The Stack

One thing I like a lot about Psychonauts is its variability of style. Each mental world is a representation of one character’s mind, and conforms to that character’s world view, with greater or lesser degrees of abstraction and phantasmagoria mixed in. One person has decided to treat life as an endless party, and has a mind decorated like a discotheque. Another is paranoid, and has a mental landscape that’s fragmented into disjointed sidewalks floating in the abyss in varying orientation, lined with tract houses, a veneer of normality that completely fails to conceal the bizarre. One guy’s inner world is rendered in imitation of the paintings he does. Since his medium is fluorescent paint on black velvet, this makes for the strongest visual shift yet.

I think my favorite so far is the inner world of Fred Bonaparte, a classical loonie who thinks he’s Napoleon. Sort of. It’s more like he’s got a split personality and one of those personalities is Napoleon. (Or, heck, maybe he really is possessed by Napoleon’s ghost. It wouldn’t be the weirdest thing in the game.) In his mind, Fred and Napoleon are playing a hex-based wargame, which you can descend into and explore at two different levels of scale, shifting in size as appropriate to interact with the buildings on the map or the wooden soldiers. (Both soldiers and buildings occupy one hex each.) With the size shifts and the conversing with game pieces, it’s kind of like Alice in Wonderland meets Avalon Hill. And, well, the game just looks good. It’s that lush mixture of gameboard and diorama that the really dedicated hobbyists put together, the sort of thing that makes you want to learn the rules just so you’ll be allowed to touch it.

I suppose that switching around the look from level to level is really one of the fundamental techniques behind platformers: it’s a way to create variety when the gameplay has limited range. Not that this is much of an issue here.

Psychonauts has a fairly complicated system of item collection, and it’s instructive to compare it to that seen in Lego Star Wars 2 (described previously in this blog). I described the latter as byzantine and difficult to understand, with its multiple goals and overlapping effects. Now, Psychonauts has more types of collectible. In the real world sections, there are:

• Psi challenge markers: few in number and located in difficult-to-reach places
• Psi cards: fragments that can be assembled into new psi challenge markers. More numerous and usually easy to reach, but often hard to spot
• Scavenger hunt items: unique objects, most of which require solving an optional puzzle

And in the mental realms:

• Figments of the imagination: all over the place, sometimes moving
• Mental cobwebs: collectible only using a special piece of equipment, and can be turned into psi cards back in the real world
• Memories: located in ambulatory safes that flee your approach; when collected, offer glimpses into character backstory
• Emotional baggage: a two-step collectible that involves finding a tag to match with each bag; matching all five in a mind unlocks some production art

However, I find the collection in Psychonauts easier to follow, and I think it’s mainly because the effects are simpler. With the exception of the effects mentioned above of the memories and the emotional baggage (which don’t seem to affect gameplay, and can be classed as “extras”), the end result of collection is always the same: increasing your “rank”. Attaining certain ranks grants permission to buy new equipment, or allows you to learn new powers, or enhances the powers you have. By funneling everything through the “rank” concept, the game simplifies the way you can think about gaining access to stuff. I suppose that this is something that RPGs have been doing all along with character levels, but it’s unusual to see a level system that isn’t at all based on gaining experience through combat.

Now, getting all the collectibles in everyone’s mental worlds typically involves going through them more than once. Indeed, it’s completely impossible to collect mental cobwebs from the first couple of minds on your first visit, as you don’t yet have access to the necessary equipment. It’s not uncommon in these games for collection to involve repeat visits to completed levels, but it’s usually handled non-diegetically: to use Lego Star Wars as an example again, if you replay the Battle of Hoth level, there’s no sense that you’re creating a story in which the Battle of Hoth was fought more than once. In Psychonauts, however, every visit is taken to be part of the same ongoing story, even if this requires convoluted excuses. So, the first time train with Sasha Nein, his mental defenses spin out of control and he urgently needs your help to restore order, but when you go back, Sasha offers to put you through the same “training course”, explaining that the emergency was a ruse to motivate you.

