The Route-Planning Problem

When you play a game for the first time, you probably just kind of bumble through it, learning as you go, until you hit the end (or until you quit). I am certain this accounts for well over 99% of all gameplay experiences out there.

When you play a game the second time, it’s likely you’re looking for a different experience from your first outing. You’ll make different choices, or hike up the difficulty, or pick a different character. This may or may not sate your thirst for new experiences, depending on the game.

There reaches a point, though, where your familiarity with a game outpaces your boredom with it. I’m sure each of us has at least one title we will never be done playing, no matter how many times we complete it. A subset of us want to keep getting those new experiences, though; that’s when we enter the realm of challenge runs.

What a lot of people don’t realize, when they’re watching speedrunners or reading in-depth LPs, is that playing a single-player game well doesn’t usually require a lot of what we typically think of as “practice”. Surely that’s part of it; you can’t master a game without first mastering its systems. But planning takes a hell of a lot more time. I’ve “played” Super Metroid more times in my head than I have on an SNES, visualizing the game state in a bunch of different ways, hashing out ideas with other Super Metroid players, finding new tricks and shortcuts, searching for new ways to enjoy the game. The end result of this sort of theorycrafting is a new world record, or an amazing tool-assisted speedrun, or the discovery of a new glitch which breaks the game wide open.

I pay a lot of attention to route-planning when I go back to revisit old games. It isn’t just a desire to finish the game quickly — I’m not a speedrunner, so the stakes on my end are pretty low — so much as an innate need to be efficient. No point in taking two trips where one would suffice, is there? I’ll never be able to match the world records, but at home in my game room I can demonstrate my mastery over a game by knowing it well enough to chart the shortest course through it. Forgetting to do something, and having to go back and do it, is a bigger failing in my estimation than getting a game over.

This mindset has served me well in my Let’s Play videos. What you see in the videos is almost never the unabridged Brickroad experience; it’s the refined, razor-sharp playstyle of a man who has the game sitting in his working memory. Barring missteps like Shantae and Mega Man 10, I never record a game without doing a full test run first.

Pick the shortest route, then gun it. Very rarely do I have to make tough decisions about what to do or where to go as I’m recording; all those decisions have already been made, all the variations tested, and usually my plotted course is sitting in front of me in the form of a page of notes. A sort of gameplay itinerary, if you will.

I have to wonder, though, whether some games were designed with the goal of tripping route-planners up in their tracks. Legacy of the Wizard is such a game. I’ve been banging my head against this one all week.

Let me set the stage, before I go into detail. Legacy is an old NES game, infamous for being obtuse and inscrutable. The main barrier of entry is simply figuring out what the hell to do as you make your way through a gargantuan, often unfair dungeon in search of treasure. I know this game inside and out. I have it all mapped out in my head. I know where all the pieces are, and how they fit together. I know the precise sequence of events that have to be ticked off before the game will let you win. The combination is “bitch hard oldschool game design” and “I have this shits memorized” is precisely the sort of thing that makes for an LP, in my experience. What’s more, it’s trivial to play the game at 150% speed.

So why haven’t I LPed Legacy of the Wizard? Because I swear to Christ there is no efficient path through the game. No matter how you arrange the sequence of events, you’re left backtracking through huge, dull portions of the map or spending a half hour farming for gold.

I’m going to explain the problem in abstract. To begin with: this game has four main player characters: A, B, C and D. Their task is to explore a gigantic underground labyrinth. The labyrinth is split roughly into four sections, each of which reflects the abilities of one of the PCs: A, B, C and D. At the end of each dungeon section is a boss: W, X, Y and Z. After the four bosses have been killed, you use a fifth player character (E) to kill the final boss and finish the game. (E is useless in any other regard, though, and isn’t part of the problem I’m having.)

The first issue is, some characters are stronger than others. All four characters can defeat bosses W and X, but only A and B can reliably kill Y and Z. C and D are much too weak to get through the health of the latter two bosses, so any route has to ensure that A and B are used to kill them.

There is a boss at the end of each dungeon section, but bosses and dungeon sections do not correlate. The first boss you kill is always W, regardless of who you’re using or where you are. The last boss is always Z. So, the best route is obviously to plan to do A and B’s sections last.

Ah, but there’s a hitch! Of the four characters, only D can traverse his entire dungeon section without the use of special equipment. The equipment the other characters need is strewn throughout the dungeon (either in treasure boxes, or in shops). So, whatever route you plan, you have to make sure your next character has what s/he needs to get to the boss. Here’s what that looks like:

  • There is a big money bag sitting near the entrance. This is helpful, but can only be collected once.
  • A only needs one item, found very close to the beginning of the game. Anyone can collect it. Every run will start with the player getting this item.
  • A’s dungeon contains one each of B and C’s items.
  • B needs three items. One is found in A’s dungeon. One is found in C’s dungeon. One is found here in B’s, but only C can collect it. Alternately, that last item can be purchased for an exuberant sum of gold. (I don’t recommend this.)
  • C needs two items. One is in A’s dungeon. The other is very deep in B’s, and can only be collected by B. Alternately, that second item can be purchased inside C’s dungeon for a huge sum of gold.
  • D doesn’t need any items, and his dungeon doesn’t contain any items. It does, however, contain a lot of gold.
  • Finally, whichever character fights bosses Y and Z needs a defensive item that can only be bought in a shop (accessible by C alone, with the item from A’s dungeon) for damn near max gold.
    If that looked like a tangled-ass mess to you, all I can say is well, yeah.

