The Stack

I seem to be getting into the home stretch now. The overall structure of the game involves five castles, a central one on a mountaintop and one in each of the cardinal directions around it, connected by a ring road. I’ve done missions in all of the castles except the central one now, and the only quest currently active is to activate the last remaining teleport gate of the eight on the ring road. This will take me back to near my starting point.

My characters are also nearly all up to the point where they can wield the strongest weapons available for purchase. And that means it’s time to sit down and do some serious crafting.

Crafting in this game involves combining items with monster leavings, such as oni horns and kappa claws, or various kinds of gem. Each item has specific effects: dragon stingers make a weapon poisonous, tengu feathers give a bonus to dexterity, etc. Each weapon and piece of armor has a certain number of enhancement slots; the more powerful items tend to have more. It’s kind of like the Final Fantasy 7 “materia” system, except that enhancement items can’t be removed. To effect an enhancement, you have to pay a blacksmith, who charges more for more powerful effects. More powerful effects also take more time to produce, and while the blacksmith is working on an item, he can’t do anything else, like start on new items or repair damage to the stuff you’re already using.

So there are four limiting factors on how powerful an item you can craft: enhancement items, enhancement slots, money, and time. Money was the chief limitation at the beginning of the game — at least, it was after I had spent it all crafting a wicked bow for my Archer. But for most of the game, it’s been enhancement items. I have lots of certain items, such Elder Kappa Shells, but not the things I really want to use. Shortages trigger the hoarding instinct: rather than use my last precious gems on a suit of armor that I’m just going to throw away once I can wear something stronger, I’ve been saving them up.

And now comes the moment I’ve been saving them up for. There’s no need to wait for a better item to use them on. It’s time to make maximally amazing stuff for all seven characters. Which means spending maximal amounts of money and time. The money isn’t a problem (at least, I don’t think it is), but the time factor means that there’s probably more than an hour of solid crafting built up. I hope it’ll all be ready by the time I get sent to the central castle.

Since my last post, I’ve got both the Berserker and the Ninja up to the same level as the rest of the party and completed the Berserker’s special quest. It turns out that power-levelling characters gets easier as you go along, because there’s a positive feedback loop: the more advanced the charcter is, the more often he succeeds in hurting things and getting XP. The Wizard is still lagging, though, because he’s spending half his time dead. Resurrection is essentially free in this game, but there’s a limit to how frequently you can do it.

Fortunately, the Wizard is the one character that can be effectively twinked at low levels. To explain why, I’ll have to describe the magic system.

Every character class, even the Brick, has its own set of forty magical abilities that can be purchased and enhanced with “spell points”. These abilities are divided into four groups, corresponding to the four elements but with “lightning” substituted for air; spell points are specific to an element. Some of these abilities are spells that have to be cast and last a short time, some are continuous status effects that last as long as they’re selected, some are permanent enhancements that are always active. The list varies from character to character. Obviously the Wizard has access to the most powerful spells, but they’re only powerful if they’ve been enhanced with spell points. For example, there’s a basic missile spell for each element that starts out doing 1-5 points of damage. After you invest a maximum of 10 spell points in it, it does 5-55 points.

Clearly, spell points make a big difference. Assuming that you don’t try to make him use a sword or something, the Wizard’s capacity for dealing damage is limited mainly by the number of spell points he has to spend, rather than by his experience level, stats, or equipment. So, how do you get spell points? Ordinarily, you get a spell point for each of the elements when you gain a level, but you can get bonus spell points by sacrificing magic items. Each item is worth a fraction of a spell point, and they build up.

At the point I’m at in the game, there is an abundance of magic items that I don’t want to keep, mainly armor and weapons that are too heavy for the Wizard and too wimpy for everyone else. This is how I know that the Wizard’s missile spell caps out at 10, even though I’ve only levelled him five times.

I said earlier that there were twelve levels in this game. This is incorrect. There are twelve main quests. The first quest is simply to escape the castle where you start the game, so when I left that area and the first of twelve empty boxes filled in, I mistakenly thought that the rest of the quests would be similar: a linear series of endpoint-reaching exercises, as is common in videogames. But it turns out that some quests are given in parallel, with at least one of them spread out over multiple areas.

