*double-checks* Whoops. My bad. I must've divided by an extra sixty. Yeah, that's more reasonable. Still, my point about taking ten to solve an obscure code still stands. Instead of raising the time by four minutes and taking twenty (ie spending eighty minutes, though it should be noted you still have a five percent chance of solving it in four), you can reduce the DC by twenty by spending 2^4 = 16 minutes. That lets you automatically figure out any message that uses obscure code words (eg my Mister Abernathy letter), complex cyphers/symbols (transliterated to Mandarin Chinese), or is written in a distantly-related language (such as Latin). That's for anyone of average intelligence with no particular aptitude for reading. Someone of above-average intelligence who enjoys reading (let's assume a first-level, 12-INT guy with maxxed out Read) can do it in eight minutes. We're talking about a guy who can probably write a decent paper on a novel or short story, assuming he takes good notes (or is familiar with that author). Speaking of good notes, notes actually do you no good. You can read at half speed while taking notes for a -5 to the DC. You can read while comparing with a secondary source for -5 to the DC. You can just read at half speed for a -5 to the DC. A few suggestions:
- The DC reduction for slow reading needs to be lowered. I'd say -2 would work better. As it is right now, sixteen minutes of reading takes a whopping 20 off of the DC, which can completely negate high-level encryption. Leave the modifications for secondary sources and note-taking as they are.
- There should be some sort of cap on how long you can spend studying a page before you figure it out.
- The DC for certain forms of coding should be increased.
- Some sort of limits should be imposed on what you glean from a text, depending on the DC or the DC modifiers. As it stands, you either know it perfectly or can't make heads or tails of it.