Thursday, 20 May 2010

VVVVVV's Lesson

If you want to set friends at each others' throats over a game, there's no better question than "how hard was it?" You're asking "how much did the game frustrate you," or, finally, "how much do you suck at playing games?" The conversation enters a death spiral as one party says old games aren't playable, another says every good game is hard, another says that real men play Wizardry, and someone or everyone is called a "baby."

But it doesn't have to be like this! If everyone plays Terry Cavanagh's VVVVVV, the world might be mended.

There's a sequence in VVVVVV that deserves to be famous. Everyone who plays the game will remember it; Kieron Gillen already wrote about it. It's the titular sequence of rooms called Veni, Vidi, Vici!*, and in it you learn how to play games again.

In VVVVVV, you use a simple gravity-flipping mechanic (no jumping) to progress through various snappy platforming challenges, packaged as individual rooms. Checkpoints appear in nearly every room and you respawn fast; Quintin Smith probably said it best by saying that VVVVVV's checkpoints "allow players to exist forever in the scorching heat of insurmountable challenges."

The main plot, rescuing stranded members of your crew, is tough; the VVV rooms themselves are an optional challenge. They're a crazy dare you can't help but accept, because you'll always remember that you couldn't do it if you don't do it.

Like every good dare, they start with a taunt. You stand on a platform on the first screen, "Doing Things The Hard Way" (every room in the game is named something), separated from the extra credit bauble by a waist-high barrier. Damn your non-vaulting legs! It's a mark of the designer's wit that he begins this notorious segment with a literal stumbling block.

The only thing to do, then, is fall up, speeding through 6 screens bristling with spikes, flashing by quickly enough that the lurid solid colors conspire to give you a seizure. The 6th screen (Getting Here Is Half the Fun) lets you rest on a dissolving platform for about half a second before you head all the way back down. Going back is the hard part.

This sequence is the heart of the game, and while it may not be the heart of gaming itself, it is one of several hearts keeping the monster that is gaming alive. 

The VVV rooms are pure memory. You come untethered from a checkpoint for probably the longest period anywhere in the game. You only need to press the left and right keys, and the action key, once, in the 6th room, at the peak/nadir of your jump/fall. Completing the sequence seems impossible at first, but as you fall through these screens you learn them literally forward and backwards; you find yourself getting just a bit farther, miraculously, with most attempts. You learn the exact amount of pressure to apply on the Left arrow when exiting Vidi to place yourself in the center of the narrow path between Vici!'s walls of spikes. You teach yourself to move precisely.

In player input, the challenge resembles outracing the Quick Man lasers in Mega Man 2, which also required a precise, memorized series of left and right taps from the player, who evaded death beams by falling as they emerged from the sides of the screen. Players had to commit the sequence of platforms below to memory to scramble off them in time to avoid the beams. This famously difficult setpiece could be avoided if players had acquired the Time Stop ability, but where was the fun in that? 

VVVVVVV's memory challenge is harder than Mega Man's (I know because I just replayed the Quick Man stage, piece of cake) but it also isolates the hard section by installing a checkpoint right at the base of it; Mega Man 2's killer sequence arrives in the middle of a level full of other difficulties. Cavanagh removes every complication, every concern beyond the task at hand, leaving only a test of memory and performance.

Memorization isn't often a glamorous activity, but it's one of gaming's essential pleasures. In any difficult platformer (and I can't name a satisfying easy platformer), you will encounter a group of obstacles that seem impossible to overcome; you will invent a plan, or trick, to handle each of these individual obstacles in turn; you will learn to execute this plan perfectly; you will feel very good when it finally works.

You might remember part of level 5 of Castlevania where you can push an axe-throwing knight off the screen, which is easier than trying to kill him. Or you remember a part of Ninja Gaiden 6-2 where it is safer to jump over a bird than kill it (because killing it would cause it to respawn near the platform you're trying to reach, but if you leave it alive the NES won't be able to draw two birds at once, and will have to wait until the first exits the screen). There are hundreds of tiny strategies that led to your victories over these levels and their bosses. You might not even remember that you know them.

