Go to the game's main page
Review
It has been a long time since I've played a game that just didn't work for me... not even a little bit. The dominating design ethic of Frobozz Magic Support seems to be "player vs. author grudge match" at its most refined, something much more common in 1997 (when this game was released) than it is today but already facing heavy criticism at the time.
I played hundreds of moves of this game and made zero progress without spoilers. The game seems to purposely withhold necessary information as much as possible, making everything that you encounter seem like a red herring. This is apparently by design. Significant changes in plain view of the PC go unremarked, requiring the player to repeatedly inspect the environment in detail to even have a chance to notice them. The main sidekick, which would typically be used to provide orientation, almost exclusively alternates between a small library of useless stock responses and inane non-specific commentary. The included hints weren't even particularly hintful.
It may be the case that someone who has played Spellbreaker will be able to use knowledge from that game productively and get off to a much better start; since I haven't played that game, I couldn't tell you. After looking over the first several hundred moves of the ClubFloyd transcript -- and noting that the only real guidance about how to win appears to be an "encrypted" message (using a "code" that is clever but deeply unfair), my confidence in this game as entertainment was reduced to zero. The key to the code is (Spoiler - click to show)vaguely hinted at by the "mythical" spell noted on the syllabus, or you can just find the decoded text in the transcript.
On the plus side, it is very competently implemented as a program, with no observed bugs and a minimum of typos in the text. The most notable technical flaws are an indistinguishable noun issue involving granola, a picky distinction between >READ and >EXAMINE affecting the same puzzle, and an object with a blank name that can be a disambiguation candidate for the word "black."
Recommended only for would-be authors, with the advice to study the game's puzzle design and try to think up ways to make it more fun and/or more fair by modern standards -- as is, it's pretty much an encyclopedia of examples of what not to do. I see that the SPAG review cites a contemporary interview from the year of the game's release in which author Nate Cull reportedly claims to dislike the types of puzzles of which this game is made. If that's true, then the implication is that this was intentionally designed to be a poor game.
If approaching the game as a player, be advised that the parts of this game depend on the whole, so you should explore every available scenario to see the universe of items before attempting to solve any one scenario. Even that often won't be enough, because some key interactions are utterly opaque, so if all else fails it is probably safe to assume that mind-reading is required and that hints or the walkthrough should be consulted. Also, here's the orientation that the game should give you: (Spoiler - click to show)Touching the sphere lets you cycle through pending support calls. When a support scenario is displayed, the compass rose will indicate a direction that it is possible to go. And to avoid end-of-game aggravation: (Spoiler - click to show)Don't forget to log every task as completed as you go.
