So let's see. We have a wizard at initiative count 15. He casts a full-round action casting time spell. At initiative count 14 on the same round, he's done casting.
Now compare. We have a wizard at initiativec count 15. He casts a 1-round casting time spell. At initiative count 14 of the same round, he's still casting. At initiative count 15 on the next round, he finishes casting.
1st, initiative count is an abstract, artificial construct. It is also
the glaring shortcoming of the turn-based action-resolution system. The thing is, the only "fix" requires that you over-complicate game play -- that's the trade-off. It is what it is.
Crack open your DMG, and turn to page 24:
[spoiler]
Simultaneous Activity
When you play out a combat scene or some other activity for
which time is measured in rounds, it can be important to remember
that all the PCs’ and NPCs’ actions are occurring simultaneously.
For instance, in one 6-second round, Mialee might be trying
to cast a spell at the same time that Lidda is moving in to make a
sneak attack.
However, when everyone at the table plays out a combat round,
each individual acts in turn according to the initiative count for
his character. Obviously, this is necessary, because if every individual
took his turn at the same time, mass confusion would
result. However, this sequential order of play can occasionally lead
to situations when something significant happens to a character at
the end of his turn but before other characters have acted in the
same round.
For instance, suppose Tordek hustles 15 feet ahead of his
friends down a corridor, turns a corner, and hustles another 10
feet down a branching corridor, only to trigger a trap at the end of
his turn. In order to maintain the appearance of simultaneous
activity, you’re within your rights to rule that Tordek doesn’t trigger
the trap until the end of the round. After all, it takes him some
time to get down the corridor, and in an actual real-time situation
the other characters who have yet to act in the round would be
taking their actions during this same time.
[/spoiler]
(side note: this is the kind of stuff you miss if you rely on the SRD)Now juxtapose that with this, which I present once again (and now we've
both posted it):
a round [...] means a span of time from one round to the same initiative count in the next round.
In light of the previously-discussed
actual given definitions, I think the example given in the 3rd paragraph is quite analogous to this one (insofar as adjudicating effects and activity).
When you begin a spell that takes 1 round or longer to cast, you must continue the invocations, gestures, and concentration from one round to just before your turn in the next round (at least).
IE: There is a period of time where you're doing " the invocations, gestures, and concentration " in a 1 round casting time spell, where you are NOT required to in a full-round action casting time spell
(this is covered in the DMG quote above; but ...) the only spells that have a
"full-round action
" casting time are ones that are spontaneously metamagiced. Spontaneously applying a metamagic feat to a spell is actually an exception to the "casting time for spells" rule -- since it is itself a specific and isolated exception, it cannot be used as the standard by which to judge other elements of spellcasting.
If a spell’s normal casting time is 1 standard action, casting a metamagic version of the spell is a full-round action for a sorcerer or bard. Note that this isn’t the same as a spell with a 1-round casting time—the spell takes effect in the same round that you begin casting, and you aren’t required to continue the invocations, gestures, and concentration until your next turn. For spells with a longer casting time, it takes an extra full-round action to cast the metamagic spell.
Seriously, the rules even explicitly say that a 1-round casting time and full-round action are different.
No, no they don't. The rules say that
casting a spontaneously metamagiced spell and
casting a "1 round casting time" spell are different. That is a
very different thing than saying that a 1-round casting time and full-round action are different. To the contrary, they rules actually
do explicitly define a 1-round casting time spell as being a full-round action (albeit with an exception).
In both cases, they are each an exception to a "normal" rule; but they are each exceptions to
different rules. Since the rules for which each are exceptions are completely different, the comparison is invalid.
Speaking of exceptions -- (given the definition) the 1-round casting time is simply an exception to the "full-round actions are resolved on the character's turn" rule.
Which has been the biggest part of my point all along.