In addition to the main quests, there seem to be three “single” quests which require a specific character. The first one I was given involved talking to an old friend of the Leader. Only the Leader can do this. The second, which I have not completed, requires the Berserker. An NPC mentions three powerful adversaries who have set a challenge, and the Berserker immediately declares that he accepts the challenge, even if he’s not in the party. Which he wasn’t, because he’s one of the charcters I had decided to ignore. But I can’t actually do the challenge without shifting him into the party.

I haven’t checked the docs on this, but I’m guessing that the single quests are optional. However, I’m too much of a completist to just leave this challenge unattempted.

It’s going to be hard. The party is nearing 20th level at this point, and the Berserker was abandoned at level 1. In most RPGs, this wouldn’t be a big problem: you could just stick the low-level character in the back of the party and tell him to defend himself while the big boys win him some XP. That doesn’t work here, because the party doesn’t share XP at all. In most RPGs, that wouldn’t be a problem either, because you could just take the low-level characters back to the earlier areas and level by means of random encounters. But, as in Diablo, the monster supply here is finite, and I’ve been pretty thorough about clearing areas as I pass through them. There just aren’t any low-level XP farms left. In order to reach a level where he’s capable of completing the quest, the Berserker will simply have to fight a large number of things that are tougher than him.

Well. Can the other characters help, with buffs and twinks and healing? Sadly, there are no healing spells, and no buffs other than self-buffs. They can twink to a certain extent, but the better items have level and/or stat requirements that prevent this. Mainly what they can provide is cash. There’s no reason not to give the Berserker the best equipment he can use and a big stack of healing potions.

Despite difficulties, I have managed to get the Berserker up to level 5. I should bring out the Ninja and Wizard and start levelling them too, just in case I need one of them for the third single quest.

Early in Throne of Darkness, I decided to keep things relatively simple by playing only with the four simplest characters: the Leader, Brick, Archer, and Swordsman. (The special roles of the Ninja and Berserker weren’t obvious, and playing the Wizard would require learning the magic system immediately.) I did try using fewer than the full complement of four, but doing that for very long is risky. You never know when you’re going to find yourself suddenly outnumbered.

When more than one character is active, you directly control one of them at a time, switching between them at will. The ones you’re not controlling at the moment act autonomously, following you around unless they see something that they want to kill, like the dog in Nethack. I suppose the developers were trying to make the single-player campaign more like cooperative multiplayer play, since that’s generally acknowledged to be the best way to play Diablo.

This scheme yields ironic results. To explain: Each character earns experience points independently, and they seem to get them by hurting foes, not necessarily by killing them. So the more often a character can successfully strike a blow, the faster he’ll level. Now, the characters you’re not controlling at a given moment are pretty efficient about finding and attacking foes. They’ll rush at things that aren’t even onscreen yet, leaving the player-controleld character chasing after them. Melee is rapid, and it’s easy to misclick, either by clicking the spot that a moving foe has just vacated or not being able to distinguish friend from foe. These clicks are interpreted as instructions to move. Consequently, in most battles, the player-controlled character spends a lot of time running around aimlessly instead of fighting. Overall, then, the characters that you’re not controlling directly will tend to level faster than the one you are. If only there were a way to just abandon control of everyone, let the computer play the battle for you so that everyone would participate optimally! Which, come to think of it, is the idea behind the Final Fantasy XII “gambit” system..

I mentioned that you control a team of seven charcters in this game. Presumably this number was chosen by a Kurosawa fan. They’re not the same seven samurai as in the movie, though. The character classes in Throne of Darkness are:

The Leader
The Brick
The Archer
The Swordsman
The Wizard
The Ninja
The Berserker

(A ninja isn’t really a kind of samurai, of course, any more than a stoolie is a kind of policeman. Actually, if you read the character bios in the manual, a lot of them turn out to have been raised as farmers and the like, rather than as members of a hereditary military aristocracy. But why bother complaining about historical implausibility in a game with tengus and oni in it?)

