Post/Author/DateTime	Post
Ulden Throatbane 04-21-04, 08:50 PM	This post is inspired by and dedicated to the incredibly huge number of sorcerer threads on this boards. I will not go into the points for or against lifting the increased casting time on spontaneous metamagic. Rather, I am proposing a new way to think about metamagic feats, that may help in finding an appropriate fix (while I personally don't like the increased casting time, I am not above compromise on this issue). Whether a spellcaster prepares his spells or cast them spontaneously, metamagic feats effectively increase the number of spells he or she knows by an amount proportionate to the spells that he or she knows already that the feat could modify. For example, let us take a 6th level wizard. Assuming that no additional spells have been acquired in her adventures, and that she is choosing her spells gained from advancing in levels from the highest level possible. Assume that the wizard is using the elite template (int 15 at 1st level) The number of spells she should know, then, would be as follows: 0 1 2 3 19 7 4 4 Now let's give our hypothetical wizard the still spell feat. This effectively increases the number of spells she now knows: 0 1 2 3 19 7+19 4+7 4+4 Her total number of known spells is now as follows: 0 1 2 3 19 26 11 8 Similar numbers can be generated for a sorcerer: Spells known before still spell 0 1 2 3 7 4 2 1 Spells known after still spell 0 1 2 3 7 11 6 3 Of course, these new spells have a worse save DC than normal spells cast at the same level, something we shouldn't forget. Now lets review the trade-off between prepatory and spontaneous casting: -Spontaneous: Know fewer spells, and very limited in how more spells can be acquired but can cast them as needed. -Prepatory: Know more spells, and can acquire new ones with relative ease, but must have the spell prepared. Given this trade between the two casting mechanics, the balance between Spontaneous and prepatory casting is in the ratio of spells known. However, since wizards can spend money to add spells to their spellbook, it is impossible to tell what the actual ratio would be without setting a percentage of the treasure he should have at his level that was devoted to adding spells. I will get to this later. Getting back to the question of metamagic, if we treat the feats as simply increasing the number of spells known, then the most important question is whether or not the ratio of spells known remains the same as it did before the feat was taken. A significant alteration towards one side or another would mean that a restriction on metamagic use needs to be put in place. To show what I mean by this, let us pretend for a moment that metamagic feats did not exist, and instead metamagic spells had to be learned like any other spell. In our hypothetical scenario, we have a 10th level sorcerer and a 10th level wizard (both whom recently leveled) debating for themselves whether or not they should learn the 5th level spell quickened true strike. Both the wizard and the sorcerer face an opportunity cost in taking this spell. For the wizard, this opportunity cost is expressed in the gold cost of adding a 5th level spell to his repetoire. Between 9th and 10th level, his wealth should have increased by 13,000 gp using table 5-1 in the DMG (this amount includes magic items and the like). Assuming he doesn't have Boccob's Blessed Book, quickened true strike will cost him 750 gp (the cost of scribing the spell plus the price of scribing a spell from a spellbook). That's roughly 1/20th of the wealth he gained during 9th level (he also has the option of using two of the spells he automatically gains from leveling up). The sorcerer faces a more difficult decision. The cost is one of his 4 known spells at 5th level. The opportunity cost, then, is every 5th level spell he passes up on to learn quicken true strike instead. Of course, if he does choose it, he gets the same advantage of spontaneous casting as he does for the rest of his spells. Back to the question at hand: would metamagic feats alter the ratio of spells known between a sorcerer and a wizard. To answer, we must first establish what the ratio should normally be. To make things fair, we will rate the ratio of each spell level independently. We will also measure the spells known at level twenty, since any power difference will be most noticable at the highest level. A level twenty sorcerer has the following known spells: 0 1 2 3 4 5 6 7 8 9 9 5 5 4 4 4 3 3 3 3 For the wizard, however, its much trickier to determine how many spells a wizard should have. Not only