Good Robot #39: Teaching Players to Good Robot

By Shamus
on Dec 1, 2015
Filed under:
Good Robot


Some other indie developers were nice enough to play the game and send us their feedback. A common theme in the feedback was that things were too chaotic. Or too random. Or unfocused. Or too busy.

Looking at the game, it’s easy to see why, and it’s easy to see how we slipped into this state. We made a system that let you make wildly divergent robots, simply by tweaking a text file. Since creating robots is easy and variety is good, then more robot types = more good, right? Aren’t games always bragging about how many enemy types they have? Isn’t this a selling point? “Fight over 12 different enemy types!” it says on the back of the box.

Only 12, AAA game? Pshaw. We have that many in the first 15 minutes of the game!

It made sense at the time, but when the feedback rolled in it was a forehead-slapping moment for all of us. Of course this will seem like random chaos to someone who hasn’t played the game constantly for 6 months.

It’s like a version of Half-Life 2 where your first fight is against a group of foes with the behaviors of an antlion, a zombie, two soldiers, a metrocop, a strider, and a gunship. It’s not about being “too hard”, really. Even if you lower the hitpoints and damage output on the gunship and the strider to bring them into line with (say) a metrocop, the player still can’t be expected to parse all this chaos. They’re not going to appreciate the differences between the soldier and the zombie when both foes die in the same panicked volley of weapons fire.

The Debate



I have to be careful and diplomatic here, because the team isn’t in agreement on what needs to be done just yet, and I don’t want to use this blog as a bully pulpit to sell my vision of the game and discredit theirs. There’s a lot of debate on how much of a tutorial we need, what form it should take, how long it should be, what concepts it should cover, and what coding changes we should be willing to apply to make it happen.

Sure, as a senior member of the team I could probably throw a tantrum and get my way, but bullshit like that is how you end up with Kai Leng. Assuming you’ve got a good team of smart people, then you should be worried about any idea you can’t sell to them. Maybe they’re just too dim to see the grandeur of your vision, but maybe there’s something wrong with your idea and you need to discover that flaw through debate. To put it another way, if the smart people on your team aren’t buying your idea, how will that idea fare with players? You planning on arguing with them, too?

All of this sort of sums up my creative / managerial / philosophical outlook: Hire smart people and trust them to know their jobs.

But here is my thinking on the matter anyway:

Building Complexity

If you watch very closely, you`ll notice the game seems to involve some sort of red thing that maybe does something?

If you watch very closely, you`ll notice the game seems to involve some sort of red thing that maybe does something?

We’ve been taking things for granted, but there’s actually a good bit of complexity in our game. This is good, because it means there’s a lot to keep people engaged. On the other hand, we need to be careful about how much of that complexity we deliver at once. Here are what I consider to be the core combat concepts of the game, from the simple to the challenging:

The Fundamentals: The player can fly across the room and hit a stationary target. The game looks like a twin-stick shooter, inasmuch as you have two sticks for aiming and moving. (Mouse and keyboard also works, and is how we do most of our testing.) But you don’t just press left on the stick to shoot left, because your weapons fire at different speeds and have cooldowns. You wouldn’t want to continuously fire the Half-Life shotgun as you walk forward, after all. It’s a weapon that requires timing. A lot of weapons in our game – like the “shotgun” style stuff – are designed so that you want to fly right up into a robot and nail it at point-blank. That feels really fun, but it’s incompatible with the spammy twin-stick style of gameplay.

So if the player is new to twin stick shooters, they need a minute or so to get used to the idea. If they’re familiar with the genre, they need about the same interval of time to see how this game might diverge from their expectations.

Basic dodging: The player can evade incoming bullets. Pretty straightforward. The Good Robot actually has a tiny bit of heft. It takes a fraction of a second to accelerate and change direction. I think we just need to give the player a little bit of time to get used to their movement speed and inertia.

Hitting fast-moving targets: The player can properly lead a target. This will vary from weapon to weapon, so every new weapon will have a bit of a learning curve associated with it. That’s fine, but the player needs to get some practice shooting at speedy targets before we dump other challenges on them.

Melee attackers: The player needs to spot and prioritize melee attackers before they can close the distance. Melee attackers are fast and do high damage. Worse, they knock you around, which tends to be disorienting. The player needs to learn to spot melee foes. Originally I made melee foes out of saw-blades to telegraph their nature, but I’ve lost track of our visual language since handing the art off to those more qualified. I spend most of my testing time on one dev level with my own testing robots. I don’t know what (if anything) we’re using to convey the idea of melee foes.

Missiles can be shot down and evaded: This is a tricky one. If the player acts on instinct alone, they might simply try to back away from missiles. This won’t work, since missiles are faster than the Good Robot. Missiles have a terrible turning radius, so you actually need to move perpendicular to their flight path to avoid them.

I’m not sure how to “teach” this. I’m very much against the idea of popups that just spell things out for the player. It’s ugly and often feels patronizing. Players enjoy figuring things out for themselves. Valve is actually really good at constructing scenarios where they funnel you into a course of action but trick you into thinking it was your idea. That kind of balance takes a lot of time and iteration. We don’t have the resources to finesse scenarios like that, but I’m not eager to give up and just tell the player what to do with popups.

You okay buddy?

You okay buddy?

The above image is a great illustration of Valve’s ‘invisible tutorial’ technique. The table is there to force you to get out the gravity gun. Now that you’re holding the gravity gun, it’s probably more expedient to use it to yank the saw blades out of the wall, as opposed to crawling under them. Once you pull the saw blade loose, a zombie will come around the corner and you’re very likely going to reflexively fling the blade at them. The half-zombie on the right is there to give you a subliminal shove in the right direction. And when you discover the usefulness of saw blades, you’ll feel like you figured it out yourself instead of just doing what some obnoxious NPC told you to do.

Luckily, we don’t have any mechanics quite this complex to present, but “How do we teach them to properly dodge missiles?” is still an open question.

Suicide bombers: The player needs to spot and prioritize bombers. I never did figure this one out. How can you telegraph that a robot is going to blow up? I mean, you can just make bombers look like another random robot and let the player learn to spot one through brute force, but I don’t think that’s really satisfying. Or at least, it’s better if we can develop a coherent visual language for the game. Should they be red? Blinking lights? A warning sound seems obvious, but we already have a warning beep when missiles are homing in on you, and the two sounds together would just make for cacophony and confusion. Although you could solve it by simply never having homing missiles and bombers on the same level. But that seems kind of limiting, in the long term.

Mine layers: The player needs to recognize and deal with mine layers. Mine robots dash in, drop some mines, and then sprint away. The mines they leave behind will limit your ability to move freely. It’s a pretty good mechanic. Ross actually came up with this. It wasn’t something I specifically designed the game to do. He just figured out how to make it work within the existing systems. (The “mines” are really just suicide bomber robots with their movement speed set to zero. Although now that I think of it, a really slow speed might be better, to make them slowly creep towards you and force you to deal with them.) These were interesting when they were first introduced, but they get lost in the chaos now.