The best moments of GameCenter CX, a Japanese show where comedian Shinya Arino attempts to beat infamous classic games, focus on his plans for breaking down the toughest levels. He often resorts to pausing the game and consulting with his "assistant" (who is apparently an expert gamer) as they illustrate their plan of action on a whiteboard. Sometimes the show even presents an map of the level that Arino is breaking down, with arrows demonstrating the correct path, like a sportscaster illustrating a replay via light pen. From an episode on Solomon's Key:

Episode subtitled by TV-Nihon.

In VVVVVV, the form of the plan is obvious: moderate right while rising through Veni, hard left in Vidi (with taps to adjust at the top to position yourself for the middle of the path of Vici! during screen transition), prepare to head right at the top of Vici!, straight halfway through Easy Mode, then a moderate right, etc. As you make incremental improvements each time, you really do learn it; once you've done the sequence once, it's not so hard to do it again.

 The victory screen. I had to beat it twice because I forgot to take a goddamn screenshot the first time.

When you finally beat the thing, you feel a massive wave of -- relief? Delight? Enlightenment? You realize that you can beat anything, if you commit to it; that there's nothing truly hard about gaming. I can't translate the message without losing the power that the game has when it teaches you this in gaming's original tongue. For all the posturing by elitists in the gaming community, there's no skill to games but patience. That's the lesson of VVVVVV

And once you've learned that you can complete this challenge, platform gaming at its most elemental, you see that all the games you once gave up on are doable. Alien Soldier, Contra, one of the NES Castlevanias, whatever it is -- if you learned the Veni Vidi Vici pattern, what's stopping you from learning the paths to beat those games?

Secretly, VVVVVV is not just a didactic game, but also a nice game. You don't notice until you take a good look; then you see, in the victory shot above, that once you get to the right of the stumbling block and collect the token, there's a (relatively) safe little passage for you on the underside of the platform, leading you back to the checkpoint. (A truly cruel game would make you go all the way through the Veni Vidi Vici chambers again, backwards, to save your progress.)

One of the stages in the middle of the VVV sequence is called Easy Mode Unlocked. It's a joke, like most of the room names. It's a joke about the games that do tell you when you've died too many times and should switch to easy mode; these are often the same games that make reviewers bitch and moan about difficulty spikes, as if games was only allowed to be challenging on the very last level, and probably not even then, because the focus group didn't like that.

"Easy Mode Unlocked" is the most insulting thing you could ever tell the player -- "since you're not good enough for the game as it was meant to be played, why not try it with training wheels on?" Why not take the easy way out, instead of learning or improving? But Cavanagh's joke here is a reassurance that his game doesn't roll like that. VVVVVV would never insult you by cutting you a break.


Beating the Veni Vidi Vici! chambers is far from the hardest task in gaming, but that the sequence distills something essential about effort and reward structures. It's much more effective than something like IWBTG, a clumsy, ugly game made with the single goal of being quite hard.

Without the frustrating difficulty, there would be little fun in this sequence. No matter how much artists and scripters may dress up a platformer, there's no joy in playing unless the game takes practice to overcome. Without a buildup, payoffs don't work. A game will present a diegetic story, aping the Hollywood structure, about a man (in games, almost always a man) defeating or bypassing a series of obstacles; but the experience of a game is also the story of what happened to you as you played, of the obstacles that you overcame. If a game never works to get you stuck or overwhelmed or enraged, if there are no real obstacles to your success, then that second narrative fails, and you might as well be watching a movie.

