The Mystery of the Nautilus

2002 T-bot Interactive
Designed by Benoit Attinost, Manuel Ruiz Dupont; Laurent Le Besnerais
Reviewed 2002 July 22

Rating -4 Linearity wide, segmented
Reasonability silly Connectivity moderate
Difficulty pedestrian Relevance strong
Interface 1st 360 simple Real-time occasional

You are a guest scientist aboard a navy submarine. When a large metal object is detected, you disobey orders and take a minisub out to investigate. You find Captain Nemo's abandoned Nautilus. Everything still seems to work, but you soon discover that the ship is controlled by a psycho artificial intelligence. You have to sidestep the AI's murderous plots, and find a way off the ghost ship.

There's not much more to the story than that, except for a little background on the original abandoning of the ship. You find some notes and hologram messages left by Nemo. It's really all about you escaping the clutches of a mad ship.

Everything is set up for a good adventure, and the designers' heart is in the right place: there are lots of challenges, most are contextual, and there's a nice non-linear layout, but with enough chapters (called stages) and triggers to effect a nicely flowing plot.

The game falls down, however, in the reasoning behind the challenges. It's not that the challenges are wildly silly, but that there's no way your character could know that certain goals are to be achieved. For example, at one point there's this locked cabinet, and the challenge is to break into it. But you don't know what's inside, and there are many other cabinets around. Why are we breaking into this cabinet? My character magically knows that there's something important in there. My character frequently seemed to know what was around the next corner, and what preparations had to be made. You don't play this game, it plays you.

The challenges generally aren't that difficult. Things that go together, you put together, with unpredictably fortuitous result. Often, your character will give you out-of-character hints, telling you things he couldn't possibly know. There are a few challenges that require you to simply fiddle until you stumble on the right result; realistic, maybe, but not much fun.

A major aspect of the challenge design is the frequent use of timed sequences. When the clock runs down, you die. A couple of the sequences are very short, amounting to instant death, but most are very long, up to about 10 minutes. The deadly short ones are when you go somewhere you're not supposed to go, and the time is only supposed to allow you to retreat. There's even an instance where a shorter constraint is contained within a longer one. It's not arcade -- there's no problem accomplishing the necessary tasks within the time constraints if you know what to do, although you can't doddle. You can also save within them, but there's no auto-restore. What disappointed me most, however, was that all deaths went immediately to a generic "you're dead" screen, without any little idiosyncratic death animations or effects.

The punchline is that the most intense part of the game is the final stage, where there's no time constraint, just effective music.

Somebody went to a lot of effort designing the Nautilus, making it a beautiful piece of industrial age equipment. Somebody (else, I would think) then went and hid that lovely work behind dark, blurry rendering. The real challenge to Nautilus is pixel hunting. An object can be right in front of you, and reasonably large, and yet still be difficult to see, just a slight variation in the general smudginess. The smaller objects are almost invisible. This problem is compounded by objects tending to have very small hot-spots. If you sweep your cursor over an object, the cursor won't change -- you have to sweep slow enough to give it time to change. Even if you know the object is there, it can be difficult trying to activate it, since the cursor hotspot is also in an awkward position in the middle of the icon. Just to rub it in, objects are often "hidden" overhead and at your feet. Remember that when your fighting those time constraints.

The objects themselves are often indistinct blobs. I soon learned to immediately dump objects into the inventory, them look at them there -- they're still indistinct blobs, but it shows a text label that usually gives you enough info to figure out the relevant object properties. I still had problems: pieces of rubber, they look like black blobs, but are they solid chunks, bands, pipes, etc.? There really should be an look function that would give proper descriptions of objects, inventoried and environmental.

The engine is the standard omni-pan at node, but the central activity area (i.e., the part of your view that does not activate panning when your mouse is there) is frustratingly small. This causes lots of unintentional panning when you're trying to fiddle with the environment. Worse, they placed the inventory and your PDA (save/restore/quit, amongst other operations) in the panning area, so that every time you want to use them you cause the room to spin widly. The same happens again when you return from PDA screens.

For some strange reason, they put some metal framework along the edges of your view on one side. Along with the overlarge inventory and PDA, the view is cluttered. There's no excuse for it, with this kind of game engine.

It's amazing that somebody could come up with such a bad interface at current levels of technology. The designers are surely on the most-wanted list for crimes against mimesis. There isn't enough story to be an attraction on its own. This all leaves nothing to recommend -- avoid this game.


Beware! Here are some spoiler-ridden notes on the game. They're only recommended for people who have played the game and want to see some of my rationale for my evaluations.
David Tanguay's Game Reviews
Here's a description of all the gobbledygook in these reviews. It's also a bit of an essay on the nature of adventure games.