In contrast, the character classes in The Seven Samurai are:

The Leader
The Leader’s Assistant
The Leader’s Buddy
The Stone-Cold Killing Machine
The Kid
The Loose Cannon
The Other One

(Like the Ninja, the one I’m calling “The Loose Cannon” isn’t really a samurai. So there’s some kind of balance there.)

This list may not be quite as gameworthy as the first one, but I have a long-standing fantasy about about the ideal game adaptation of The Seven Samurai. The player, as I see it, would choose which samurai to play, and that choice would fundamentally shape what kind of game it is. Playing the Loose Cannon, for example, would make the game more freeform and self-directed, like GTA: you could do your assigned missions, but you could also just wander off by yourself and try to steal the enemy’s guns. If you chose the Killing Machine, it would mainly be an action game. The Leader would be about strategy, the Leader’s Assistant would be about micromanagement. The Kid? Dating sim.

So, I’ve decided to start another game while I decide whether or not spending more time on my joystick problems is worth it. Picking one at random, I find Throne of Darkness, a Samurai-themed Diablo imitation from 2001.

Man. I haven’t thought about this game in ages.

I think I bought it mainly because it was cheap and the screenshots on the box looked appealing. I played it briefly when I first got it, but it turned out to be more complicated than I was in the mood for just then.

Installation Annoyance #1: It has two disks, and you have to swap them three times before you can start playing.

Installation Annoyance #2: Throws a silly error, even when patched. Finding a site that would let me download the patch without the hassle of “becoming a member” was frustrating enough (and I’ll save anyone else who reads this the frustration by offering my copy here), but discovering afterward that it didn’t help? Bleah. I eventually found help on a third-party forum. It turns out that the solution is to overwrite the game’s copy of msvcrt.dll with the one from WINDOWS/System32.

First session: The mechanics are pretty complicated, but I haven’t yet reached the point where I have to understand what I’m doing in order to make progress. Hopefully I’ll be able to learn and apply things as I go along. Like I said, it’s Diablo-like, but you have a team of seven characters with different abilities, of which four can be active at a time. It seems like it might be a valid strategy to play with just one or two characters at a time so they can level faster.

I still haven’t got the right joystick to work correctly with GTA3, and I’m on the verge of giving up. There’s a part of the registry under HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\MediaProperties\PrivateProperties\Joystick\ that clearly corresponds to my gamepad. Indeed, I can disable rumble effects by deleting certain keys under it, so it’s not like I’ve been completely unable to affect the way the joystick operates. According to various websites, including Microsoft’s joystick driver specs, I should be able to remap the axes by altering the “Attributes” value of the various sub-keys under “Axes”. Nothing I have done affects the axes at all in any way other than miscalibrating them. Like I say, I’ve almost given up, but I’d really like to get this working right, not just for GTA3, but for all the other games in the stack that don’t have in-game axis reconfiguration.

Anyway, in the process of testing my alterations in GTA3, I’ve noodled around in cars a bit. I’m beginning to see why this game was so popular. This is a very different game from the first two. The switch from top-down fixed camera angle to a more street-level view has a greater effect on the experience than I thought it would, mainly that it gives a better sense of motion, that you’re careening along the street and onto the crowded sidewalk and so forth. It also has a different feel from its imitators, such as Jak 2, which tend to be set in more fantastical environments. Liberty City is based on New York City. I live in New York City. The sites in this game — the decaying tenements, the tiny fenced-in parks, the storefronts jammed into grey concrete — are familiar to me, and modelled well enough to really evoke the real thing. I can’t explain why it’s enjoyable to play with an imitation of someting that I could see just by walking around outside, but it is.

Surely, Grand Theft Auto 3 is one of the games that any game-literate person must know, one of the defining games of this decade. Not only has it been tremendously influential to the industry, it’s controversial enough to have become one of the few games that even non-gamers have heard of. It’s even been satirized in a soda commercial. Strange to think that it’s taken me this long to get around to playing it.