A more extreme example: In the mental landscape of the monster lungfish, it’s Raz who’s the monster, a Godzilla-like building-crumbling behemoth attacking Lungfishopolis. When you complete the level, by destroying a certain broadcast tower (the lungfish’s mental representation of the crainial implants that the bad guys are using to control it), the lungfish of the city are freed from their brainwashing and come to regard you as a hero. On a return visit, you’re cordially greeted by a lungfish who warns you that there are still some crazy guys out there who “don’t realize that the war is over”, and invites you to destroy as many buildings as you want: now that they’re free, they realize that lungfish were meant to live in mucus-lined bubbles at the bottom of lakes, not in tower blocks.

These scenes make a virtue of a constraint: they treat an unlikely situation resulting from gameplay decisions as an opportunity for humor. Or possibly the other way around: they use the fact that this is a humorous game as an excuse to integrate gameplay decisions with the story in ways that would be implausible if taken seriously. Whichever way it goes, it’s something that Schafer has been doing since his Lucasarts days.

I think I’ve been missing things that the designers of Psychonauts wanted me to see.

Early on in the game, Raz is given access to his own mental landscape, his dreamworld. There, he sees a mad doctor in a tower menacing Dogen, one of the other kids at the camp, and threatening to extract his brain. Raz is convinced that this is more than just a dream, that he’s witnessing something real. His default dialogue with other characters changes: he asks them if they know anything about people’s brains being stolen and things like that.

But after a while, Raz started saying things that suggested that he had observed changes in Dogen’s behavior that the stolen-brain hypothesis would explain. Thing is, I hadn’t observed these changes. I hadn’t seen Dogen in some time and had no idea where he was. I think I know where he was now: when I finally reached the brain-removal room in Raz’s dreamworld, the one thing that brainless Dogen said was “TV”. There’s a TV room in the main cabin, but I hadn’t visited it lately because there wasn’t much to do there. When I finally got back there (after a long diversion involving a monster lungfish), it was empty, and Raz said something along the lines of “Finally, no one is here! I can watch whatever I want!” Now, I had never seen anyone else in the TV room, and no one had ever objected to my changing the channel. So it really seems like I was supposed to have gone back there and seen Dogen, and possibly other debrained children, watching TV and drooling or something.

If I’m right about all this, and I’ve been breaking the sense of the story by not doing things in the right order, it’s kind of surprising: Tim Schafer comes from adventure games, and keeping the plot sensible regardless of player choices is a big part of adventure game design. But here I am, circumventing important plot points without trying. In fact, I’ve been at some pains to cooperate with the designers and do what they want me to do.

That may in fact be my problem, because the game sends mixed messages. On the one hand, Raz seems to regard the rescue of Dogen in the dream tower to be his top priority, and urgent to boot. It’s a false urgency, because the the game has no real time limits, but it seemed like a sign that this was what I was supposed to focus on next in order to keep the game going the way the designers intended. On the other hand, each step in the plot also changes all the incedental stuff: the kids are all in new places doing new things and have new dialogue, the bulletin boards have new notices, etc. (I didn’t even notice that the boards were readable at first, so I missed a couple of chapters worth of notices.) Most of this is just humor not related to the main story — there’s a fairly complicated pre-teen soap opera going on if you feel like paying attention to it — but some of it’s important to Raz’s investigations, and you’ll miss most of it if you actually go directly where Raz tells you to go instead of wandering around and visiting every location again every time you exit a dreamworld. I suppose I thought I’d always have the opportunity to stumble across relevant scenes involving the other kids on the campground, but at the point I’ve reached in the game, all the other kids are gone.

I may replay from the start at some point just to get a more thorough look at everything, maybe after I’ve finished the game and can understand all the foreshadowing. It’s an enjoyable game to play, and this would also allow me to see the earlier scenes at their intended framerate.

So, I’ve been seeing slow framerates on my machine under multiple games lately: first Myst V, then certain brief bits of Lego Star Wars 2 (which was generally pretty smooth, presumably because its style allows for a pretty simple world model), and now large sections of Psychonauts. Worse, I even had speed problems in Eternal Daughter, a retro sprite-based 2D platformer which I tried when it was featured on Play This Thing. Clearly something had to be done.

The problem is, when games are slow, it’s hard to figure out why. It could be either the CPU or the graphics card, and unless you have multiple machines, it’s impossible to test them individually. So I asked myself: If I bought a replacement for one of these things, and the replacement didn’t speed up these games, which one would I regard as less of a waste? And there was a clear answer to that: the graphics card.

