For some time I’ve seen comments that the separation between good saving throws and poor saving throws is too big, and that it is difficult to find challenges appropriate to higher-level characters with different saving throw progressions.
Mathematically, I can see the argument for this. A high-level spell from a moderately optimized spellcaster will likely have a save DC in the high 20s. A Wiz17 with Int 28 (18 base, +4 from level bumps, +6 from an item; I have found in the last few years that this should be expected) casting a ninth-level spell will have a save DC of 28.
A good base save progression would give +10 at this point (17/2 = 8, +2); if the target has +9 for related ability score (cleric with Wis 28, rogue with Dex 28), +5 more for a cloak of resistance +5 — a total of +22 — he’ll need to roll 6 or better on d20 to make the save. +2 more for the appropriate feat (Great Fortitude, Lightning Reflexes, Iron Will) and he’s down to 4+ to make the save. Pretty good chances, but he’s pretty well tuned to make this save (good base save, topped out ability score, good saving throw item, and a precious feat). The guy who doesn’t tune to this extent (say, a bog-standard Ftr17 with Wis 12) has a total of +6 to his Will save. He’ll save on a… 22+, on d20 (ignoring the auto-20 rule). This drops to 17+ on d20 if he too has a cloak of resistance +5 — but shouldn’t these be uncommon, rather than standard equipment?
The spellcasters are a specific example, and one that I’ve seen as a problem. However, monster natural abilities usually have their save DCs based on ‘HD/2′ as well, and monster HD often is markedly higher than their CR (a CR17 creature can have 30 HD, for example, though those ones tend to be ‘mostly meat’ with few abilities that require saves to be made). Again, the character made to resist them will have a decent to pretty good chance, but everyone else is likely to suffer.
Considering how many of these abilities and attacks can be ‘save or die’, things can suck for the low-save guy.