Link (YouTube)

There are other concepts that the player needs to learn about: Their secondary weapon, the vending machine upgrades, buying a warranty. But these aren’t really combat concepts, so I’m sort of setting them aside for now. These are things that the player can learn somewhere outside the pressure of combat. And I actually wouldn’t mind tooltips for these, although obviously having the player discover these things organically would be even better.

My thinking is that we should adopt a Valve-style approach to teaching concepts. Let the player face A. Once we’re sure they’re comfortable with A, then set A aside and let them face B in isolationBy “isolation” I don’t mean B is the ONLY thing on the screen. I mean B is the only COMPLEX thing on the screen. There can still be simple familiar trash mods around for the player to mop up without worrying about overwhelming them.. Once they’re competent at dealing with B, we can let them face A and B together. Then take A and B out of play again and give them C. And so on. Late-period Mario levels work in a similar way.

That sounds simple enough, but it’s a little more problematic than that because the game is a bit like Spelunky. There’s not a fixed progression of maps but instead a pool of possible encounter types that the player will draw from, and this is driven both by player choice and a little bit of dice rolling. It’s easy to say, “Make the player master A before they face B”, but it’s not quite that simple when you don’t have direct control over when they encounter B for the first time. And imposing strict linearity is probably a terrible idea for a game designed specifically for repeated play-throughs.

Complexity Can be Good!

Nobody else on the team supports my idea of just removing all the guns and enemy robots and adding more purple lights. Savages.

Nobody else on the team supports my idea of just removing all the guns and enemy robots and adding more purple lights. Savages.

The problem isn’t that there’s no solution to the need for clarity and teaching. The problem is that there are a dozen ways to approach this and we need to nail down the best one as quickly as possible.

On the upside, I think we’ve really solved the blandness problem that made me abandon the project a year ago. The new ideas that everyone else brought to the game have given it the kick it needed. And I think it all came together in a short time, all things considered. But now we need to fit these ideas together so the player can make sense of them.

We have a little under a month to solve this. We want the art assets finalized before the end of year, which would just leave Arvind and I on polish and bug-fixing duty until the game goes “gold”.

Fingers crossed.

Enjoyed this post? Please share!


[1] By “isolation” I don’t mean B is the ONLY thing on the screen. I mean B is the only COMPLEX thing on the screen. There can still be simple familiar trash mods around for the player to mop up without worrying about overwhelming them.

A Hundred!5105 comments. Quick! Add another to see if this message changes!

From the Archives:

  1. Daemian Lucifer says:

    Learn from diablo.All of the diablos have a bunch of enemies that appear in batches.For example,one level of tombs in diablo 2 can have any of the 6 types of enemies,but only 4 of those appear in a single run(and usually its 2 types for the first part and the other 2 for the second part=.

    • Lanthanide says:

      That’s really only true of Diablo 1. In Diablo 2, the concept exists a little bit, but not really. Most levels will have mostly the same monster types each game. Act 5 has completely static monsters.

      Diablo 3 has completely static monsters (except for rifts).

      Btw, the way it was done in Diablo 1 is that each enemy had a weighting, and a level range. There were 4 variates of each base monster type, and there was overlap between them so multiple varieties of the same monster could appear on the same level, often overlap with 3 of them and very occasionally 4. So for the goatmen, you could get Flesh Clan on lvl 3 to 5, Stone clan on lvl 5 to 7, Fire clan on lvl 5 to 8 and Night clan on lvl 7 to 11.

      Each level had a weight cap, and the game would randomly select monsters from the list of eligible monsters for that level, subtracting that monster type’s weight from the level cap. So most of the time you might get 1 type of skeleton, 1 type of melee goatman and 1 type of goatman archer. But occasionally you might be 2 types of skeleton and 1 goatman archer. Or 2 melee goatmen and a goatman archer.

      It meant that you could never get all types of enemies on a single level, and exactly which ones you got on that level would be random. But by careful adjustments of the weights of enemies, it would be possible to craft certain ‘random’ outcomes. There’s a type of monster in Diablo 1 called the Hidden who are invisible until they get very close to your character. These guys are usually found in the catacombs. It’s possible (but quite rare, something like 1 in 100 or so) to get a level spawn that has 3 different varieties of the Hidden monster and nothing else. Because the Hidden are quite different from all of the others in the game, it ends up being a very memorable experience to get a level that is full of them to the exclusion of all else. There are similar possible combinations in the Hell levels as well, where you can get 3x succubus class enemies, or 3x magistrate class enemies. There are also some other ‘old favourite’ combinations, like Balrogs + Soul Burners and Blood Knights + Advocates.

      Anyway, if you took an approach like this, you could have a bunch of ‘tutorial’ style enemies that only occur in the first 3-4 levels of the game, and adjust the weightings so that those enemies will always turn up on at least 1 of the first 4 levels of the game, mixed in with ‘real’ enemies.

      • James says:

        I can only speak from Diablo 3 examples, but yea monsters are theme’d around the level (except rifts ofc), but there is still a great deal of variety in each theme so that it doesn’t become monotonous or dull.

  2. Daemian Lucifer says:

    The missile tutorial:Have a boss that has at least one stage where they fire a volley of relatively slow,but tough(so that you cant shoot them all),missiles that you cant run away from(constrained by the boss arena),so you have to dodge them.After that,have faster missiles.

    • NoneCallMeTim says:

      This seems like a good start, given that you have essentially two inputs in this game: moving and shooting.

      If you present a single tough enemy of the seeker missile type in a reasonably sized area, where there is plenty of room to move, the missiles will go after the player, who will only have a few options – move or shoot. Or both.

      Depending on how the rest of the game is set up, if the player goes down a particular path where they meet a missile robot for the first time, it could be set up that before the main bulk of the bots, there is a single room where this tutorial section is.

      Would it be cheesy to have the enemy literally say “Bet you can’t dodge these?” the first time you meet them. Or no dialogue from the robots?

      • Trix2000 says:

        Actually, I could see an enemy literally saying “Dodge this!” and firing missiles if they’ve spent a while trying to hit an evasive player with slow-moving lasers or something. It both serves as an interesting ‘reaction’ by the enemy to the player’s input as well as an implicit challenge of “hey, can you keep dodging even these?”

        Having it be a LOT of low-damage missiles in a small spread might be good too, such that a player who relies only on shooting them down would be hard-pressed to get them all in time, but dodging would easily bypass the lot. Reducing their turning radius may or may not also help in this situation to sell the idea.

        The only other idea I had was to construct an environment where the player rounds a corner and is suddenly confronted by incoming missile(s), with the only path they can run to (in reaction) being perpendicular. This might be trickier to construct though, and I’m not sure it would work in the game’s current implementation.

      • Jakale says:

        My first thought was have them go down a vertical corridor and get missiles shot at them, maybe holes in the corridor walls where missile bots are hanging out, but that may not be an option depending on how much random generation is going on.

        You could also get a bit of subliminal messaging going on with your start menu, like Megaman X did.
        Example: give the menu options vending machine/reconstruction station graphics that a good robot sprite uses to select that option. Then people may be more inclined to think of those as potentially helpful when encountered in-game.

        • Nimas says:

          On this note, I honestly think looking at how the original Megaman did this could be very helpful. This (although you may well have seen it) has a good run down of some things it does.