See, my nVidia GeForce 7800 GTX card was giving me problems anyway. The GeForce drivers for Windows XP have a known bug that affects 3D objects displayed on a 2D bitmap background: frequently, random patches of the background will cover up any 3D elements. It’s really a quite striking visual effect, reminiscent of the scenes in Labyrinth and Indiana Jones and the Last Crusade where things that look like background turn out to be foreground, except not done on purpose. I wish I could post a screenshot that captures it, but (a) to get the full effect you really have to see it in motion and (b) to reproduce it I’d have to reinstall the GeForce card. Anyway, it makes several graphic adventures on the Stack unplayable. It’s been a a few years now since I first encountered the problem, and nVidia has issued multiple driver updates, but the problem remains. I’ve contemplated dealing with it by installing Windows 98, which uses different drivers that don’t share the problem, but switching to a non-nVidia card will spare me the trouble.

So, Psychonauts runs perfectly smoothly on my new Radeon, and I confidently expect to start blogging about Myst V soon. The new card has larger numbers associated with it than the old one, but I don’t want to say that that’s the reason for the the performance increase, as the old one was released after most of the games that gave it problems. More likely there was some sort of hardware malfunction. In researching the problem, I saw one person on a support forum who suffered abysmal framerate simply because the graphics card’s cooling fan wasn’t spinning. That wasn’t the case here (I checked), but it could easily be some similar failure.

Psychonauts is one of those games that I’d heard people raving about. And yet somehow it seemed to hit the bargain bins pretty quickly, and is already on Gametap. I don’t know why. It might have something to do with the character designs, which are caricatured to the point of grotesqueness. As always, this is something you get used to over the course of play — by now, I’m actually able to see the protagonist, Raz, as handsome and well-formed ¹ Raz actually kind of reminds me of a young Michael J. Fox, but I think that’s more a matter of his earnest manner than his physical appearance. , because he’s really one of the least distorted characters in the game. But seen for the first time, he’s a spindly hypercephalic freak, and I can see that turning off some potential buyers.

The premise provides some reason for making the characters freakish: it emphasizes the fact that they’re not normal. The whole thing is set at a summer camp which is really a secret training ground for tomorrow’s elite psychic warriors. It’s kind of like Harry Potter: kids behaving like kids while at the same time displaying abnormal powers, adults who have mastered those powers giving classes, some kind of secret plot unfolding in the background that only the child hero can unravel.

Structurally, it’s a lot like the Harry Potter videogames, too: you’ve got a largish school/campground hub area containing secrets and collectibles, and various challenge areas accessible from it. The challenge areas in this case are the abstract worlds inside people’s minds. The first such world, in the mind of a drill-sergeant-wannabe coach, is war-themed, by which I mean that the Ditkoesque floating platforms are covered in barbed wire and concrete bunkers and fragmentary bomber fuselages. Interestingly, this level doesn’t contain any actual fighting: war is presented, not as a situation in which you have the opportunity to triumph over an enemy, but merely as a situation of constant danger.

Later scenes do contain combat, but the game is basically a platformer — a fact that must have come as some surprise to fans of Tim Schafer, the creative lead, who’s best known for his work on graphic adventures. (Psychonauts has some adventure-game elements, but what doesn’t these days?) The sense of humor, though, definitely hearkens back to Monkey Island. I’ll note in particular one bit where an unpopular kid with an incongruously deep voice tells a humorously long and boring story: “…Then we went up a hill. Then we walked four miles. Then we walked two miles. Then we walked three miles. Then we walked half a mile. Then we made a U-turn. Then we stood still for a while…” Past a certain point, it’s constructed from randomly-selected phrases, but they’re still delivered with perfect comic timing. That means some coder took the trouble to tweak the delay in a loop for maximum comic effect.

Alas, I’m probably missing similar timing elsewhere. Much of the dialogue in the game occurs in the hub areas, where you can talk to the other kids or show them inventory items or try to set them on fire with your mind or whatever, and in those areas I’m suffering from framerate problems. Presumably this is because the campground is harder to render than the mental worlds, being more open and more detailed. It’s not so bad as to make me stop playing, but characters’ mouths often become noticeably out-of-sync with what they’re saying, and the audio playback pauses to let the video catch up. I wish I knew where the bottleneck is here — whether it’s the CPU or the video card or what. All I know is that altering the settings in the in-game video options panel doesn’t seem to help.