Maybe everyone thinks everyone knows about this second, crucial narrative; paying attention to it might be the foundation of New Games Journalism anyway. Why, then, do developers and game writers act like they forgot (or never learned) the reason for making games hard? Why do you see articles like this one, which claims that "an increasing amount of data" [citation needed] says games should be easier**. Games should tone down the challenges that might frighten customers away; developers need to child-proof all these dangerous edges! Never mind that hard games are already becoming rare. What data is being referred to, how was it collected, and what games are these stats were drawn from? Davison cites "telemetry data," which I know is pretty impressive, because MI6 sent me to retrieve it from a silo in Goldeneye.

Here's a funny story you might know already: did you hear why Demon's Souls was released by Atlus in the US? 1UP claims SCEA passed on it, figuring that it would only sell "15,000 copies" in its initial run, due to "unorthodox design and unrelenting difficulty." This turned out to be a huge mistake; it sold half a million copies, the majority of them in the US. Hard games can sell. I'm guessing the people Davison talked to were relying on the same kinds of market research that told SCEA that Demon's Souls would never be a hit. (I don't actually know who Davison was talking to, or what research they used, because he never cites any evidence.)

If you want a test case for an easy, short game, coast through the 5-hour Splinter Cell: Conviction some time. The story of that game ended for me the minute I figured out that when I crouched behind a crate and began shooting every enemy in the area, I was playing a shooter that was easier than Gears of War. Sure, I half-heartedly stealthed my way through the rest of the game, but I had already seen the rigging behind the stage; I could never really believe in it again. Even as I sat through the arcane horseshit of the game's cutscenes, I was painfully aware that Sam's enemies were not really black ops motherfuckers, but a gang of half-blind screamers who couldn't shoot straight. I couldn't lose; I was much better at aiming for the head than they were.

Developers should stop thinking of difficulty as a rough patch to be smoothed out. It's a narrative device. If you want gamers to feel they've triumphed against insurmountable odds, actually pit them against something overwhelming. Require some effort before you heap up rewards. We're all wise to fake scripted scares, "urgent situations" that aren't actually timed, and a whole bag of tricks used by developers to disguise situations that are truly tame. Create difficulty spikes, difficulty hills, whatever, as long as it fits the plot. If you want players to beat your game, make it worth beating. 

*I assume Veni vidi vici is the titular phrase, and not Vi veri veniversum vivus vici, which is short a V anyway.

**I didn't want to write an entire post about Davison's piece, but here are a few ideas:
1. Maybe consumer buying habits aren't motivated by the amount of time they will actually play a game, but rather the amount of time they imagine they could play it? If that was true, then developers would still have an incentive to make longer games. I often buy games that promise many hours of gameplay, but I tend to rent or borrow games that are 5 hours or shorter.

2. Most people play only the first four or five hours of a game? Maybe someone should tell them to buy fewer games. (Just messing with you, video game industry!)

3. I heard the "90% of people only play the first few hours of a game" line at least 5 years ago, and I never saw a citation for it then, either. But, assuming it's true, it's not a new issue.

4. Davison makes an interesting point about making games more modular. However, DLC is almost always  poorly conceived, poorly balanced, buggy, and well below the standard set by the original title. So, I would suggest that developers start making their games much harder and simultaneously ramp up the production of game-breaking superweapons and items as launch-day DLC. Companies that follow this strategy will be able to seed feelings of helplessness and inadequacy in players, then convert that weakness into cold hard cash.

5. Do games that attach significant bonuses to their "game completed" Achievements coax more playtime out of people than games that don't? ME2 had a lot of end-related bonuses, but the game was pretty damn easy as well.

6. Finally, a "won't someone think of the art" plea: if studios begin capping the possible length of their games as a matter of policy, they also limit the variety of stories they can tell. One of Red Dead Redemption's best storytelling tricks comes from its willingness to tell a story that goes on and on. A film can only be around three hours long before they start getting impractical to show in theaters (owners want to show a film as many times per day as they can). Games can be much, much longer, and their scope can be as great as a serial format TV show. If every game shrinks to 5 hours, they'll probably all imitate blockbusters, and be the worse for it.