    • Phill says:

      My suggestion for the missile tutorial would be to set up an area with nothing much in it, so the player speeds along quite fast. Then have them ambushed with a missile coming in from the side, at relatively close range. If they do nothing but keep moving, the missile curves past and misses them and hits the scenery (not enough room to turn).

      Follow it up with a second missile coming from straight ahead. At the combined closing speed, I’d imagine most players would try to dodge to the side (above/below) rather than stopping and reversing. And again, give them enough room to dodge but not enough for the missile to turn around and chase them, so they can see the poor maneouverability and the missile whacking harmlessly into the walls.

      Far from guaranteed of course – some people will manage to die, and some will not get what you are trying to teach, but that’s going to happen with *some* people no matter what you do.

      (And maybe you can then follow up with a large-ish room with a single missile-launching enemy that ups the challenge to more than one missile, and a moving launch platform)

    • Abnaxis says:

      How about starting off with missiles that pretty much dodge themselves, i.e. that an enemy deliberately fires wide so that the projectiles can only hit players on the return path? Like, make the enemy tough/fast, only have them shoot when they’re within X units of the player, and have the initial velocity be like 30° offset from the vector connecting the player to the enemy. Is that possible/easy with the ini file format you have?

      • AileTheAlien says:

        I was about to suggest the same thing; An enemy that fires off-aim from the player would nicely show off the turning radius of the missiles.

      • DjordjBernardChaw says:

        If they fire two at a time in an offset, this might also keep it from looking goofy.

        “Of course they fire at an offset, they couldn’t fit them on the robot otherwise because they’re too bulky.”

        Or something like that.

  3. boz says:

    How can you telegraph that a robot is going to blow up?

    In Rainbow 6 Siege PvE modes, you have regular white masks and bombers as AI enemies. Bombers wear white armor, make distinctive sounds and have a bluish, highly visible, flashlight (and they still kill a lot of people).

    • Phill says:

      Other people have suggested a flashing red light on the robot, which is one option. Another would be to have the whole robot flash (red, white, yellow, whatever). I guess you have to find out what stands out clearest from the visual noise so the player finds it hard to miss.

      Have the flashing/pulsing happen more and more rapidly as the time to detonation runs out (with audio cues also pulsing more and more rapidly at the same time). I think 95% of players will assume that something bad is about to happen, plus the increasing frequency of visual/audio pulses adds to the sense of tension. Hopefully.

      • Lazlo says:

        Something that I’ve seen that I believe is relatively new in WoW is their graphic for time bombs. It’s intuitive enough that it was pretty obvious the first time I saw it: a black circle that starts with a yellow radius line at 12:00 and starts sweeping clockwise, leaving an area of non-black that changes from yellow to orange to red as the line finishes its full circle and becomes all ‘splodey. I don’t know if the bombers in Good Robot are purely time-based, or proximity based, or a combination of the two, but I think that graphic could accommodate either one. A visual shake as it’s nearing explosion is also a good indicator.

        • AileTheAlien says:

          The visual shake is another good trick. In all of the Mario games, Bob-ombs shake when they’re about to explode. Conveys nicely that they’re about to do something drastic, and that they’re crazy / suicide-bombers.

          • Alexander The 1st says:

            Bob-ombs also usually have a “Starting explosion” animation though.

            That said, using pulsing red lights on the suicide bombing enemies seems…like a bad idea in a game with a *lot* of other flashing red lights.

            As such, I’d recommend going the opposite direction – as they get close to exploding, have them dim out to black. If their eyes go, then their central engines, then their arms…I feel that might stand out more.

            In a sense, taking the lack of light to be a greater signifier in a game where everything *usually* has a light.

            I’m reminded of hearing about how cops used to do voiced over advertisements on TV about the dangers of drug growing, but then they realized that it would not get heard in a grow op because of all the lights and fans and such…so they switched to silent ads that just displayed an image – because the lack of the TV making background noise stands out so much more.

            • PDK1359 (number inconsistently used) says:

              I like the idea of going with silence rather than more noise.

              Here’s my take on the idea; have an enemy that’s mostly a solid color, with a couple lights on it, the lights die if the player gets too close or the timer runs out and then as an oh-shit warning, a circle of light starts rapidly expanding from the center-of-mass or the bomb sector of the bot, followed shortly by the actual damaging shockwave.

              Question? Can you stagger the damage of explosions? Say N-range is high damage, Nx2-range is less damage & Nx4 range is slight damage? Because that could go a long way to helping a player recognize explosive threats without actually killing them.

              Another idea; not all explosives necessarily have the same twitchiness. Some bombs are nervous, go off early, others are disciplined go off in your face or not at all. Maybe even a class of bombs that go off if you shoot (perhaps the right weapon?) near them, because they get surprised/scared, but can be used to set of chain reactions.

    • James says:

      ive seen some streams and at times the bombers seem to have supernatural levels of awarness to find players and just bumrush them.

    • Peter H. Coffin says:

      A common trope is a flashing light, that increases frequency as time to detonation approaches.

    • Supahewok says:

      My own suggestion would be to incorporate the nuke symbol. There isn’t a person on Steam who doesn’t immediately know that that means both “danger” and “explosion”; even ol’ Grammy playing The Cherry Blossom Murders knows what the nuke symbol means.

  4. LazerFX says:

    Re: Missiles:

    I’d say the easiest way to teach them to dodge missiles is simply give them missiles, and have the opponents dodge. Then, when they’re used to that, have the opponents fire missiles. They’ll already be cued to dodging them, and boom.

    • Syal says:

      My suggestion is similar: give the player a missile weapon early on, have enemies run directly away from them and be unable to avoid them (maybe they drift a little so you can see how slowly the missiles turn), and then introduce an enemy that shoots them down and runs away. Then players can figure out, “okay, I can shoot them down, just backing up doesn’t work, does anything else?” without making enemies immune to missiles.

  5. Da Mage says:

    I mean, in any game with a suicide enemy I’ve seen, they have a large red blinking light on them as they move toward you.

    Even in Fallout 4 the super mutant suiciders have a large red blinking light on the bomb they are holding. Makes them easy to spot in a fight so you know to make them a priority. It’s a clear design choice as the normal mini-nuke does not have that light.

    • DanMan says:

      Agreed. Also, the art style already seems like bad roboids have glowing eyes. Could easily just make the eyes flash if they’re suicide. If it’s fast enough, it should look like it’s intentional and not a glitch and players should learn it fast enough.

    • NoneCallMeTim says:

      Indeed. Blinking red lights solve all problems. Maybe blink faster as they get close to the player?

    • Trix2000 says:

      I think the sound for those was also critical, because that’s always been the first indicator to me of “Oh crap, WHERE IS IT?” The light doesn’t help so much if you’re not facing the enemy.

  6. arron says:

    How can you telegraph that a robot is going to blow up?

    If you’ve got a lot of noise, I’d make it mostly visual. The usual sound for “bomb about to explode” is a beeping noise, which you might not hear over everything else.

    What I would do is to have something on the robot blinking a red colour. And either some electric lightning or have the robot shaking if it is static, like the explosion is not going to be contained. For a twin stick shooter, you need to see these enemies fast.