Outside of the cities, there are free-standing buildings scattered throughout the map. Some of them, such as the training camps that improve the stats of visiting heroes, are intrinsically neutral, and will serve heroes on either side. Others can be claimed. A banner on the side indicates by its color the owner of such a structure, and typically any enemy hero passing by can change the color of that banner by tagging it. It’s possible to leave troops behind as guards to prevent this, but this is seldom if ever worthwhile, as it leaves fewer units in the hands of your heroes.

Some of the claimable structures, such as mines and sawmills, provide resources to their owner every turn, and it’s very important to claim as many of these as possible — even if you don’t need the resources, it keeps them out of enemy hands. Others produce troops on a weekly basis, provided that a hero comes along to pick them up. It seems a lot less important who owns these; it just matters who can reach it, tag it, and recruit the troops first every week.

But on level 3 of Conquest of the Underworld provides a motivation to claim everything you can, even things you have no intention of using. It has to do with the “fog of war”. Every level I’ve seen so far starts out with most of the map dark, revealing it through exploration, as is typical in these games. But in levels 1 and 2, once an area was explored, it stayed explored. In level 3, the only things you can see on the map are the areas that can be seen by your heroes or from your buildings. Dark areas aren’t even considered as navigable by the game’s pathfinding algorithm if you’ve seen them before and know there’s a way through. So keeping those roadside towers under your control is an important convenience, even if you have no intention of ever recruiting what they produce. (Heroes can only have so many creature stacks under their control at a time, so recruiting everything you see isn’t always an option.) And when an enemy hero slips through, he brings darkness.

I don’t remember episode 1 well enough to know if this returning darkness is the state of most levels or if it’s a special feature of the underworld. It shows the game’s age somewhat, though. More recent games with fog-of-war effects tend to have three states: not just “visible” and “not visible”, but “visible”, “not currently visible but explored”, and “unexplored”.

I’m back to level 3 of Conquest of the Underworld which is where I was when I decided to start over. So far, it’s proved pointless. My main heroes (Tarnum is allowed to take two other heroes with him between levels) all have Earth Magic and the ability to learn the Town Portal spell, but the spell hasn’t been offered yet.

Heroes mainly learn spells from the mage guilds that you build in your cities. Each mage guild, when built, gets a random assortment of spells. I wonder how random it is? It might be possible to repeatedly load a saved game and rebuild the mage guild until it gives you the spells you want. Which would be cheap. But putting a die roll in the way of crucial permanent effects encourages cheap behavior. That’s why most CRPGs these days don’t randomize hit point gains from levelling: to eliminate the temptation to quit without saving when you don’t get enough.

Somehow, even though I knew the second level better this time, it took me longer to finish. Consequently, I got to see some plot events that I had missed the first time around — events in the form of narration in a dialog box. Some events are just color text (as when Tarnum gets a letter from Queen Allison, his boss for this adventure, asking about his progress), some have effects on gameplay (as when Allison’s letter is accompanied by funds for recruiting more troops). But past a certain point, you know they can’t be essential to the story: any sufficiently skilled player will miss them.

So, I decided to start over from the beginning in Conquest of the Underworld. To explain why, I’ll have to desribe the game mechanics a little.

In the Heroes of Might and Magic system, everything that you do outside of the cities you control is done through entities called “heroes”. Heroes have various stats and skills, can wear magic items to affect those stats and skills, and improve with experience like a character in an RPG, but they don’t participate in combat directly. Instead, they gather troops and creatures under their command.

Creatures are produced every seven turns — every week of gametime — in cities and certain free-standing structures. But they can’t do anything on their own. Any creatures that you recruit in a city will simply wait there and defend it until a hero scoops them up and carries them away. It’s useful to think of them as wargaming minis that the heroes carry around in a box.

Now, the kind and number of creatures that a city can produce is determined by the structures that have been built there. For example, in order to produce Griffins, a city must have a Griffin Tower. Turn it into an Upgraded Griffin Tower and you can produce Royal Griffins, which are stronger and faster. Build a Griffin Bastion as well and you can get an additional three griffins per week. But it takes time and money to build these things. By the time a city can produce the best units in any quantity, the action has moved far away. Getting your troops to the front as quickly as possible is a big part of the strategy of the game.

