle in your way but you're interpreting it as "Look at how the game teaches you to use your abilities without explicitly telling you anything." I feel like you could make this case with almost any game.

For example, imagine you're playing a first-person shooter game for the first time and there are no tutorials whatsoever. That would be bad game design. But what if I described it like this: "The player comes across an enemy. The player quickly learns that the Right Trigger shoots their gun and can be used to defeat enemies. The player moves forward and comes across multiple enemies. This sequence hones the player's ability to defeat many enemies at once." Suddenly that sounds like pretty smart game design, but it isn't. That's pretty much how you're describing Shovel Knight though. The game isn't doing anything, the player is learning everything by themselves through perseverance. I do think that this is a fine way for the player to learn how to play a game, I just don't think it's anything special or worth pointing out as great game design.

Also, this is something you used as an example of good game design:

On the next screen, the player encounters another moving platform, this time over a bed of spikes, and a lone piece of treasure to tempt them into falling onto the spikes. If the player lands on the spikes, they will learn that spikes are instant death -- and then they will learn how checkpoints work by respawning at the checkpoint on the previous screen.

So basically, by dying the player will learn that things can kill them? How is that good game design? Maybe if an enemy jumped into the spikes and killed itself to show the player how spikes work, but as it is you have no idea that spikes instantly kill you until you hit them yourself. That's not good game design.
 

Neiteio

Member
For example, imagine you're playing a first-person shooter game for the first time and there are no tutorials whatsoever. That would be bad game design. But what if I described it like this: "The player comes across an enemy. The player quickly learns that the Right Trigger shoots their gun and can be used to defeat enemies. The player moves forward and comes across multiple enemies. This sequence hones the player's ability to defeat many enemies at once." Suddenly that sounds like pretty smart game design, but it isn't. That's pretty much how you're describing Shovel Knight though. The game isn't doing anything, the player is learning everything by themselves through perseverance. I do think that this is a fine way for the player to learn how to play a game, I just don't think it's anything special or worth pointing out as great game design.
Your example is a person attacking in a first-person shooter. The equivalent in Shovel Knight is the player swinging their shovel. I didn't call special attention to this, other than to show that they put an enemy and non-enemy element in close proximity so you learn from the onset your shovel has more than one use.

Besides, learning to attack is not comparable to the less obvious things the game teaches you without overt explanations. For example, coming across the sand block in the ground and discovering through experimentation that you can pogo downwards to clear those blocks is something most games would hold your hand and straight-up tell you ("Jump up and tap down to shovel into the ground"). And then on the next screen, the maneuver you used to go down can be used to go upwards -- but you have to figure this out, and the game nudges you toward the solution without ever telling it to you. By modern standards, neither solution is an obvious maneuver, yet the level layout leads you into learning it organically, as can be seen in blind playthroughs (Let's Plays), where people are stumped by that bit, not knowing the shovel drop is a move, but then they learn it in the course of regular gameplay. All while thinking of the first level as just that -- the first level. Not a tutorial.

It's very smart design. Many other games would take the easy way out and sign-post everything with a blatant explanation. But Shovel Knight makes the tutorial level feel like a full level and teaches without every breaking your immersion or interrupting the flow.

So basically, by dying the player will learn that things can kill them? How is that good game design? Maybe if an enemy jumped into the spikes and killed itself to show the player how spikes work, but as it is you have no idea that spikes instantly kill you until you hit them yourself. That's not good game design.
The spikes convey their danger by their very appearance, a basic tenet of game design that hearkens back to Nintendo's age-old stance: "If it's dangerous, make it look dangerous." So even mildly experienced players will avoid them. However, those who fall in will learn early on they're instant death, only to be reset to a checkpoint the previous screen and learn what that glowing orb was all about. If they proceed without falling into the spikes, they will learn it later, but it's there for players who perhaps didn't grow up with 8-bit/16-bit games and don't sail through it with ease. Other levels have more learning opportunities on spikes, which I may detail in a future post.
 
I don't see how this is an example of masterful game design. A bunch of games do this kind of stuff. All the game is doing is putting an obstacle in your way but you're interpreting it as "Look at how the game teaches you to use your abilities without explicitly telling you anything." I feel like you could make this case with almost any game.

