Anubis said:
For saving throws and attacks, the text specifically states the automatic successes/failures rule. I would say that since nothing ever says a 1 is an automatic failure with spell penetration, then it doesn't apply. Besides, if it did, a 20 would be an automatic success, and that kills the logic.
How so? I see the possibility of an automatic success on a SR check as following the spirit of the game: a world in which anything can happen.
Think of it this way, perhaps:
Can you honestly see a level 1 wizard being able to hit a great wyrm prismatic dragon with magic missile successfully? No! The SR is simply too high, meaning the wizard doesn't have enough "oomph". Basically, with an attack, you can always get lucky and land a good hit regardless of defenses. Magic works differently, as does SR.
What you see as "oomph," I see as "magical aptitude." Most people don't have any noteworthy defense against magical attacks (i.e. SR); they're completely open, so spellcasters don't even have to roll, normally. When someone DOES have such protection, though, the caster has to see if he can break through their magical defenses, just like any fighter would have to do when trying to hit a certain AC. Since it's no longer a sure thing, it has to be rolled, and since it's treated like AC, SR checks fail on a 1 and succeed on a 20. Thus, I believe that luck IS involved, and one can indeed get lucky, or UNlucky.
DnD is a game of the suspension of disbelief: dragons (with wings that, in the real world, wouldn't be able to support them) can fly and breathe fire (and other nasty substances); an animated table can attack the people sitting down to enjoy a meal conjured from thin air; a holy man can obliterate waves and waves undead soldiers sent by his unholy rival. I have no problem with a level 1 wizard getting a
magic missile through on ANY creature not IMMUNE to magic.
On the same token, I think it'd be extremely silly for a level 1 character with SR 1 to be able to somehow miraculously block a meteor swarm cast by a level 1000 wizard.
As silly as, say, a level 1 goblin warrior being able to somehow miraculously block a +2000 Greatsword (+4000 versus Goblinoids) wielded by a level 10000 ranger (primary favored enemy: goblinoids)? Where does the suspension of disbelief end?
I hope this clears it up. That's not even a good house rule. Auto stuff should be left to attacks and saves only. I would compare spell penetration more to a skill check than an attack roll.
And I would compare spell penetration more to an attack roll to a skill check. And, with some more food for thought, another section from the SRD, this time from the section on scroll activation -- another type of caster level check which, I do believe, works very similar to that required by SR:
the 3.5 SRD said:
If the user meets all the requirements noted above, and her caster level is at least equal to the spell’s caster level, she can automatically activate the spell without a check. If she meets all three requirements but her own caster level is lower than the scroll spell’s caster level, then she has to make a caster level check (DC = scroll’s caster level + 1) to cast the spell successfully. If she fails, she must make a DC 5 Wisdom check to avoid a mishap (see Scroll Mishaps, below). A natural roll of 1 always fails, whatever the modifiers.
Bold, again, is emphasis mine. Now, as to whether the bolded section is referring to the caster level check or the Wisdom check remains to be seen, but due to the passage declaring Ability checks as untrained skill checks,
the 3.5 SRD said:
An ability check is a roll of 1d20 plus the appropriate ability modifier. Essentially, you’re making an untrained skill check.
how could THAT possibly be, given that skill checks never automatically fail on a roll of 1?
I thank you for your profound opinions, Anubis. As to the validity or viability of house rules, be them my own or someone else's, I again thank you for your insights. Truly, you have opened my eyes. I never realized the wide gulf between what is believable, and what is inconceivable.
--B--