One of the basic techniques is to use a “bucket brigade” — a string of heroes stretching from the city to the front, positioned a day’s ride away from each other, ready to pass that box of minis all the way from the castle to the hero who needs them. In this way, troops can travel arbitrary distances in a single turn. But hiring heroes costs money, even if you’re going to just use them as delivery boys, and setting up a bucket brigade takes some time and effort, especially when you suddenly have to shift it to point in a different direction.

Another thing that helps a lot is that some cities — the “inferno” types, the ones that produce demons and hellhounds and the like — can build Castle Gates. Heroes can travel instantly between any two cities with Castle Gates. But these are expensive and not likely to get built until late, and, as I said, only help for transporting troops between Inferno cities. While this scenario has more Inferno cities than may be regarded as typical, there are also an awful lot of Castle (knight-type troops) and Necropolis (undead-types).

What I have just learned is that there’s a spell called Town Portal that could help a lot. In its basic form, it teleports a hero to the nearest friendly city, but a hero with Advanced Earth Magic can use it to teleport to any friendly city. Thus, it is really useful to have the Earth Magic skill. And there’s a seer back on level 1 who teaches this skill, but I ignored it, because you have a limited number of skill slots and I thought I had more important things to learn.

Well, now I know better.

It strikes me that Heroes of Might and Magic (as revealed through the Heroes Chronicles) has a balance problem. Not that it’s unbalanced exactly, but that it loses its balance easily. It’s a very high positive-feedback game, which is to say, power is rewarded with more power, so the winners tend to keep on winning and the losers tend to keep on losing. The outcome of a scenario rests on the first few turns. If you can pull ahead then, there’s no stopping you. But there’s no stopping the scenario, either: even if victory is assured, you have to keep playing it out to get credit for it.

This is especially visible in the second level of Conquest of the Underworld. The goal in this scenario isn’t to wipe out all enemies and conquer the map, but to obtain a certain artifact that’s at the end of a sequence of map-spanning fetch-quests. Wiping out all enemies and conquering the map does, however, make the questing much easier. In fact, I find that the easiest way to approach the level is to concentrate on securing the terrain first, hitting the enemy castles while they’re still weak, and not go out of your way to cart plot tokens around until the conquest is complete. This probably isn’t the approach that the level designer had in mind. At least, I hope not, because it’s kind of boring: it leaves you with a bunch of time-consuming tasks to pursue after you’ve removed all challenge.

I suppose this is an example of one of the classic injustices of game design: if there are two ways of accomplishing something, one that’s difficult and interesting and one that’s easy and boring, players will choose the boring way and then blame the designer. But in this instance, I’m not really sure what the other option is. Some of the quests in the chain involved finding particular artifacts, with no clues to their locations. This isn’t something you can really pursue. All you can do is peek in on any ruins you pass by in the hope of lucking out, which is something you’d be doing anyway.

Somewhat embarassingly, I haven’t made it past level one of Conquest of the Underworld yet. I could blame this on my stubborn insistence on playing it on the Hard difficulty setting — there are five difficulty levels, which to me means that difficulty level 3 is Medium, darn it, regardless of what the game calls it. But really it’s more about making mistakes.

Heroes of Might and Magic is one of those two-tiered games: specific battles take place in a tactical combat mode, which is mainly about deciding who should attack what, and the battles are embedded in a larger strategy mode, which is mainly about deciding what to build when. The limited resources you need to supply your army are often guarded by monsters, and the enemy warlords are competing with you to reach them first. So there are two fundamental mistakes you can make: going for the treasures before you’re ready and getting killed, and waiting too long and falling behind the competition. I’m generally more prone to the latter mistake — part of being a completist is wanting to have all the buildings built and all the creature types available — but today, I’ve been making both.

And it isn’t just fundamental mistakes, either. It’s stupid litle things. Like going after heavily-guarded things that I don’t really need, and not saving often enough. I’ve never really decided what level of saving is appropriate for a strategy game. Doing it every turn seems kind of cheap, like you’re trying to bring it all down to tactical decisions instead of strategic ones in order to avoid the consequences of not really knowing what you’re doing.

But hey, maybe I really don’t know what I’m doing. In which case I should really dial it down to Normal difficulty. That’s how I passed the first few levels of Warriors of the Wasteland, when I really really didn’t know what I was doing, due to complete inexperience with the game.