    It’s a cliche I know, but it’s immediately recognisable by instinct to the player. Unlike the service robots in System Shock II that were suicide bombers.

    Their requests for your current location as they walked towards you used to scare me when on the cargo storage level.

    • Syal says:

      Or maybe they just glow brighter and brighter the longer they’re on screen.

      • Matt Downie says:

        Do they explode on proximity to you, or after a set amount of time? The visual feedback should be different in either case.

      • Jonathan MacAlpine says:

        Mr. Gutsy’s (Mr. Gutsies?) in Fallout 4 do this, and it works really well. It’s associated with a sort of rising-in-pitch crescendo humming — their reactor overloading. They glow red, then yellow, then white, then… BOOM!

        Especially given how silhouetted everything is in Good Robot, I think an effect like this would communicate bombers well.

        • Cuthalion says:

          I like the “rising-in-pitch” suggestion. Didn’t The Dark Knight do that with the Joker? The note gets slowly louder and higher and higher pitched as it builds?

          That would let you do something other than rapid beeping so that you could save that for missiles. Plus, it might be just as great at building tension.

      • RedJack says:

        Perhaps the glow could have a cycle which gets faster the closer it is to going off? I picture a stopwatch, or better still a red radar-type sweeping hand that spins quicker and quicker as a sound tone becomes a high-pitched shriek. Maybe a final rapid-fire beep beep beep BOOM! to finish it all off?

        The timer imagery coupled with the urgency of the sound might sell the idea of “get outta the way!” and tie together pretty well with the established look.

        It’s much clearer in my noggin than I’m able to explain here. Sorry.

  7. Leonardo Herrera says:

    Question: is the actual game this dark? Can you play with color saturation? (background, enemies, player, obstacles?)

    Well, some ideas that others have already mentioned: introduce new enemies in batches, never have more than X types of enemies on screen at once, some effects could be subdued (the trail from the Good Robot could be less visible, it tends to blend too much with some weapons), and blinking red lights to indicate danger (at different blink rate for missiles, mines, and suicide bombers.)

  8. Groboclown says:

    How about soething akin to what the old arcade machines did – have a rolling demo of prerecorded game play that shows off the elements you want to teach. Doom did this too, and it helped to teach the player what was a pickup, a door, etc before there were the fps conventions for these things.

    • TMC_Sherpa says:

      I was going to be snarky and suggest loading screen tips (because everyone *loves* those) but a looping intro is way better. It even solves the problem of having the starting levels being too easy after you’ve played a few times.

      • DjordjBernardChaw says:

        The loading screens sound too short to have tips. I think there was a post about how the loading screens had crept up to 10 or 20 seconds, and it was a potential barrier to testing.

        I know if I have a game on an SSD it’s not difficult to miss loading tips, even on games with otherwise significant loading times. It’s generally not a good idea to have parts of your game rely on having or not having specific hardware setups.

        • Shamus says:

          For the record, the 10 seconds was the loading screen for the whole application. I think I shaved it down to five.

          The between-level loading screens are only a few milliseconds. No time for tooltips. :)

          • TMC_Sherpa says:

            A few milliseconds is plenty. Especially if you start them with something like “The map is important and we are really sorry we forgot to include it in the keybindings or the tutorial. Boy do we have egg on our face! To cut a long story short press…” Thanks for loading the new level before I can read the drivel game, really appreciate it.

            Anyway it doesn’t matter because there will be a printable paper manual. Right? A PDF? Something? Anything? I’ve been dorking around with one for Rogue System and it’s kinda fun[1]

            [1] For various definitions of the word fun

    • AileTheAlien says:

      One refinement I’ve seen, that really seems to help, is to have the rolling demo under / behind the game menu, before the player starts the game. Like, in the old games, the demo would only play if you sat on the main menu without going into a game, but that’s annoying. So, as the player is messing around on the menu, deciding if they want to start a normal game, survive-as-many-waves-as-you-can game, or is fiddling with the audio sliders, they’ve got a demo playing in the background, so they don’t have to wait for the demo to start. Also, a demo/show-me menu, where they can explicitly see some demo reels in full-screen / not-the-background would be cool.

  9. Squirly says:

    For what it’s worth, the first thing I thought when you mentioned differentiating suicide bombers from the rest was “red blinking lights”. Maybe as a dome-like hat or something. Keep the missile lock-on sound for the missiles, make the suicide bombers stand out on the screen.

  10. Diego says:

    Hah, the missile tutorial is easy! A long corridor with a missile launcher at the end and one corridor at 90 deg (so like a T). The missile gets launched when the player is past the intersection so he has to run from the missile and notice it can’t outrun it. Luckily, there’s an intersection right when he thought he was gonna get blown to pieces.

    OK, probably easier said than done.

    • DjordjBernardChaw says:

      If the missiles explode upon colliding with terrain, this could also introduce one of the ways to nullify them to players.

      At the same time, it sounds like the game is almost entirely procedurally generated, so this tutorial may be ideal for technical, or replayability reasons.

    • Paul Spooner says:

      Yeah, that’s basically how Mario teaches you how to dodge bullets. A long corridor with alcoves on the sides. Too long to traverse without getting shot. The alcoves teach you to get out of the way. Then you just widen the corridor, so there are no alcoves, and suddenly the player is dodging!

  11. The Rocketeer says:

    “It made sense at the time, but when the feedback rolled in it was a forehead-slapping moment for all of us. Of course this will seem like random chaos to someone who hasn’t played the game constantly for 6 months.

    I’m pretty sure this is what happened with Black Mesa, the Half-Life remake.

    The people making it played the original for hundreds of hours, and then played their own remake in the Source engine for hundreds more hours while testing, and thought, “Well, the Hard difficulty setting is about my speed, but then I’m probably pretty good by now so the balance is probably about right.” Of course, in reality, the game is like sticking your hand in a fucking vise.

    Or maybe that’s just what they were going for. But that seems to have become a sort of hallmark of enthusiast projects that receive little external feedback.

    • It’s a pretty common human thing, really. If you either don’t remember how hard x was at first or x was never hard for you, it can be very difficult to see that it might be hard for someone else. I had an elementary school teacher tell me once that “You don’t remember how hard it was to learn to read, but I promise you it was. Learning to write is just as hard, but you’ll get it, just like you did with reading.”

      The major reason I’ve never tried to teach in a classroom setting (though I love teaching stuff to others, likely because I love learning new information) is because while I’m quite good at empathy and understanding other people, I’m not very good at explaining or presenting things in a different way. I still remember my short tenure as a physics 101 tutor and the hour-long discussion/argument about assuming gravity is down. It’s a common conceit in basic mechanics problems that unless you’re told otherwise, gravity is down. I couldn’t seem to get her to understand that, that yes, in real life this may not be true, but basic mechanics uses a simplified universe. Also, at the time, I hadn’t consciously realized all that. I got told gravity is almost always down, I accepted it as a RULE, and couldn’t understand why she couldn’t just do the same.

      • Paul Spooner says:

        It’s a really great moment though, when you see someone realize that it could be any way, and not just what they are familiar with. When I was tutoring, I would just rotate the piece of paper. It’s a reference frame! It can be any orientation you like!