For example, imagine you're playing a first-person shooter game for the first time and there are no tutorials whatsoever. That would be bad game design. But what if I described it like this: "The player comes across an enemy. The player quickly learns that the Right Trigger shoots their gun and can be used to defeat enemies. The player moves forward and comes across multiple enemies. This sequence hones the player's ability to defeat many enemies at once." Suddenly that sounds like pretty smart game design, but it isn't. That's pretty much how you're describing Shovel Knight though. The game isn't doing anything, the player is learning everything by themselves through perseverance. I do think that this is a fine way for the player to learn how to play a game, I just don't think it's anything special or worth pointing out as great game design.

Also, this is something you used as an example of good game design:



So basically, by dying the player will learn that things can kill them? How is that good game design? Maybe if an enemy jumped into the spikes and killed itself to show the player how spikes work, but as it is you have no idea that spikes instantly kill you until you hit them yourself. That's not good game design.

I think you're giving the designers too little credit. OP broke down the first level screen by screen, and I think doing so demonstrated how there is a clearly designed progression at work here: from introducing mechanics in isolation and under "safe" circumstances, to testing the player's understanding of said mechanics in increasingly dangerous situations, culminating in a scenario designed to test them all in concert. Design like that doesn't just happen.

Your example feels a bit disingenuous -- without more detail, I can't really comment on whether or not it would represent "good" or "bad" game design.

I disagree that allowing the player to experience insta-death spikes firsthand is somehow bad game design -- especially where, as in Shovel Knight, the pit is closely preceded by a checkpoint. I think that's as bout as "safe" as you can make insta-death. Besides, your idea of using an enemy wouldn't teach the player about the Dark Souls-like loot mechanic.
 

Goldmund

Member
One thing I always found strange is that one can't destroy the checkpoints in the Planes, which is somewhat of a missed opportunity considering that the player learned just one screen earlier how to pogo off bubbles and might have tried that technique on the checkpoint, seeing as they have a common look, to learn that it is in fact destructible. It would obviously have made the introduction of one-hit-kill spikes more difficult. Instead they clue in the player by removing the wings later.
 

Neiteio

Member
One thing I always found strange is that one can't destroy the checkpoints in the Planes, which is somewhat of a missed opportunity considering that the player learned just one screen earlier how to pogo off bubbles and might have tried that technique on the checkpoint, seeing as they have a common look, to learn that it is in fact destructible. It would obviously have made the introduction of one-hit-kill spikes more difficult. Instead they clue in the player by removing the wings later.
I think they didn't want players to get frustrated breaking the checkpoint (not knowing what it is) and then resetting to the start of the level if they fall in the spikes on the next screen. They chose to first establish them as checkpoints, leaving players to discover the risk-reward mechanic of breaking regular checkpoints later.
 

Elija2

Member
It's very smart design. Many other games would take the easy way out and sign-post everything with a blatant explanation. But Shovel Knight makes the tutorial level feel like a full level and teaches without every breaking your immersion or interrupting the flow.

So basically, you're just praising the game for not explicitly telling the player anything? Because that's the only thing that Shovel Knight does differently from most games and I don't consider that to be particularly smart. I think it's a good way of doing things, I just don't think it's smart.

The spikes convey their danger by their very appearance, a basic tenet of game design that hearkens back to Nintendo's age-old stance: "If it's dangerous, make it look dangerous." So even mildly experienced players will avoid them. However, those who fall in will learn early on they're instant death, only to be reset to a checkpoint the previous screen and learn what that glowing orb was all about. If they proceed without falling into the spikes, they will learn it later, but it's there for players who perhaps didn't grow up with 8-bit/16-bit games and don't sail through it with ease. Other levels have more learning opportunities on spikes, which I may detail in a future post.

How is that different from any other obstacle in any other game?

I think you're giving the designers too little credit. OP broke down the first level screen by screen, and I think doing so demonstrated how there is a clearly designed progression at work here: from introducing mechanics in isolation and under "safe" circumstances, to testing the player's understanding of said mechanics in increasingly dangerous situations, culminating in a scenario designed to test them all in concert. Design like that doesn't just happen.

