Review written by Andrew Plotkin
So that's how it's been released. Which is why you might see this game as Agon 2. The original trio of episodes are now packaged as Agon: The Mysterious Codex.
That all being clear...
I enjoyed the original episodes, and I was disappointed with this one. That's the whole review, and I don't want to spend too many pages hashing over depressing details, so I'll start with a positive note: Lost Sword has a fine plot. You interact with several characters; they have stories; they have relationships with each other (if not a lot of on-screen interaction). You go through stages of discovery and exploration, the plot develops, you go back to discover more about where you've been, and it all winds up with a chase scene, a dramatic escape, and a happy ending.
All these things are nicely framed into the game mechanics. It's not a barren puzzle-fest; the puzzles and locks fit naturally in the story, as do the clues you discover. Nor does the game devolve into a series of fetch quests. You have to talk to a lot of characters, and you do some of them favors, but by and large these interactions don't feel like artificial obstacles. And they're interleaved with puzzle-y, but realistic, interactions -- devices and challenges that make sense in the environment (even without the "everyone is a puzzle fiend" convention of adventure gaming).
So why am I disappointed?
Pacing. And pacing and pacing. The first thing you do in the game is talk to someone. The second thing you do is also talk to someone. And then to another person. Then you solve one (very simple) puzzle, and then you get to talk to someone else. Each of these people has opinions about every other person in the game, plus various plot points. The dialogue menus just go on and on.
Now, this part of the game is fairly broad -- several avenues of exploration open up early, including both puzzles and character interaction. The walkthrough I found runs in a different order than I did. But if you're the kind of player who likes to nail down all the easy stuff before attacking the puzzles -- and I am -- then you will sit through a good hour of spoken dialogue before the game gets moving.
(This, by the way, is a wonderful example of why giving your players a choice does not solve your problem. "The players who like dialogue will go through all the dialogue, and the players who like puzzles will skip it and head for the puzzles!" No, wrong person, you are wrong. I went through all the dialogue because I figured there were clues in it, and because I am a completist package-hunting nerd. Also I didn't imagine that the game pacing would be improved by skipping stuff, because, well, why would anything in the game be a waste of time? Well, there were clues in there, and the dialogue wasn't a waste of time exactly -- it all existed to set up a detailed story. But I think it could have been tightened way the heck up. And mixed in better with you actually doing stuff.)
(At least the voice acting was good. And the accents. I give thanks for these small things, because I know how much I complain when they're taken away.)
Another disappointment: three of the big puzzles.
Sound puzzles are always risky, because players have such widely different capabilities. Some players, for example, are unable to hear sound. Others have the sound turned off (for any number of reasons that are not your business as a game designer). These are big problems for sound puzzles, and have been since Myst.
But even for the majority of players that do hear game audio, distinguishing sounds is a gamble. Remembering the sequence of high-screechy-groan, low-whining-hiss, scrapy-bell-rattle is not easy for everybody -- and writing it down on your notepad is harder than you might think. If you thought about it at all. What about melodies? I'll tell you, I'm weirdly bad at remembering them (and I sang in high school choir for four years).
In Lost Sword, we have guitar melodies. You hear one; then you go elsewhere in the game and listen to several, one at a time. Pick out the one you heard. Whoops, it's not one of them, go back and try some more. You can recheck the original at any time... give or take a walk across Toledo... but nothing washes a recently-heard tune out of my head faster than hearing three newer tunes in a row. So this was just really hard for me.
And the puzzle is tuned (no pun intended) to discourage guessing. You aren't just trying to pick the right answer off a list; you have to look up some stuff about the answer, figure out how it fits in, and then guess a little bit more. This is actually a fine example of how to design a puzzle to prevent back-solving. Unfortunately, I was guessing about the melody -- simply because I wasn't confident about my ears -- and therefore the whole thing was impenetrable. Advantage: walkthrough.
And then they went back and did exactly the same thing with a visual puzzle. Scattered around the city are various numbered images. By the time I saw two of them, I knew they were puzzle clues, so I copied them all down on paper. And when I got to the puzzle that used them... my notes were useless, because I hadn't copied them obsessively enough. (They're fiddly, ornate, decorative patterns. Sometimes seen from a distance. This isn't a Rhem symbol-puzzle, where the symbols are artificial and ugly but easy to notate exactly.)