  12. Diego says:

    Also, good that the blandness problem is on its way to be hammered :) The complexity seems to be adding a lot of depth to the game.

    This is plaguing my project for a while too and, just like you, is keeping me from working on the project until I figure it out.

  13. I think you may have mentioned a good solution in passing there Shamus; Spelunky!

    Certainly Spelunky is a randomised game designed for repeat playthroughs but it also features a tutorial introduction through a fixed set of stages that teach the basic mechanics of the game. You could consider implementing the same here, where the first time a player hits “Play” they are sent to a fixed, pre-generated tutorial level that guides them through movement, missile dodging, basic enemy types, etc. After they complete (or fail) the tutorial session add an option to the main menu called “Tutorial” that lets them play it again and set the “Play” button to start a new, fully randomised game.

    This should give those familiar with twin-stick shooters enough of a soft introduction that they can pick up the differences from what they already know whilst also providing those newer to the genre a suitable opportunity to learn the ropes.

    • krellen says:

      This was also going to be my suggestion, though as I have not played Spelunky, I would not have suggested that clever means of “hiding” the tutorial. I was instead going to suggest a “story mode” linear/tutorial section and an “endless run” base gameplay section.

    • Dan Efran says:

      I don’t think a mutating main menu makes much sense. How would you know it’s the same player coming back with more experience, rather than her friend sitting down to play for the first time?

      I also don’t like pop-up hints: they smell like an admission of failure from the game designer.

      Even if the game is randomized, why not just keep track of which combat styles the player has encountered and survived? Hold back on new challenges until they demonstrate a bit of mastery.

      Like, maybe they’ll encounter missiles before suicide bombers, or maybe vice versa, but they definitely won’t see both at once until they’ve seen each alone. And if they see missiles first, they’ll keep getting mostly missiles until they survive a missile attack successfully (like until they shoot all the missiles or dodge them for 30 seconds).

      Tracking this information and using it to influence the random level design would complicate the code, but (I think) not unreasonably so. It’s kind of like tracking ‘achievements’.

    • DanMan says:

      I like the idea of the ever-present tutorial. No matter how good your mechanics introductions are, some people just simply aren’t going to get it and will be frustrated. Some people actually PREFER to be told what to do and how to do it.

      However, I am personally against having a mandatory “easy mode” tutorial. For me, a lot of the fun of these kinds of games are discovering the mechanics. Maybe I spend 5 hours in the game and I always dodge the missiles and on the next play through, I discover I can shoot them out of the air. That feels like a victory to me and gets me excited for the other little details that maybe I’ve missed. It keeps me coming back.

      Whereas a tutorial that teaches the mechanics, now I just have to put what I’ve been told together. Feels like school again.

      • Syal says:

        I was just reminded of the challenge rooms from Bastion and Transistor. Those could work really well for teaching certain skills along the way. A roomful of missiles with no weapons and a ‘destroy all enemies” condition could teach how to dodge and how to run missiles into walls, or each other.

      • Maybe a quick popup that offers the tutorial but doesn’t require it? First time you hit Play it pops up and asks something like “Hey, want a quick overview of how this works or would you rather just jump in?” And after that have the tutorial as a menu option.

      • Harold says:

        One of the things Shamus wanted to do was have conveyance, and the reason that’s almost impossible to set up is because you can’t really control when exactly different enemies would spawn. A non-verbal tutorial level would be perfect.

  14. Zoe M. says:

    Make a set of one-off training encounters. (EG a single, high-health missile-firing robot). Substitute them in for the regular encounters, but ONLY the first time you encounter them. Kinda like the inverse of a mastery-test boss.

    • Kylroy says:

      Or a standalone skippable tutorial that runs through each mechanic individually? Though that has the problem of a player being shown a mechanic long before they encounter it. Maybe introduce two mechanics each level, and have a skippable tutorials for both new mechanics at the start of each level?

      • Akri says:

        The way Crypt of the Necrodancer does it could work, where the practice sections are accessed separately from the main game. These rooms can be entirely scripted, making it easier to guide the player toward the correct answer without needing to outright spell it out for them. And because they’re separate from the main game, only players who need/want the practice will ever have to deal with them.

        You could even make it so that these rooms only unlock once you’ve encountered the relevant challenge in-game, if you want players to try and deal with enemies “in the wild” rather than mastering all their skills in the practice sections first.

    • I think only once is a bad idea, really. To me that’s a good road to frustration. People learn at different speeds, and more than once I’ve missed an important bit because I’m still busy getting the last bit down.
      Practice rooms sound lovely, though. You could even have the player encounter X enemy or Y weapon or Z whatever in the wild and at the end of the level have a little screen of victory that mentions that you’ve now encountered X,Y, or Z for the first time, and thus a practice room has been unlocked in the tutorial menu so you can go fiddle with the new thing in isolation if you want.

  15. CrushU says:

    If there isn’t already, there should be a way to tell the game ‘Spawn enemies of supertype A’ Like, a set of enemies that are the basic trash mobs, with the game handling which specific types it’s spawning…? If that system’s in place, you can have it spawn successively ‘harder’ enemies, first by just introducing more types, then harder versions of previously mastered types.

  16. Fizban says:

    I’ve been playing a lot of Binding of Issac the last couple days and it feels like that game is full of this problem (I still just wing it with colored foes, at least half the time my guess on their effect is wrong anyway). I’m pretty sure they combat it by unlocking enemies and rooms as you go, the “Everything is Terrible!” and such, even before the new dlc I’d still occasionally get rooms I hadn’t seen before. It still only goes so far, I didn’t even consider trying the game until I’d watched at least half a dozen runs on youtube and that was back on the original. Trying to set up gradually unlocking enemy pools might be more trouble than the benefit would be worth. There’s also the challenge modes, but those aren’t very teachy and making teachy “challenges” would just seem disingenuous.

    I think a tutorial would be fine. I like the example gameplay footage idea, but it’s easy to miss if the player just doesn’t pay attention. The tutorial doesn’t have to use text popups (though it might if you fail), just stuff like the missile dodging side-corridor and the basic enemy types and such. Snarky one-liners could also be good, though I’d be careful on the wording lest people do the opposite of what’s intended.

  17. Cerapa says:

    I don’t think I have ever played a game where I have enjoyed fighting against suicide bombers. They tend to be highly dangerous, but killing them is unrewarding compared to the danger. Either they just die, like any normal enemy, or they explode, making it so you can accidentally blow yourself up. This is especially bad when you have a shotgun style weapon, which means you have to be close range to an exploding enemy to kill it.

    • Syal says:

      Make them explode backward in line with the weapon trajectory that killed them. Then they’re walking crowd control weapons.

      • Blue_Pie_Ninja says:

        This idea would be even better if weapons with more stopping power blew the enemy backwards more, or made the explosion backwards more powerful.

      • Cuthalion says:

        I’ll echo this: suicide bombing enemies feel less like an annoying hindrance and more like a fun adversary if blowing them up… blows them up. Allowing you to use them strategically if you’re smart about it.

        Combine this with Blue_Pie_Ninja’s suggestion, and you might even find a way to make short-range and melee weapons viable against bombers, which was one of Cerapa’s objections.