I think design like that does just happen. There are a ton of games that go through the progression of "1. New mechanic is learned in easy situation. 2. Mechanic is used in more difficult situations."
 

Goldmund

Member
I think they didn't want players to get frustrated breaking the checkpoint (not knowing what it is) and then resetting to the start of the level if they fall in the spikes on the next screen. They chose to first establish them as checkpoints, leaving players to discover the risk-reward mechanic of breaking regular checkpoints later.
I think they take three strikes, right? The sphere starts to crack only a little at first. Maybe that would have been enough to make the player stop immediately. It's pretty clear that you might want to reconsider destroying something that's established as extremely positive by the audio cue.

A very minor gripe, of course; and their solution is perfectly acceptable.
 

Ferrio

Banned
What are some examples of games where this wouldn't work?

If Portal can take such a complex mechanic and perfectly ease the player into it, I find it difficult to think of a mechanic that couldn't be done this way.

Portals mechanics are simple. Go in portal, come out portal. What you can do with that comes with experimentation.

That's something that wont' happen in games with tons of subsystems and stats, and more than 2 buttons.
 

Ansatz

Member
What are some examples of games where this wouldn't work?

If Portal can take such a complex mechanic and perfectly ease the player into it, I find it difficult to think of a mechanic that couldn't be done this way.

The combat system in Bayonetta I'd say requires explanation. But I think everyone would agree that's completely fair.

It's the other stuff like camera paning to show you where the game wants you to go, outlining points of interest with flashing lights, telling you what to do through text. That sort of thing can be worked around by creating level design and graphics that makes the player instinctively understand what you want to convey.
 

Neiteio

Member
So basically, you're just praising the game for not explicitly telling the player anything? Because that's the only thing that Shovel Knight does differently from most games and I don't consider that to be particularly smart. I think it's a good way of doing things, I just don't think it's smart.
I'm praising the game for very smart and thoughtful design that teaches one new concept after another, and reinforces each in turn, all in the context of an actual level that never comes off as a tutorial.

The layout of the level, the arrangement of the enemies and obstacles, and the order in which everything is sequenced is a very deliberate act on the part of the developers that teaches you without you feeling like you're being taught. The game thus avoids the pitfall many titles fall into, where the opening act is clearly a tutorial and less fun for it.

Here, you learn through emergent gameplay -- a series of "aha" moments that are satisfying in their own right -- and the game trains you in subtler ways such as learning air control against the dragons, an essential skill in the boss fight at the end. It's a first level that teaches inputs without spelling them out, and application without demonstration. The layout of the level ensures you will learn everything you need to know, and then develop the proper technique to use what you learned correctly. :)

Honestly, I stand by every point in the OP, which I tried to make as clear as possible. Every moment is establishing concepts, big and small, which all entwine and build up to something more. No radio chatter telling you what to do, no illustrated picture pop-ups, no training rooms, nothing but pure gameplay, gameplay designed so well you learn just by playing, and without thinking, "I'm in a tutorial."
 
I think design like that does just happen. There are a ton of games that go through the progression of "1. New mechanic is learned in easy situation. 2. Mechanic is used in more difficult situations."

But it doesn't just happen. Level 1 in shovel knight isn't just an "easy" level -- it systematically teaches the player how to play without ever saying a word. I think that's praiseworthy.

The alternative -- say, throwing up a controller schematic and simply starting the game with an "easy" stage -- is lazy design, in my opinion.
 

Neiteio

Member
How is that different from any other obstacle in any other game?
No one ever said that particular bit doesn't have precedent in other games. What's exceptional is the way all of these elements fit together in the level as a whole: Whether it's a familiar fixture (checkpoints, spikes) or less obvious mechanics (shovel dropping down, and shovel dropping UP), every lesson is strung together in one cohesive level that eschews the trappings of the modern tutorial. You learn everything through playing, thanks to the carefully crafted level layout, and this includes not only inputs, but their applications (when and where to use them) and techniques (how to use them well). No "tips and tricks," no control schematics, no button prompts, no pics, no text, no dialogue whatsoever. Nothing resembling a training room whatsoever. Everything comes off as a real level, because it -is- a real level -- just an incredibly well-designed one that teaches through its design.
 