Mega Man 2 image taken from the forums. Email me at


  1. Funny that I read this only 10 minutes after sending a message to my partner about our options for easy mode and skippable levels for our small game in production.

    Although, our game falls plainly into the puzzle-platformer genre, not the twitch-platformer genre, so the difficulty is less about 'executing what you already know' and more about 'knowing what you must execute.' (Though both involve a lot of trial & error and patience, like you mentioned.)

    The last option I wrote to him was that we could keep our game the way it is and have players "rough it out" but that "I think that's dangerous in a puzzle game."

    I'd like to know if you have any thoughts on difficulty in puzzle platformers, which can be just as punishing as twitch platformers, but have many more variables and input options (with twitch platformers, most of that is "where to jump" & "when to jump"). Puzzle platformers can run into Infinite Frustration Zones where some players have no idea where to begin (or, sometimes, no idea what they're doing wrong), even when well designed; and I'm worried about how to mitigate that in my game without diluting the puzzles themselves.

    In any case, I really enjoyed this article. I was up way late last night playing Bit.Trip Runner, doing what you describe: trying to conquer the frustratingly difficult because I knew that, in the end, it all came down to memory and patience; like Contra, like Mega Man, like VVVVVV, like all the platformers I've played before it. And, you're right, it is satisfying.

    It's true that real men play Wizardry, though. ;)

  2. Good read, but one should be also aware of the negative sides of high difficulty - nice take on the subject: though the problems described in that link don't apply to games like VVVVVV...

  3. I agree that games need difficulty, and I agree that John Davison's post is mostly wrong.
    But I only want games to be as difficult as VVVVVV's main rooms, not the optional ones. I gave up on Veni Vidi Vici after a minute, because while I could see that I was getting better, and that I would eventually beat it, I also realized that it would take me probably an hour. And I just didn't want to do that. At the end I had 14 out of the 20 shiny things. And out of those that I missed I only gave up on three. The others I didn't find. So yeah, challenge is cool, but keep the extremes away from me.

    As for John's post: I think he is misreading the stats. When it turns out that a lot of people don't play a game for more than a couple of hours, the more probable explanation is, that they didn't like the game. If you look at the Steam stats, you simply can't explain the drop of players at the beginning of HL2:Episode 2 with the game being too long or too hard.

    I'm probably going to do a proper response to him some day...

  4. "VVVVVV" refers to the spikes in the game IIRC.

    "3. I heard the "90% of people only play the first few hours of a game" line at least 5 years ago, and I never saw a citation for it then, either. But, assuming it's true, it's not a new issue."

    Maybe not a new issue but it's becoming a bigger issue since, first of all, games are much more expensive to make, and secondly, game developers are becoming more aware of it and are just now getting ways of substantiating it.

    If you make a game you want people who play it to finish it.

    I like VVVVVV too but it has nothing to do with the GamePro article. VVVVVV is a very niche game for a completely different audience than the one they're talking about. Playing games for the challenge can be fun but the kinds of games that facilitate that are becoming less relevant. Demon Souls sold 0.5M copies? Great, but it's still just an exception. It's not the model for what most people are looking for in a game and anecdotally a lot of people are finishing that game either.

  5. I think in a lot of people's cases the title stands for Veni Vici Veni Vici ... Veni Vidi ;)

  6. The biggest merit of VVVVVV, in my opinion, is that all the challenges are self-contained. Having Veni Vidi Vici in another game where every time you fail, you have to replay the previous 5 minutes of the game would be incredibly frustrating for me.

    For example, I played the Call of Duty 4 campaign on Veteran difficulty because that's what the game suggested to me after running the training course. It was fine until one of the flashback missions, where the enemies throw incredible amounts of grenades at you. After surviving and carefully pushing forward for 10 or more minutes, I'd eventually get killed and have to repeat those 10 minutes all over again.