  18. RedJack says:

    I’m echoing the idea for the fixed set of training levels mentioned above, but maybe you could bury them in the narrative a little by making them the only “mapped” areas of the game (in other words, these beginner levels were the only areas that the little robot was allowed knowledge of/access to before running out the door to tackle his/her adventure).

    This would allow G.R. to have existing knowledge of the first bit of the game–which you could make as complete or decayed as was needed–and still retain the openness of the later stages.

    For what it’s worth, those are my two cents. Hope everyone is having a good one.

  19. You could have a more constrained tutorial type level or levels that go the first time the user plays your game. I would also ensure that in general, complexity goes up as the player progresses. Only one or two complex enemies can spawn in the first level or two, for example.

    As to specifics, for missles have an enemy that fires two missiles perpendicular from the player, so they see the slow curve and just fall into dodging if they keep moving around. If you wanted a scripted sequence to teach, have the player enter a room from the left, with a powerful on the bottom. As soon as they start moving for the power up, an enemy enters from the right and fires a missile.

    For explody robots, I would make the whole robot flash red, flashing faster the closer it is to exploding. You could also add some smoke or steam particles.

  20. Drew says:

    In light of the fact that the player character is a robot in this game, it could be interesting to introduce some kind of terminator-style view and breakdown of new enemies when you first encounter them. For example, the game could quickly pause, zoom in on the new enemy, and highlight its various features. If each capability is represented graphically, it could even highlight each part as it described them. Highlight the missile launcher tube and indicate “launches homing missiles”. Highlight the afterburners and indicate “moves quickly”. Et cetera. If there’s no physical indicator of capabilities, simply listing them out could be useful. The idea is that the Good Robot would be capable of scanning the enemies and determining their capabilities, and you, the player, could get that information. Then you return to gameplay.

    The issues:

    1) Does this break the flow of gameplay too much? I guess that depends on how many new enemies you encounter over the course of a game, and how quickly you encounter them.
    2) Is part of the fun of the game NOT knowing what each enemy does and figuring it out from watching and learning? If so, this isn’t a good idea.
    3) Can the game design allow for spacing out the introduction of new enemies to prevent overdoing it with this information?
    4) Is this kind of zoom-pause-list too much development effort at this stage of the game?
    5) Would it be super annoying to have to look at these descriptions every time you play through the game, or would it remain fun and useful?

    Another bonus is that this could also give the game’s creative team an opportunity to provide some more personality when writing the scanned descriptions of each feature, and also provide a name for each encountered enemy. Any opportunity to add flavor is a good one in my book, as long as it’s not cumbersome and annoying.

    • Lachlan the Mad says:

      I really, really like this idea.

    • Goblin says:

      Maybe you could start the player with 1 (or 3 or 5) scans and allow them to buy more if they want them?

    • Blue_Pie_Ninja says:

      I guess this could be in a menu so that once you have encountered a new enemy had have versed it for a bit, a notification comes up that Good Robot has scanned the new enemy type and added it to a database. This database could have all the known information on that specific enemy type and it’s features.

      Also that could also tie into the idea of a practice room which acts as a simulation of fighting with that new enemy that was encountered.

  21. Van Tuber says:

    Elegant subtle tutorials are nice and all, but no one is going to hate your game if you have a tutorial mode with pop-ups.

  22. Raygereio says:

    How can you telegraph that a robot is going to blow up?

    Having a blinking light and/or beeping sound with the interval between blinks/beeps decreasing as the timer goes down is the common way of saying “This thing is going to blow up”.
    But you’re going to get people who won’t get that.

    A demonstration would, I think, be the obvious answer. For example when it’s first introduced, have the Suicide Bomber blow itself up to clear some rubble allowing its buddies to get to you.

  23. ThaneofFife says:

    The majority of people seem to be advocating that suicide bomb robots flash red. While that would work for me, I foresee two potential problems with that approach:

    (1) Will people who are red-green colorblind be able to differentiate the suicide bomb robots from all of the other colors on the screen? If not, then you may want to think about implementing a color-blind mode.

    (2) How frequently are objects on the screen flashing? Rapid screen flashing can trigger seizures in a substantial minority of people. They can also give other people (including myself) headaches. I would recommend either limiting the amount of flashing on the screen, or, failing that, putting a seizure warning in the product description.

    On tutorials, I like the idea of having a fixed tutorial level, as several people suggested above. Just make sure that you give people an achievement for completing it. ;-)

    • Syal says:

      Call it ‘pity cheevo’.

      • To quote Mumbles, “Please do not.” I speak only for myself here, but while that might be intended as gentle ribbing, to me it says “You suck for needing or even wanting help.” (and for emphasis, to me, at the moment. I very much doubt you meant it in any sort of hurtful way and I am very likely extremely sensitive at the moment as my brain has gotten the “Ponder deep thoughts nonstop” and “Sleep” functions confused again and two nights of no brain shutdown does not make for a good-thinking Melfina.)

        Sometimes it can be tremendous fun to figure out how things work. And sometimes you just want the manual. Right now, hand me the manual. I have way too much frustration in my real life right now, I have no desire to deal with frustration in a game, and my reaction would be, well, let’s just say extremely negative.

        • Syal says:

          I’m always in favor of including either insulting achievements, or achievements somewhere between sadistically hard and literally impossible. (“Rescue all 50 toads” when the game has 45, and such.) So ‘pity cheevo’ would be for completing the mandatory tutorial. Maybe not even completing it; maybe just starting it.

          I always use tutorials. I’m actually pretty bad at things that don’t spell stuff out. Shamus praises Half-Life 2’s soft tutorial style, but that stuff frustrated me a lot.

          • I never would have thought someone who uses tutorials would be in favor of the “joke/insult” easy achievements, so yay, learned something new! To me they generally seem to come from and appeal to the same sort of person who would mock you in gym because they could lift 100 pounds and you could only lift 10. It feels very much like someone hurting me to make themselves feel better, even though I know that’s not the conscious intent (and if it were coming from Shamus (or a commentor here) they’d get the benefit of the doubt). I know it’s a bit of a gaming tradition, but someone who doesn’t might not realize it’s intended as a joke and it strikes me as perhaps not the best idea to risk insulting your potential audience right off the bat if that’s not how you mean to go on.

            As far as sadistic achievements, yeah, go for it. I’m not going to bother, but I know there’s gamers who love ’em, so why not? I’d stay away from not possible achievements because that feels like a mistake, a bug, or a lie. Two of the three I’ll forgive, the lie I won’t, especially from a company. It’s a good way to lose my business.

            I’m not really a gamer anymore (more of a person who likes reading about them but doesn’t really seem to want to play ’em anymore), and I’m probably not in any part of Good Robot’s intended market (unless it’s because I’m a reader/commenter here). I don’t think I’ve ever played a two-stick shooter in my life. So I’m trying to give outlier marketing feedback, I guess. A voice that’s familiar with, but no longer part of the gaming echo-chamber as it were. Hopefully it helps!

            • Syal says:

              But then you paint the weights pink and white with rainbows and kittens, and you have a free beginning weightlifting class that starts with a quarter pound and increases in quarter pound increments, all the while telling you how outclassed you’re going to be by this next quarter-pound increment, and you bolt six or so of the heavy weights to the floor. That 100 pounder guy might still be there, but he’s going to look really stupid if he’s mocking people.