My reasons for not playing it yet are not good ones. They stem from my completist leanings: I don’t like to play series out of order if I can help it. Thus, I didn’t want to start GTA3 until I had finished GTA2, even though there’s no continuity of story or anything like that. And it took me a while to get around to playing through GTA2 simply because it wasn’t all that good. Its faction system was an interesting experiment, but it encouraged somewhat tedious gameplay. The easiest way to complete many of the missions was to pacify the gang whose turf you’d be invading in advance, which you could do by killing your unresisting allies in the target gang’s rival gang. Still, I finished GTA2 a few months ago, and then took a months-long break from the series.

Even now, I haven’t really made a serious go of it. I’m having some difficulty getting it to work properly with my joystick, a PS2 Dualshock controller connected to my PC via a PSX-to-USB adaptor from Radio Shack. The problems I’m having are problems I’ve had before: the right analog stick seems to have its axes swapped, so that pressing forward and back rotates the camera and pressing left and right zooms in and out. Various websites suggest registry hacks to fix this, but nothing has worked yet. I suppose I could just go to keyboard/mouse controls, but that just seems wrong for something that’s primarily a PS2 game.

I reinstalled The Neverhood and played through a bit of it to see if my earlier comments were at all accurate. If anything, I understated things. I called the look “handmade”, but I didn’t specify that the scenery had finger gouges all over the place.

I also mentioned the impression of three-dimensionality. This goes way beyond the look of the graphics. The Neverhood goes to great pains to give an impression that the gameworld is a single continuous physical object, using the tricks employed by graphical adventures from Myst onward. You get maps, glimpses through windows of distant locations that you’ll visit later, puzzles based on adjacency of locations you can’t walk between directly, and physical manipulation of large landscape features to alter what locations are accessible. Presumably many of the backdrops were assembled and photographed individually, but some of the exterior scenes had to have been done by moving a camera around inside a large clay model. Skullmonkeys, by contrast, is clearly a disjointed series of levels assembled out of sprites. How disjointed? Travelling from level to level involves jumping into a “warp gate”. Even though some of the level graphics in Skullmonkeys are quite attractive (particularly the “Castle de los Muertos” level, which involves running across battlements in silhouette against a red sky), I have to call The Neverhood’s overall approach more impressive.

The one area where Skullmonkeys really beats Neverhood is in its framerate. The Neverhood‘s animation looks unbelievably clunky after playing Skullmonkeys for a while. It doesn’t take long to get used to it, though.

So, let’s talk story. You’ve got this nameless city, ruled with an iron fist by one Thomas Malik, who keeps the city in an illusion of perpetual daylight. To sustain his magics, he secretly abducts travellers and extracts their souls, rendering them insane. He earnestly believes that everything he does is for the best. Opposing him is a Gnostic sect led by one Evaine, who might possibly be heir to the dynasty of Queens who ruled the city long ago, although I never found any definitive confirmation of this, and strongly suspect that there is none to be found. Malik’s enforcers are hunting for Evaine on the grounds that she’s a rebel and a terrorist, but in the end you learn that the two of them have some personal history together, which makes it seem like the whole city is in the grip of some kind of twisted lover’s spat or something. In a way, it reminds me of Aeon Flux.

As with most heavily plot-based adventures, reaching the end is not difficult: if you just keep plowing ahead wherever the plot guides you, important events will keep occurring until you reach the last important event, which is the ending. But it’s striking that you don’t have to do it this way. Usually, “plot-based” means highly linear and skimpy on simulation, both of which stem from an author’s decision to give priority to their one preimagined storyline over player choices, but that is emphatically not the case here. For example, there’s a point in the story when Malik asks for the player’s help in finding Evaine. In my first play-through, I took him up on it. I suspected that Malik was the bad guy by then, but accepting missions is what you do in these games.  The option to refuse a mission is, in nearly all games, a fake choice; if you choose it, the game will either find an excuse to force the issue or just end. Playing from the beginning a second time, I tried putting up a sterner resistance to Malik in order to see what would happen, and was surprised that he let me go. If I didn’t want to help him, I could just walk away and try to cope with the mysteries of the city on my own. It did not break the plot, although the decision would certainly have consequences.

The only part where player agency really goes away is at the very end, which is a little ironic, because the solution to the final puzzle is precisely to assert your agency, to refuse to be railroaded into doing what the game tells you to do.