The Atomic Heart by Stefan Blixt [Comp03]

IFDB page: The Atomic Heart
Final placement: 10th place (of 30) in the 2003 Interactive Fiction Competition

Comp03 is starting to feel like Old Home Week. First there was the return of Mikko Vuorinen, and now here’s a brand new game from Stefan Blixt, author of Comp97’s Pintown and Comp98’s Purple. Pintown, in my opinion, was an unwinnable disaster, and I found Purple to be very poorly implemented as well, albeit full of fun ideas. The Atomic Heart, I’m sorry to say, is in just about the same shape as Purple.

The game’s concept is that old science fiction nugget about the machines gaining consciousness and battling mankind, but it’s enlivened by the IF presentation, in which the PC is a nanny robot (albeit with a very minor twist, revealed only in game-ending scenes) who bears no hostility towards humanity. “PC as robot” has certainly been done and done much better by other games, as has the twist gimmick, but not in combination (so far as I know) with the “robots vs. humans” plotline, and I liked the idea of playing a robot whose mission is to save humanity from other robots. There were a number of interesting details in the writing, and the structure of the story was good too, with surprising revelations, an exciting climax, and a satisfying ending.

Unhappily, though, the game’s incredibly shoddy implementation demolished any chance I had of enjoying its story. Even more irritating, most of the game’s problems are attributable to nothing but carelessness. For example: synonym problems. There’s a photo that can’t be called a picture. There are “trainers” that can’t be called shoes. There’s this exchange:

>examine bulk integrity module
I only understood you as far as wanting to examine the bulk integrity
module.

“Well,” I had to ask, “isn’t that far enough?”

There’s more, though. Newlines are added or subtracted willy-nilly, giving the game a sloppy appearance. Losing endings are utterly unmarked as such, making them feel more like bugs than dead-ends. There’s a database that supposedly can provide all kinds of information to the robot, but when CONSULTed, even about the topics the game says it knows, all that ever seems to happen is that “You discover nothing of interest in the database.” Then there’s the delightful old “Which do you mean, the eighteenth century bottle or the eighteenth century bottle?” problem. Worse yet, one of the game’s main actions requires incredibly finicky syntax, and that action must be performed again and again in a successful game session.

After about an hour, I was so frustrated with the game’s inability to understand basic things that I started going straight from the walkthrough, and was gobsmacked to discover that even the walkthrough is loaded with commands that don’t work. I can think of no possible excuse for this. Apparently the walkthrough is a command transcript of the game’s author, or someone who has similar inside knowledge, playing the game through to conclusion, but if even the freaking walkthrough document can’t easily figure out how to phrase its commands, isn’t that a really big clue that the game is under-implemented? I think so.

I have to say, I really can’t figure this out. You take good idea, write an interesting story, make up some cool puzzles and such, then you put it together in such a slapdash way that almost nobody could enjoy it. I dunno, maybe there’ll be a big handful of glowing reviews for this game and I’ll discover that I’ve just become uptight and overly picky, but my experience with it was just so aggravating. The game even credits testers, but if the evidence of the walkthrough is any indication, just because a problem is obvious doesn’t mean it will be fixed. But WHY NOT? Why why WHY release something that is so much worse than you know it could be? Take some pride in your work, for heaven’s sake.

I guess this is all getting a little ad hominem, and I don’t mean it to be, but games like this just make me want to pull my hair out. There is just no plausible reason for a game to have problems like this, not with testers and playthroughs that clearly found them. Look, your job as a game author is to make sure your game is the best it can be. Do your job.

Rating: 4.4

Pintown by Stefan Blixt [Comp97]

IFDB page: Pintown
Final placement: 28th place (of 34) in the 1997 Interactive Fiction Competition

Playing Pintown was an extremely frustrating experience. The game was loaded with errors, both in its writing and in its coding. Even the walkthrough had bugs. I went through two hours and two interpreters trying to get the program to respond in such a way as not to crash the game, and after 45 minutes I stuck with the walkthrough. Even with the walkthrough, it took me another hour to finally get the game to stop crashing, and once I had done that a runtime bug derailed three puzzle solutions and magically eliminated the person following me. By the time I figured out that the game is unwinnable (at least according to the walkthrough provided by the author) I really didn’t care anymore. The game’s puzzles were infuriatingly arbitrary, its plot made no sense, and its prose was both very unhelpful and heavily burdened with grammar and spelling errors. I recognize the fact that the author of Pintown is Swedish and therefore might not have the best grasp on English. That’s why beta testing is important — not only does it get rid of those pesky game-killing bugs, but testers who are fluent in English can help correct all those mechanical mistakes.

In Pintown you play a musician (though the game will only respond to “play guitar” in one location at one particular time) who’s just come off a bender where you had a major row with your girlfriend. Now it’s your task to find her and make peace. Of course, the game gives you no hint as to where she might be, and characters who seem to have no programmed responses to any questions or actions aren’t much help either. A vital part of the plot hinges on your getting into a parking garage, but you can only get in there at one crucial point in the game, and your are given no hints as to when and how that point arises. I only managed to get that far with the walkthrough in hand. As to how the game ends, I have no opinion since I never got there — the game’s bugs prevented me.

This game recalled some of the more disappointing moments of last year’s competition, and I found myself once again asking that question: why would anyone who cares about his work at all enter a buggy, error-laden, unwinnable game into the IF competition? There’s clearly a precedent (upheld by many of this year’s entries) of high quality among many of the competition games. And surely authors understand that the idea of the competition is to encourage the writing not just of IF, but of good IF. So why would people humiliate themselves by entering poorly written, untested games? It’s a puzzle too tough for this adventurer to solve.

Prose: The prose in Pintown was weak. On the one hand, it’s one of those situations where there are so many technical errors that the overall prose quality is dramatically impacted for the worse. Then again, even without all those errors, I still think the prose would be bad. There’s a character who responds to every subject with “I don’t know.” Room descriptions leave out crucial objects. Sentences often make little sense. Clearly, there was a lack of effort here.

Plot: It’s hard to use the word “plot” in connection with a work like Pintown. The game’s basic goal, according to the designer, is “simply to get along with your girlfriend.” Of course, this is hard to do when she’s nowhere to be found until the endgame. Actually, the main goal of the game is to be in the right place at the right time, then to steal a car, set some random events in motion with an arbitrary action, then to clean up your apartment in exactly the right way, and finally to rescue a cat and “discover” your girlfriend’s hiding place. These plot points are relatively unconnected by any sort of logic.

Puzzles: The puzzles are very much of the “read the game designer’s mind” variety. First, you wake up and the first thing you need to do is go back to sleep. Then, you need to follow someone without really knowing why; if you don’t do this, you’ll never finish the game. Then, you steal a set of keys and a van to drive to your apartment. There is no alternate solution to the apartment puzzle. You can’t call a cab, or walk there, or ask anyone where it is. You’ve got to be there at the right time to get those keys and steal that van. Well, you get the idea.

Technical (writing): There were many grammar and spelling errors in the game. They include words missing letters, misspelled words, and made-up words (“trafficated?”)

Technical (coding): But if the writing was bad, the coding was downright awful. Playing the game under WinFrotz led to many game-killing bugs of the “FATAL ERROR: Illegal Object Number” or “Illegal Attribute Number” variety. Under JZIP, the game would just stop responding at random points. In addition, the game state was unstable enough to eliminate a follower and return two solved puzzles to an unsolved state, all at once. Unsurprisingly, beneath these major bugs was a panoply of minor bugs, including a dearth of synonyms, important missing verbs (like “pet cat”), and readable materials that only respond to “examine”, not “read.”

OVERALL: A 1.5