  24. Cybron says:

    Maybe it’s a little too simple but I’ve always been a fan of a seperate, non-immersive tutorial that you can pick off the main menu. Just a little how to play run section designed to teach basic concepts. This is how Dungeon Crawl: Stone Soup does it. It’s short, to the point, and should get the basic mechanics across in a controlled environment.

    Sure it doesn’t have that same “I figured it out!” appeal, but there should be plenty of that anyways if your procedural encounters impose emergent challenges for the player to solve. You don’t have to tell them exactly how to do well, just introduce them to the concepts and let them figure it out. To go back to DC:SS mentioned above, they show me I can fight, but they don’t explicitly tell me that I can use corridors as chokepoints to control how many dudes are attacking me.

  25. WWWebb says:

    Here is where I think the random levels cause the most problems. People have made lots of good suggestions about how to handle introducing a new mechanic. Personally, I like the “zoom and demonstrate” method (possibly because I’ve been playing a lot of Broforce), but
    1) Shamus generally complains when games take away camera control
    and 2) it would be really tricky to smoothly integrate into an otherwise random level.

    To do it that way, a level would need to have a static, tutorial section where new things introduced, and then flow into the random level. And of course, that assumes that you’re controlling WHICH things get introduced in what order and that only one new thing shows up in each level…even less randomness.

    Groboclown’s suggestion of an arcade-style demo reel is actually pretty good. It could even have story content if you made a (skippable) intro cutscene where Good Robot gets tormented by the other Bad Robots:
    -you watch a melee attacker hit him
    -you see a shooting attacker shoot him
    -you watch him get hit and then dodge a missile
    -you watch him run into a mine ( robot whoopie-cushions!)
    -you watch a mine come awake and chase GR (a bomber)
    -at some point GR dies and comes out of a warranty repair machine
    …but wait, now he has a gun! Time to do whatever it is GR is going through these levels for!

    That’s not interactive (it’s also a big ask for the art team in a month), but it’s the only think I can think of that keeps fully random playthroughs.

  26. Khizan says:

    Missiles: Make it so that the first time the player encounter missiles they’re launched by a ground-based silo or something and they launch vertically and then home in. This lets the player see the turning radius of the missiles before they have to deal with the missiles launched directly at them.

    • 4ier says:

      I was going to suggest stationary missile emplacements, too. But I would also have the first couple in a long hallway with a low ceiling, so sideways is the natural way to go, and the missiles can’t turn fast enough to avoid the ceiling. Maybe have the launchers angled towards the entry to the hallway so backing away is more quickly and obviously punished.
      After the first two or three missile emplacements, raise the ceiling up so the missiles can start homing in.

      • 4ier says:

        Another thought: if you have stationary launchers, they need to be able to have some aiming/tracking ability before they fire, so players don’t camp out just behind them and let them waste missiles firing in the wrong direction.
        Aiming speed probably needs some play-test fine-tuning, but it probably needs to be fast, so the game stays ‘avoid the missile’ rather than ‘avoid missile lock’.

        • Khizan says:

          Fixed launchers aren’t what kills you; fixed launchers are to box you in while other things kill you. If you’re hiding from the launchers you’re letting the launchers dictate your positionings, which means they’re being effective.

          • 4ier says:

            Eh, fair enough.
            I was thinking of the tutorial case, where there isn’t any other threat, so the player feels clever and then bored. I may have over-thought it. :)

  27. David W says:

    On the original subject: a Valve-style tutorial goes with a Valve-style game. Nearly no penalties for failure (maybe 10 seconds lost), a long linear game with probably just one playthrough. If you do come back to the game for a second playthrough – it’s almost certainly not immediate.

    That’s not how you’ve described Good Robot, or at least not how I understand it. You have instead permanent death penalties, with escalating cost. Most playthroughs will end short of the finish line, and the intended player behavior is to immediately hit the New Game button and jump right back in.

    You didn’t describe the opposing position in the debate, but if I had to pick an opposition to your suggestion, that’s what I’d base it on: escalating difficulty with valve-style tutorials mixed in will get really, really, annoying on the fifteenth playthrough. Especially if that fifteenth playthrough is the result of a new concept that you haven’t got the hang of, while the early parts are so easy that they’re tedious: that’s a recipe for DIAS gameplay. DIAS gameplay where you have to work your way through 20 minutes of easy stuff before you’re challenged again. DIAS gameplay where the easy stuff can still kill you if you’re bored and not paying attention, which means you have 30 minutes instead of 20.

    You don’t want restarting the game for another try to feel like a punishment for someone who’s learning the game but hasn’t quite gotten to the finish line. Which means that the early levels have to be enough of a challenge that the guy who just died 3/4 of the way through, can still enjoy himself when he restarts, instead of trying to race through back to the spot he died.

    I think this is why so many of the suggestions above are for a separate or skippable tutorial, or a series of tutorials – but all outside the framework of the ‘main game’. Because you can’t afford a drastic difficulty ramp during the game, if you want to avoid DIAS gameplay – but you still need an opportunity somewhere for people to learn the basic skills.

    • tmtvl says:

      When you say Valve-style tutorial, I think the training course from HL1, completely separate from the game and accessible from the main menu. That is, in my opinion the only way to do it, making anyone, even first time players, sit through a mandatory tutorial is really not the way to go.

  28. Cybron says:

    Another thought – have a basic difficulty (with few robot types of standardized appearance) then have advanced mode where you throw in the robotic, heavily armed kitchen sink.

  29. Goblin says:

    I don’t remember if you’ve described the lore for what the vending machines sell and why, but perhaps they could include a “Anti-bomber sim”, “anti-minelayer sim”, etc. which would provide an option to run a tutorial level and a bonus to damage against that type of enemy.

    Maybe start the player with one for free, and with a (skippable) explanation…?

  30. John Beltman says:

    You seem to be confusing your spaceship 2d scrolling shooter with a role-playing game.

    Did Galaga have a tutorial? 1942? Raiden? Pac Man? River Raid?

    Just let them get into it. If they die they can just start again. Given them a manual they can read from the main screen if they like. They will only look at it once anyway. Consider only giving them a subset of information in the manual so that there are still things about the game they can discover on their own. That’s half the fun.


  31. HiEv says:

    I don’t know if this fits with your theme, but perhaps, between levels, you could have Good Robot’s creator send him a message warning him about the new kinds of threats coming up for the next level. Perhaps you could even make it so that just before the final level it has his creator getting caught by the final boss, so in the end he gets to save/meet his creator.

    In any case, it would give you an excuse to give the players a quick heads-up about the new enemy types and give Good Robot a bit more motivation for what he’s doing.

    Just a thought.