Arkage

Banned
Every moment is establishing concepts, big and small, which all entwine and build up to something more. No radio chatter telling you what to do, no illustrated picture pop-ups, no training rooms, nothing but pure gameplay, gameplay designed so well you learn just by playing, and without thinking, "I'm in a tutorial."

Coming at this from the viewpoint of an educator though, I don't believe you could say that this style of tutorial is the best, nor that it is more effective than other types, without the all important caveat of "for me". You're apparently a visual learner due to the creation of this thread, but for others visual context clues may not work the best. Maybe reading an explanation works best for someone, or hearing a narrator give instructions works best, etc. There are multiple learning styles and to claim this one is The One To Rule Them All is excluding the way others might prefer to learn a game's mechanics.
 

Neiteio

Member
Coming at this from the viewpoint of an educator though, I don't believe you could say that this style of tutorial is the best, nor that it is more effective than other types, without the all important caveat of "for me". You're apparently a visual learner due to the creation of this thread, but for others visual context clues may not work the best. Maybe reading an explanation works best for someone, or hearing a narrator give instructions works best, etc. There are multiple learning styles and to claim this one is The One To Rule Them All is excluding the way others might prefer to learn a game's mechanics.
I'd hazard to guess that when most people start a game, they would rather feel like they're playing it than reading text, or studying schematics, or listening to chatter from HQ. And like I said, you can pull up blind playthroughs where people learn all of the skills described in the way I described, just by playing. It's a commendable feat when pulled off. Would I penalize a game for failing to do the same? Of course not -- different games have different demands. But do I admire a game that tries to teach through action rather than words, pictures and other cues? Of course. :)
 

boxter432

Member
just downloaded and played the first level, partly due to this thread. first level was great..said "awesome" or something similar after pulling off some of the simple yet cool bounces. didn't read the whole first post and died on Black Knight the first time, he was much faster than I expected! timed some jumps poorly as he did his jumps
 

Bleep

Member
Coming at this from the viewpoint of an educator though, I don't believe you could say that this style of tutorial is the best, nor that it is more effective than other types, without the all important caveat of "for me". You're apparently a visual learner due to the creation of this thread, but for others visual context clues may not work the best. Maybe reading an explanation works best for someone, or hearing a narrator give instructions works best, etc. There are multiple learning styles and to claim this one is The One To Rule Them All is excluding the way others might prefer to learn a game's mechanics.

Unobtrusive tutorials make the early parts of games much less annoying on repeat playthroughs. Having the player feel like they are figuring out mechanics for themselves is also a more rewarding feeling than being specifically told how to tackle certain obstacles. It doesn't work for every game, but if the mechanics are straight forward and clear enough to work this way I will always prefer it to dialogue/text tutorials. If games need to have written tutorials to explain more complicated mechanics I prefer it to be a separate option from a menu that isn't part of the main game. Strategy games and older PC titles like Deus Ex and Half Life had separate tutorials from the main campaign so that the main game can be properly paced without forced instructions on how to play.
 

Neiteio

Member
Unobtrusive tutorials make the early parts of games much less annoying on repeat playthroughs.
This is also true. :) What I like about Shovel Knight's first level is it's perfect to replay when I feel up for some platforming goodness but don't want to commit to one of the longer levels. Sometimes I'll play it a couple times in a row to get my fix. It has a very satisfying progression to it, and it's fun to revisit after beating the game... You can see how daring you've become since you started. :)
 

Raitaro

Member
Fantastic OP Neiteio. Fully in line with Egoraptor's sequelitis and the cool discussions about Mario's stage design as some have pointed out, and something I'd wish we'd see more of on websites or youtube channels as well (feel free to point some out if there are more!).

If I could make a living out of studying how players of different skill levels play through opening levels in games then write reports on that in the form of the one you wrote to help improve level design and the understanding of player psychology, I'd be in heaven. (Do tell me if such jobs actually exist, and where! And if not for game levels, testing the way people use general training/educational tools, devices or software would be good as well. Isn't this part of Q&A perhaps?)

