Ugh... what the fuck has happened to the Brilliant Gameologists?
Re: 19 Int. It doesn't show any kind of contrast at all because it has no association with reality. What's important in this argument isn't the biggest boost from the baseline, it's what you can do to make a better Wizard.
Re: 30 Int. Loses 7 spells relative to 36 Int, not 8. 30 Int gives another bonus 5th-level spell.
Re: Nerveskitter. +5 init for one encounter, vs. +2 init for all encounters. Further, you can cast Nerveskitter on someone else, which means that getting a free +2 init yourself means you can buff the party Cleric with +5 initiative and still have the same chance of going first.
Re: Mirror Image. Mirror Image is a shit spell at 17th level. Any schmuck and throw out a quickened Manyjaws to instagib all your images and hit you for damage, aside. Greater Mirror Image is useful against precisely one full attack as a panic button.
Re: Quickened True Strike. Just one? Really? You have 78 spells per day, and only one of them is going to require an attack roll from you,
ever?
Re: Vampiric Touch. Sure, it drains 35 HP on average. Also, it's now a ranged touch attack, which means it can benefit from the +2 to hit that improved Dex gives you. That said, it's temp HP, which means it interferes with spells like False Life or feats like Minor Shapeshift.
Re: Greater Resistance. It's a Resistance bonus, which happens to be the most common type of bonus to saves. If your Cleric isn't persisting Shield of Law/Holy Aura on the party, you're casting Superior Resistance yourself. In either case, this doesn't work.
Re: Persistent Heroism. LOLno. Heroism is a touch spell.
Re: Money. I brought up the Pearl of Power IX because you were whining about having fewer spells. It's actually one of the most overpriced, useless, pieces of trash in the game. Better items are the Normal Quicken rods, which turn 3 of your 6th-level spells into effective 10th-level spells. Cheaper are Greater Extend or Greater Sculpt rods, which turn 3 of your 9th-level spells into effective 10th-level spells. There's also the Reach rod, which takes those up to 11. Tell me, how much Int does it take for a 17th level Wizard to cast 11th level spells?
At this point, I don't care what your opinion is. You decided before even reading the OP that it was wrong, and made absurd, closed-minded responses to justify your position.
Ah the usual. I never troll/flame like what's in here. Heck I still use my original account at GitP. But apparently I'm o-so-controversial. So here goes:
This applies to mostly core, mostly unoptimized, and mostly "normalish" D&D campaigns: usual human wizards who are waiting to go archmage and actually play with a party, not faerie initiate necropolitan incantrixes who use clones 1-ups. Those guys max int because its all that helps them anymore. Everyone else has to split precious resources for unpleasant situations.
I think this describes it pretty well. If you have a 28 PB, then the net benefit of increasing other stats before putting Intelligence to 18 are really high. On the other hand, if you have a 36 PB, you'll run out of places to put those points, and still have enough left over to get 18 Int anyway. With gold it's the same way, if you have all the gold you could ever want, then you're not making a choice between Int and Con/Dex/Wis, you just max it all.
That said, I don't think core or optimization level matters nearly as much as the available resources.
EDIT: I think this thread has run it's course. I was hoping for a higher level of discussion than this, like maybe some discussion about where the cost/benefit breakpoints are, or examples of what a high Dex/Con Wizard can do better than a pure Int Wizard.