    Even though it didn't take nearly as many tries as VVV, it was a very, very frustrating experience. VVVVVV, on the other hand, never made me feel that way at all. I would play another VVVVVV, but I'd probably tone down the difficulty if I were to play another Call of Duty.

  7. @Scypher: Thanks for the awesome comment! I wasn't thinking about puzzles at all when I wrote this, and I can only imagine how difficult it is to design them. If your test players can't handle the difficulty, maybe the solution is to prepare them more via the early, easy puzzles, which ease players into thinking the way your later puzzles require them to think? Doing more "foreshadowing" (or whatever you want to call it) seems preferable to me than directly watering down late puzzles. But again, I know nothing about designing puzzle games.

    @teoliit: Whoa, spikes as the title makes a lot of sense.

    I feel like there are two concerns getting confused here (as much my fault as Davison's). "How many people buy the game" and "how many people finish it" are separate issues. Nobody has actually provided the data to substantiate the claim that "fewer people finish harder games"; furthermore, nobody has connected it to sales, which is really what matters when your games are getting expensive.

    You may be right that designers want everyone who plays their game to finish it. I think it would be better if their main concern was making a good game, though. I don't mean to say that "all good games are hard," like the joke at the start of the post. However, I feel like some recent games lose credibility by being too easy, and too concerned with staying in the player's comfort zone.

    I don't believe VVVVVV is a title that restricts its audience because of its difficulty. It's true that it's an indie game and it hasn't penetrated the mainstream. It didn't have an advertising campaign and you can't play it on the Wii or XBLA. But there's nothing about the content that makes it "niche." The checkpoints are frequent enough that anyone with patience can complete the main story, and I think its approach actually teaches people to be more open to gaming challenges. (Even if they don't beat the Veni Vidi Vici! rooms.)

    Demon's Souls is a very hard game that sold well (for its budget) with little to no advertising. I'm not saying that a game that is AS hard as DS could outsell MW2, but I think a pretty hard game with AAA presentation (I hate that phrase) could sell millions. If Uncharted 2 had featured harder, more elaborate puzzles and climbing sections, I think it might have still sold as well as it did and been a better game. (As it stands, the climbing and puzzling in that game is just busywork.) That's just my guess, though.

    I think difficulty is a device that developers should be willing to use. Difficult games go against current industry trends, so they're a risk. If developers want to make better games, they should consider taking that risk.

    @Thiago: I wonder when Cavanagh decided to add that brilliant checkpoint system -- whether it was part of his original idea, or if he put it in when he realized that the game he was making was really hard. Gillen pegged the game perfectly by describing its "fairness." Its clarity and precision are really worlds apart from something like the grenade spam you describe, which is just Bullshit Hard.

    I think it's a matter of bad design (in CoD4's case) rather than being too hard, though. There should be a way to beat a section of a game that isn't just dumb luck, some element to figure out or a task to improve at.

  8. I want to second Sagan's comment about gamers not finishing games because they just don't like them.

    It's funny how developers always want to blame the gamers, "We can't make long games because gamers have short attention spans. We need to make easier games because gamers are lazy. We need to make dumber games because gamers are stupid."

    But they never seem to consider the possibility, "We need to make games that don't suck." Or, in the case of something like Half Life 2, "Even if we make a great game, not everyone is going to like it. We need to ditch the idea that one game can be a blockbuster that every single gamer will enjoy."

  9. By the way - that 50% quote is from a few places. One of the better ones was Valve saying that about 50% of the people who bought Episode 1 completed it. And, of course, it's only 4 hours long.


  10. Sadly, this is exactly what stopped me playing VVVVVVV (and puts me off most platform games, sooner or later). Not my kind of challenge. I want something with a clever solution, not a trial-and-error solution.

    I see the platform.
    I see what I have to do.
    If I practice, I will be able to do it.
    I will be able to say I have done it.

    I cannot be bothered.

  11. @CdrJameson: If you don't like the genre, that's fair enough. The bulk of the game is more doable than the vvv rooms (or the other optional challenges), but its challenge is more platforming than puzzle-platforming, so it fits your pattern.