That said, I'm affraid I can't really get too deep into SK's opening level design itself yet as I'm waiting for the game to be released in Europe just as some others here are. One question that did pop into my mind while watching people like the Two Best Friends play this game, is more of a visual design nature: which castle is featured in the background of stage 1? As far as I can tell, none of the two initial knight castles look like that tall tower you see there, so which castle are we seeing then?

I love it when games, apart from everything you touch upon, also provide visual clues that allude to future segments, such as showing something in the background that you will later get to visit (which Dark Souls 2's Majula area does as well, though I'm not 100% sure if all those distant areas are actually in that game either as I haven't played it to completion yet). Hopefully, this castle is featured in the game world of SK in other words, making the castle in the background of stage 1 more than just nice eye-candy.
 

mclem

Member
It's the way old school games like Mega-Man did it.

Actually, I'm not sure Mega Man's a great example. I remember Air Man in MM2 - which if I recall correctly, the manual suggested you try first - starting with quite a difficult jumping section with instant pits all over the place. Then leaping from moving platform (cloud) to moving platform after clearing the lightning throwers off them.
 
If you absolutely must go beyond being incredibly subtle in order to properly teach the mechanics to the player, there are suggestions:

  • Make the tutorial just slightly less subtle, and blend it with the world. Portal is a brilliant example of this, as explained by Extra Credits. Portal brilliantly does a lot of what Shovel Knight does, but what instructions or explanations it does provide directly to the player is either non-intrusive (the controls are out of the way and can easily be ignored by players after their first playthrough) or blended with the world (GLADOS, basically).
  • Make the tutorial really fun. Apparently, the original God of War is a fantastic example of this.
  • Separate the tutorial from the main game, as previously suggested by Bleep. It allows players already experienced with the mechanics to ignore the tutorials and jump right in. This also has the advantage of not necessarily requiring the tutorial to make sense within the context of the main game, meaning you can do stuff trippy dream shit like in Kingdom Heats 1, for example. Vanquish has a separate tutorial that involves the main character testing his suit's various functions, if you want a recent example.
  • Provide an optional help menu that explains pretty much everything. This is especially useful for players who want to get back into a game from where they left off and forgot some of the mechanics from not having played it in a while.
  • If you can justify it, try and have another character or some kind of 'ghost' or 'hologram' demonstrate various mechanics before the player performs them, especially if the mechanic isn't immediately obvious just from the level design alone. This alerts the player that they can perform the mechanic, and will immediately attempt to figure out how.

Just some ideas.
 

mclem

Member
If you absolutely must go beyond being incredibly subtle in order to properly teach the mechanics to the player, there are suggestions:

  • Make the tutorial just slightly less subtle, and blend it with the world. Portal is a brilliant example of this, as explained by Extra Credits. Portal brilliantly does a lot of what Shovel Knight does, but what instructions or explanations it does provide directly to the player is either non-intrusive (the controls are out of the way and can easily be ignored by players after their first playthrough) or blended with the world (GLADOS, basically).
  • Make the tutorial really fun. Apparently, the original God of War is a fantastic example of this.
  • Separate the tutorial from the main game, as previously suggested by Bleep. It allows players already experienced with the mechanics to ignore the tutorials and jump right in. This also has the advantage of not necessarily requiring the tutorial to make sense within the context of the main game, meaning you can do stuff trippy dream shit like in Kingdom Heats 1, for example. Vanquish has a separate tutorial that involves the main character testing his suit's various functions, if you want a recent example.
  • Provide an optional help menu that explains pretty much everything. This is especially useful for players who want to get back into a game from where they left off and forgot some of the mechanics from not having played it in a while.
  • If you can justify it, try and have another character or some kind of 'ghost' or 'hologram' demonstrate various mechanics before the player performs them, especially if the mechanic isn't immediately obvious just from the level design alone. This alerts the player that they can perform the mechanic, and will immediately attempt to figure out how.

Just some ideas.

I'd say Half-Life 1's Hazard Course (and in turn, the tutorials from Op4. Can't speak for Blue Shift) is a good example of most of those points being done very right. It's optionally accessed from the menu, it's thematically tied very nicely with the game, it's demonstrated by the Hazard Course Hologram, and it covers some mechanics that *are* a bit unconventional for FPSes at the time (Long jump, mantling)
 
Top Bottom