  32. PlasmaPony says:

    Do the suicide robots start to explode on proximity or on a timer? I was thinking that if they explode on some sort of time maybe you could have a countdown appear on the robot itself counting down. Like once you aggro them they go off X amount of seconds later and you have that time to destroy them or get away. This can be paired with the explosion causing damage to nearby enemy robots and you could make suicide robots into both a boon and an obstacle. If there’s no timer and it’s pure proximity, find a color that you don’t use for many other enemy types and make them flash that when they aggro you, picking up before they explode. Or some other tell that other enemies don’t have. Maybe they shake and send out smoke or something, with it getting more intense before they explode. Or make their visual design scream BOMB in some way, perhaps to the extent that they look like a bomb or have a flashing explosive strapped to them. Kind of like how Valve uses (or rather used before all the hats) a very distinctive figure for the characters in TF2 so even at a glance it was clear what they were. If the visual design of the enemy looks like a bomb in some way, players will quickly put 2 and 2 together.

    As for tutorial, I think the best bet is to have a brief tutorial sequence accessed from the main menu. The player goes through a brief series of quick rooms that introduce you to some of the hazards. This allows the player to get an overview of basic gameplay without hampering the flow of the procedural generated levels, lets players go back if they need a refresher (or say if a friend is over trying it out), and also lets people skip it if they want to learn as they go. Stuff like “controls work like this, fly here. Good!” and then go to a next room with “dodge enemy fire”, and then progress through missiles and such. Then you can give a quick room tying it all together and then send them on their way to the main game.

  33. Probably Steve says:

    Just from looking at your screenshots and gifs, I don’t think the problem is enemy variety so much as the fact that there is glowing bullshit all over the screen and enemy bullets are almost invisible.

    But maybe I’m just getting old.

    If you do nerf the early game, make sure you add a “fucknuts mode” switch so that people who *have* played the game for 6 months won’t be bored for the first half of every session.

  34. Pyrrhic Gades says:

    RE: teaching a player that you can blow up missiles.

    Howabout an end level bossfight in which the boss fires a fuckton of missiles at once, such that the player wouldn’t be able to dodge ’em all. The player’d stop thinking of surviving and start trying to kill the boss before the boss kills him, and in the ensuring chaos he accidently blows up one of the incoming missiles. Once the player survives the first missile barrage, the player will then ask “WHY AM I STILL ALIVE?”, and then he may notice a corrolation between the good robot not dieing and the lasers striking the missiles.

    This’ll have to be an end-level boss (and the player’d have to be use to the concept of killing bosses == win level (and that’s IF levels may be finished off with boss fights and its not just a case of going from A to C)) for this valvy “tutorial” to work. The player shooting the boss (thereby winning the level) before the boss can kill them is a desperate gambit, and if the player doesn’t see the win he may just give up at the sight of so many missiles.

    Ofcoarse this may backfire in that player may see this as a cheap difficulty spike, and then cry bullshit after having their good robot blow up…

  35. Duoae says:

    I’m not a game designer but these are my thoughts on the issues:

    RE: Evading and shooting down missles:

    Maybe this is one of those instances of emergent player behaviour being the saviour of the learning curve…

    Start missiles off as being rather under-powered and so each hit hurts but not overly so. Then gradually ramp them up damage-wise. Players will note how much they’re being hurt and try and form tactics to avoid that damage. Starting with lower damage amounts mean that the player will be given more chances of learning the ‘correct’ behaviour.

    To mitigate instances where the player does not learn this ‘correct’ behaviour, (I know it’s a pain) but the game could keep track of how many missles hit out of how many were fired and reduce their damage based on some formula – probably a type of inverse exponential where as the no. of hits approaches 1 the damage increases to a plateau. This is essentially an invisible form of adaptive difficulty that selects only on a non-intuitive function of the game.

    RE: Bombers:

    Since robots are a black silhouette, having them inverse colour to a white silhouette and back again would be more dramatic and visually distinctive to separate them from all the other noise on the screen – assuming that there are not currently any robots that perform this type of colour swap. Think back to the bob-ombs in Mario Bros. You can increase their colour switching rate the nearer they get to the player to instil a sense of urgency and desperation.

    As for sounds – bear in mind that for audio/visual deficient players a single clue is usually not sufficient if it is colour or sound based. The above flashing technique will work for all colour blind players and deaf players.

    RE: Save stations/warranties/non-combat stuff

    Is there a mother-bot/computer programme in all this backstory? Could she/he/they/it be the information giver but only at the save stations/warranty stations/stores/etc? Might also be a story hook for the player to latch onto and empathise with.

  36. silver Harloe says:

    re: teaching to lead targets

    I would consider literally gating this. There’s a door, to pass the door you have to shoot the Thing. The Thing moves fast, but is otherwise unthreatening. You have to fire ahead of the Thing’s movement to open the door.

    re: suicide robots

    Inverse gate. Pick a graphical warning for robot about to splode, like flashing lights or whatever. Some bombers approach from the opposite side of some door. The Thing is on their side instead of the players, so the player can’t open the door. One of the bombers does the graphical-leadup to exploding and suicides against the Thing, opening the door. Maybe have a pair of doors, so the player sees the suicide sequence twice before it starts to happen against their hide.

    But I’m weird, and you might not have doors as a feature and it’s probably too late if you don’t, so just my thoughts.

  37. Nidokoenig says:

    Good way to teach a player about an enemy ability is to let them use it, and it stops a lot of bleeting about cheapness that isn’t. Want to show that missiles home? Let the player use missiles, they’ll pick up the idea when a volley kills an enemy and the remnants start veering towards another enemy. Want to show suicide bombing? Make it the default death animation and make it enough to do severe damage to even bosses, allowing players to burn a warranty or two on a tricky boss. They miss out on the no-death score bonus anyway, so let them.

    Want to show a player can do something? Have enemies do it. The Wonderful 101 has Vorkken, who shows up as a mid-boss basically at the earliest opportunity after the player gains a new team member and weapon, each time using that ability. Have a strong enemy that teaches the concept, like a sniper with a laser pointer so you can see him lead his shots. If the player has missiles, smarter enemies can dodge them and players can try it.

  38. Chris says:

    Regarding the bombers, maybe instead of having them change color over all, give them a super-bright glow that flashes or bright outline-light that spirals inward right before they explode. Not a fan of making it sound-dependent because if alot is going on in game someone could easily miss the signal until it is too late. Maybe have slow/over-sized bombers to start, and then release their smaller/faster model after a few levels?
    Could take a page out of Popcap’s book in terms of introducing enemies by giving a glossary on the main page that gives details about an enemy once a player has encountered them for the first time, ala Plants-vs-Zombies. Granted PVZ takes waaay too long to get thru the tutorial stuff.

  39. WJS says:

    My first thought was: “Why shouldn’t you have the same warning sound for bombers and missiles?” What, fundamentally, is the difference between the two? Hell, if missiles are an earlier game things than bombers, that could be part of how you introduce the latter.

Leave a Reply

Comments are moderated and may not be posted immediately. Required fields are marked *


Thanks for joining the discussion. Be nice, don't post angry, and enjoy yourself. This is supposed to be fun.

You can enclose spoilers in <strike> tags like so:
<strike>Darth Vader is Luke's father!</strike>

You can make things italics like this:
Can you imagine having Darth Vader as your <i>father</i>?

You can make things bold like this:
I'm <b>very</b> glad Darth Vader isn't my father.

You can make links like this:
I'm reading about <a href="">Darth Vader</a> on Wikipedia!

You can quote someone like this:
Darth Vader said <blockquote>Luke, I am your father.</blockquote>