    @Kieron Gillen: I believe Valve's figures, but I don't think 50% completion is that shocking a number, and Ep 1 is just one game. The "over 90% of gamers don't finish games" number cited by that GamePro column is the claim I'd really like to see a source for.

  12. @Chris/Kieron: I find 50% completion on even a short game quite high.

    As a perennial non-completer of games I did a quick intra-personal study. I'm not particularly good at games, and as a game developer I'm expecting to see a lot of games in not much detail...

    I have 35 games on Steam, not including minor expansions.

    4 cannot be completed (multiplayer)
    6 I don't want, but came with bundles

    Of the 25 remaining,

    7 have been completed*
    4 have received significant play, but not to completion.
    8 have received only cursory play.
    6 have never even been played, or in some cases installed.

    These are not bad games - the 'not installed' category includes STALKER and Dawn of War II - I just haven't got round to them. Of the 18 uncompleted games there are 13 I'd like to complete, but haven't.

    This gives me a 20% completion rate, which is surprisingly high.

    My interest piqued, it's time for a quick flick through my CD wallets:

    XBox: 1 game completed out of 26 possible completions, and that's Psychonauts again.

    Dreamcast: gets 2 of 12 (Rez, Jet Set Radio)

    Playstation: 5 of 22 (Front Mission 3, Silent Hill, Command & Conquer, Incredible Crisis, The Misadventures of Tron Bonne). Hang on, that's quite a lot.

    Playstation 2: 4 of 51 (Detonator, Ico, Red Dead Revolver, Transformers)

    Wii: None of my Wii games can be completed. DOES THAT SAY SOMETHING? Hmmm...

    PC CDs:
    19 of 148 (Lego Star Wars, Mass Effect, Deus Ex II, Rome: Total War, Max Payne, System Shock 2, Civ 2, Deus Ex, Colonization, Dark Forces, Hidden & Dangerous, Freespace, Gunman Chronicles, Mafia, Blade Runner, Call of Duty 4, Broken Sword 2, Planescape: Torment, Under a Killing Moon)

    What does this show? That I should probably buy a more modern console, and that someone's nicked my copy of Mechwarrior 2. No, hang on, that with 31 of 249 games complete I'm still only at at 87% non-completion.


    I expected my own figure to be around 97%. I buy games like they were sweeties, and have no intention of even trying to complete most of them and yet even then I can't manage to fail to complete a decent, respectable 90%.

    In conclusion:

    That 90%+ figure sounds like a load of old crap.

    Thank you, good night.

    (Actually I'm going to think on this further. Why these games? Why not the others? Why did I stop?)

    *Half Life & expansions; HL2; Portal; UFO multiple times; Psychonauts, but only because I'm not counting the Meat Circus.

  13. Good read. I think it's immeasurably difficult for games designers to get the difficulty balance right, since there are so many factors.

    There's something to be said for your idea that games are more about patience than they are about skill. But there's no denying that when a person runs out of patience, then that's it - they are no longer enjoying the game.

    It's different for everyone but sometimes the relief of finally achieving victory over a task just wasn't worth all the frustration. It affects you like a particularly bad hangover. You should have good memories of when you were enjoying yourself, but you don't. You just feel irritable.

  14. Nice one dude.It took me around 500 deaths to get that trinket, but the feeling was really unique.And around the death numero 420 I noticed that the game's name is actually a abbreviation for Veni, Vidi, Vici, Vici, Vidi, Veni (or so I think) and the feeling was once again unique.
    One of the greatest games that came from the depths of the indie mind..

  15. (One other thing - I think the 'Easy Mode Unlocked' screen name is a joke referring to the fact that you pass all three VVV screens on the way to the beginning of the awxrion in the (relative) safety of a perfectly straight (though spike-lined) tunnel. But obviously it passes by so quickly that not every player is going to notice this.)