First: the blogger appears to have come to old school play via 5e, in contrast to my background in OD&D and 1e. Background influences your models when making new material. Consider the way we each used “spell slots”, for example.
- For 5e players, spell slots are a resource to be used. So, The Kernel in Yellow’s approach is to require one spell slot per emergency casting and allow additional slots to be “spent” to improve the spell.
- For me, “spell slot” is an artifact of the way we’re talking about the spell prep process and cannot be “spent” for any reason. So, I treat casting from a book as spell prep and ask "Should there be a difference if the caster were already ‘full’ of prepped spells?’
Then we get to casting time. Do you start with regular spell casting as your model and balance emergency casting by making it slightly longer, but still viable as something you could cast in combat? Or do you start with spell prep times, normally undefined, and try to define them?
The tables we each use diverge a lot more. I don’t have a “spell does nothing” effect on my table, and the range of effects aren’t that extreme. 97% of the time, spell casters are going to get more or less the effect they want unless they rush things. Backfires are pretty difficult to get without divine intervention, as is the double-strength maximum result of Fantastic. Kernel in Yellow’s table is aimed more at penalizing emergency casting at least part of the time, to prevent it from being the go-to choice for spell casting. Plus, that Fireworks result is pretty extreme, causing the caster to lose 1 to 4 additional spells.
A side note: since I allow learning new spells via Read Magic, some “emergency” spell casting is really going to be experimenting with a spell the caster has never cast before. Sometimes, the spell will be unique and of little use anywhere but in the dungeon. This is one reason I didn’t want to lean too hard on penalizing spell casters who cast from spell books.
A lot of the differences can be summarized as “My model is spell prep, Kernel in Yellow’s model is spell casting.” With the second model, your design question is going to be “Why don’t casters always cast spells from spell books?” Your answer will probably be “because the results are usually bad,” and you’ll work to balance it against ordinary, “safe” casting. With the first model, my design question was “What happens when you try to prep a spell in a dungeon, instead of taking your time at your home base?” My answer wound up as “It will probably be a little unpredictable, but not much. Mostly, it’s just annoyingly long, so people rarely do that.”
You’ve got to decide which route you want to take: giving players a boost and then balancing that, or giving them something that adds to the rich background details.
This work is licensed under a Creative Commons
(CC BY-NC-SA 4.0) license.