    At first it looks like you have to do A’s dungeon first. That’s where you get C’s item, after all; without it, much of the game is inaccessible. But if you do A’s dungeon first, that means you’re going to have incredibly difficult boss fights later, as the weaker C or D have to step up to the plate. Not impossible, but definitely in the zip code of a crap shoot. Of course, the desired item is buried very deep into A’s dungeon, so grabbing it and then leaving means a holy crapton of backtracking later on. Not desireable!

    At second glance it looks like doing D’s dungeon first is a good choice. It’s an opportunity to pick up a bunch of gold, after all. Plus, he’s the character least suited for boss fights. This is a no-brainer! Except, all that gold does you no good without anything to spend it on — and you can’t reach any of the good shops without C’s item from A’s dungeon. So you still have to clear A’s dungeon first, in order to make any headway. Oh, and leaving the gold there fore later isn’t an option, because the treasure box sits in the way of the path you need to take. You can’t reach the boss fight without opening it.

    So you do A first, then a shopping spree, then D, then take C into her dungeon with enough money to buy the item she needs. This pits her against Boss Y, who has more health than C can dish out damage. Damaging monsters uses up Magic, you see, and C only has 100 points of it. Because she needs two items to reach the boss, and her third slot is taken up by the defensive item required to survive, this leaves no room for a Magic-restoring potion. The C vs. Y fight is not impossible, but it is extraordinarily luck-driven, and her level is very long. Too long, in fact, to leave to chance.

    Okay, so I may have fibbed a little earlier: D’s boss can actually be reached by anyone. This means you can leave D on the bench the entire game, clear A’s level first, then give C her easy fight (against X), then have everything you need for B’s level, and then bring A back out agains the toughest boss in D’s level. This is the shortest route I know of in the game, requiring no backtracking whatsoever, but there are two enormous problems with it:

  • Right before the boss is a trap which instantly kills every character except D. A has an item that can circumvent the trap, but seems to do so on a coin toss. See C’s “level is too long for a luck-driven battle” point, above.
  • Leaving D’s level for last means you have to either buy the defensive item (required for the Y and Z fights) or C’s second item without the help of a bag of gold. This is possible, but again, it’s entirely up to luck; sometimes monsters drop gold, sometimes they drop poison. You might get away with doing this, or C might end up farming monsters because she came up short.
    You have to do A’s level first. But you can’t do A’s level first. It’s a damn conundrum, is what it is.

    Backtracking half of A’s incredibly long, slow level looks sloppy. Farming gold with C looks sloppy. Fighting through D’s level twice because the coin toss came up wrong looks sloppy. You can’t play this game “well” without looking sloppy, or without manipulating luck! (Or using savestates. But, eh, savestates. It burnssss us.)

    So what’s a route-planner to do, when all the available routes suck?

  • 6 comments to The Route-Planning Problem

    • Shifter

      Why not go for the luck route at 300% speed, there for, if you do have to do it twice it’d take as long as a normal 150% run… Your insane btw!

    • I’m totally with you re: efficiency. Knowing there is a slim, perfect, backtracking-free route through the game is the best feeling in the world, and figuring it out is my personal criteria for mastery. Like you, I’m never going to be out there breaking world records or wowing the world with my mad skills, so the best I can do is work at the game until it’s a pleasingly fat-free experience.

    • Grab the mattock before recording the LP? Or, ~equivalently, find a password online that begins the game with the mattock in your inventory?

      If you felt like being creative, you could pretend the game starts with a dude finding a chest in a dungeon and using a crystal to escape and warn his family.

      Otherwise, I’d say you’re down to either pausing the recording while you re-play Xemn’s section or sucking it up and pushing a bunch of annoying blocks around twice. You could use the second time through to explain to the viewers most of what you’ve said above. And frankly, I’d argue that frustration and back-tracking are an integral part of the authentic Legacy of the Wizard experience.

    • Solitayre

      If you are trying to LP Legacy of the Wizard without backtracking five times, you are doing it wrong.

    • WIP

      I had this game as a kid and could NEVER get anywhere in it.

    • Rosewood

      I’ve been watching Legacy of the Wizard on Backloggery as they stream it intermittently on Mondays, which has been my first exposure to the game. I found it funny/interesting that when faced with grinding 80 gold a second time after dying, and the streamer asked his audience “Should I grind some more or move on to something else?” his audience asked him to grind! In any case, I’m sure a sloppy run, whatever that might amount to, would be fine with your audience, especially if you explain the game’s structure to them (and how it’s messing you over), as you have in this post.

    Leave a Reply

    You can use these HTML tags

    <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>