Premise: I know that houseruling retroactive skill points would prevent this from being a problem.
I'm aware of the possibility. It just isn't what is true for most characters I make.
This question stems from my optimization obsession. Whenever I build my characters I always try to get the best I can from the build I chose. A suboptimal barbarian/fighter half-elf, for instance, would always start with the barbarian level for the increased HP and skill points.
This question is about one of those times where the order at which you gain things changes the final result: the non retroactiveness of skill point gains and the interactions between money gain and buying magical tomes.
Tomes of Clear Thought (or wish spells) are a strange beast. They're one of the few ways to increase your ability point gain, but every time you spend money on them you start from scratch - meaning your actual inherent bonus doesn't matter, if you want a +5 you need to pay the same sum no matter if you started from a +0 or a +4.
Moneywise, it's more efficient to wait until one can buy a +5 book (or a +4, if a +5 would get an odd Int score) and do all the upgrading at once.
Skill-points-wise, I'm pretty sure that's not the case.
Let's say you can buy a +5 book at level 13, and you started from an odd score. When you get to level 14, you gain 3 more SP. Now, what if your wealth allowed you to buy a +3 book at level 12, meaning you would have less money at level 13 and a +5 could only be bought at level 14? You would have gotten 2 extra AP at level 13, 2 more at level 14, totalling 4 more by the same level.
I'm looking for an optimum-finding procedure (which I'm going to transform into a fillable spreadsheet) that considers those inputs:
- the money the character gets at every level - according to the WBL table or to the next row of it (since you can be 1XP removed from the next level and with all the money already, buy the tome and wait a week before leveling up)
- whether your DM tells you you can spend all your WBL on a single object or not (this is usually done at character creation only, but I'll probably need it to be true for every level up to the one I'm building my character at), including the allowed percentage and the level at which the character is built.
- whether the starting Int value is even or odd (it will usually be odd, to actually benefit from a +5 tome. Sometimes a high roll and a desire not to lose physical stats due to age category shifting or starting with an old character will make it even regardless, and also see below). Notice how this actually splits the problem in two, one with odd books and one with even ones.
It's already been determined that every age increase that's deemed to be desirable on the build should be done as soon as possible (level-wise. I don't expect people to age before time) since it has no cost and it can only improve the gains. If the starting Int with the age bonus included turns out to be odd, the aging can then be delayed until the first tome is bought but not later (otherwise one would need to buy a larger tome, possibly shifting one of the next purchases up a level), but the math has to be done considering the bonus already there.
The optimization should also take the following into account:
- sometimes, the resources your build can put into a starting Intelligence score are poor and reaching a odd score (e.g. 15) is not possible. Is the extra early skill points gained by starting from an even score (14) a net gain compared to starting from a lower odd score (13)? An odd score means being able to purchase your first books earlier, after all.
I'm not interested in Int gains from leveling up, because this is for characters that do not otherwise need that much Int, which means those every-fourth-level resources will be spent elsewhere, but I'm probably going to ask a second question itroducing the extra complexity once I have a valid answer to this one.