Again, the puzzle was designed to discourage guessing. There were a few possible ways to interpret the sequence, so you had to be confident about the images and then work through the variations. Again, I was unconfident about the images. Could I have gone back through the city and redrawn all my notes? Sure. I could also have looked in the walkthrough. Guess which I did.
Third example: developing a photograph. I'm sure I have played a simple, enjoyable adventure-game scene in which you go into a darkroom and develop a photograph. I don't remember which game it was. (My review collection only mentions Dark Fall, which handled it awkwardly.)
But the canonical develop-a-photograph scene gives you pans of developer, stopper, and fixer. Or maybe bottles of those stuffs plus a pan. You dip your print three times in the right order, no doubt with pointed nudging from your voiceover. The game doesn't let you screw it up. Because this isn't a puzzle, except in the broad sense of "something you have to do to win". You are not figuring out how to develop a photograph; you are enjoying the experience of successfully developing a photograph.
Or, you could play it the Lost Sword way. Giant rack of reagents. Instructions on how to mix each processing chemical, one scoop at a time. Seven mixtures, dealing with both the negative and the print sheet. Some require you to pull the photo after the right number of real-time seconds. The rules are are unrealistically strict in some places (I screwed up one run by using distilled water instead of tap water), and unrealistically lax in others (you have to turn off the lights to remove the unfixed negative from its case, but then you can carry it around outdoors.) You can make mistakes -- you can't make the game unwinnable, but you can definitely blow a print and have to start a new one.
(I think I found a bug that can make the game unwinnable, actually, but I didn't have the energy to prove it.)
This is an excellent simulation of the experience of early photography. It does not belong in this game. Getting it right requires persistence without brainpower. Screwing up is tedious and unenlightening. Restarting from scratch is painful.
I saved after every step, and I still looked at a walkthrough after my first inexplicable failure. (Turned out to be the distilled-water thing.)
So that's a lot of major problems. I had minor problems too. Like the way the game kept crashing when I looked through the musical scoresheets. Or the mysteriously-appearing-hotspot problem in many puzzles. Or the cursor bug that people still get wrong even though mouse cursors have been changing shape for twenty years now. (Short form: if the menu changes under the cursor, I shouldn't have to wiggle before clicking. Particularly when I spend fifteen minutes clicking Menu Item One repeatedly with no other interaction. See "dialogue menus go on and on", above.)
Lost Sword does, I repeat, have good aspects. I even mentioned some of them, which makes this a more balanced mini-review than usual. Could I mention more? Probably -- but when a game sours me early on, I stop enjoying the good points. That is, I appreciate the art (Lost Sword has decent scenery), but I'm not putting much effort into the puzzles any more, so I don't get much out of them.
This is a pity, because the final puzzle was an interesting blend of adventure puzzle and environmental puzzle, of a sort I've never seen before. I wish I hadn't been predisposed against it, because I really couldn't tell if it was insufficiently reactive or if I just wasn't trying hard enough.
As usual for an Agon game, Lost Sword ends with a board game. As usual, it's a square grid on which two armies alternate moves to capture each other. I'm tired of those. Go play Blokus or Transamerica and take notes, people.
This board game was actually less of a game and more of a jumping-pegs puzzle. The initial setup forces you to sacrifice your first couple of pieces, so you wind up having to play near-perfectly to catch up. It's not exactly like a jumping-pegs puzzle, since the enemy makes moves -- but you're working around a medium-dumb AI, so it's the same feel. (I'm pretty sure that two humans playing this game would always stalemate.) But heck, a jumping-pegs variation that I've never played -- I enjoyed it.
Overall: I think the developers sketched out a game design and then jumped into it, without ever considering the balance and pacing. There were good puzzles and good interactions and a good storyline; but a few obtrusively weighty puzzles loom like boulders out of a tide of dialogue, leaving the game's virtues drowned. Or at least seriously damp.