must we decide how much treasure he has devoted to learning new spells, but we have to decide how he has divided his cash between each spell level. Let us begin with how much total treasure he should have at 20th level. The table on 5-1 gives us 760,000gp. Let us also assume that he crafted himself a Boccob's Blessed Book (6,250gp, which will be deduced from the total spent on spells). For this experiment, our wizard will use 5% of his total wealth for acquiring new spells by copying other wizard's spellbooks. The amount he is spending on copying spells is 38,000gp - 6,250gp (B.B.B.) =31,750gp, or 635 spell levels copied. (for comparison, a headband of intellect +6 costs 36,000gp) If I wanted to fill B.B.B., I'd need to spend 56,250gp on spells. (note,: without Boccob's Blessed Book, I would have had to pay an additional 63,500gp on the spells I'm adding. This makes it a 10-fold net savings.) As he already has all the cantrips from the PHB, we'll assume that he distributes his new spell levels evenly. That is, for each spell level, 70 pages are alloted to adding spells (with 5 left over). If there are more pages than needed to fit every PHB wizard spell from that level, or if there are left-overs from rounding, they will be taken to the next spell level. Also, assumed that he has the following selection of spells from the ones he chose for free: 0 1 2 3 4 5 6 7 8 9 19 7 4 4 4 4 4 4 4 8 Here is the table for what spells he should gain in addition to his freebies, after rounding and re-allocating left-overs: 1 2 3 4 5 6 7 8 9 35 46 23 19 15 11 10 9 8 Of course, this example assumes that sufficient time is given to learn these spells. A wizard needs to take 2/3rds of a year (spread out over the campaign, obviously) to learn all these spells. The amount of down-time available of course varies from campaign to campaign. Unfortunately, the DMG does not provide a standard ratio of down-time to gp. We can now give the spells known for a level 20 wizard that devotes 5% of his treasure to additional spells: 0 1 2 3 4 5 6 7 8 9 19 42 50 27 23 19 15 14 13 16 Note that the wizard in this example doesn't learn every spell except at 1st and 2nd level. At last, we can establish the ratio of spells known for a 20th level wizard and a 20th level sorcerer, assuming a 5% investment in new spells and even distribution of pages. Ratios are rounded to nearest rational number * 0 1 2 3 4 5 6 7 8 9 sor 9 5 5 4 4 4 3 3 3 3 wiz 19 42 50 27 23 19 15 14 13 16 rat 1/2 1/8 1/10 1/9 1/8 1/5 1/5 1/5 1/4 1/5 1st through 4th level spells have an average ratio of 1/8.5, while 5th through 9th level spells have an average ration of 1/5. Now we check the effects of metamagic on the number of spells "known". For simplicity, we will categorize each metamagic feat by the effective spell-level adjustment it imposes on each spell. We will also not consider levels that cannot carry a metamagic feat of that level. So when the quicken spell feat, we will ignore changes in the ratio of 0 to 3rd level spells, because there aren't any. 1-level adjustment (enlarge, extend, silent, still) * 1 2 3 4 5 6 7 8 9 sor 14 10 9 8 8 7 6 6 6 wiz 61 92 77 50 42 34 29 27 29 rat 1/4 1/9 1/9 1/6 1/5 1/5 1/5 1/5 1/5 2-level adjustment (empower) * 2 3 4 5 6 7 8 9 sor 14 9 9 8 7 6 6 6 wiz 69 71 73 46 38 33 28 30 rat 1/5 1/8 1/8 1/6 1/5 1/6 1/5 1/5 3-level adjustment (maximize, widen) * 3 4 5 6 7 8 9 sor 13 9 9 7 7 7 7 wiz 46 65 69 42 37 32 31 rat 1/4 1/7 1/8 1/6 1/5 1/5 1/4 4-level adjustment (quicken) * 4 5 6 7 8 9 sor 13 9 8 7 7 7 wiz 42 51 65 41 36 35 rat 1/3 1/6 1/8 1/6 1/5 1/5 We compile the ratio adjustment into a single table to make reference easy: * 1 2 3 4 5 6 7 8 9 0 1/8 1/10 1/9 1/8 1/5 1/5 1/5 1/4 1/5 +1 1/4 1/9 1/9 1/6 1/5 1/5 1/5 1/5 1/5 +2 *** 1/5 1/8 1/8 1/6 1/5 1/6 1/5 1/5 +3 *** *** 1/4 1/7 1/8 1/6 1/5 1/5 1/4 +4 *** *** *** 1/3 1/6 1/8 1/6 1/5 1/5 Observations: -The inclusion of 0-level spells in the study gives the appearance of skewing the low-level results in the study to favor the sorcerer when in fact the worth of casting these using metamagic feats with these spells is somewhat questionable. I felt it was necessary to include them for the sake of completeness. However, we will discount the impact of these spells in future observations. -When the level adjustment is low, the ratio change is more likely to work in the sorcerer's favor, at least at low levels. After 6th level, the ratio is unchanged. -At higher level adjustments, the ratio is more likely to swing in the wizard's favor. The biggest shift is in the 6th level quickened ratio (1/5-1/8) Important notes: this test not account for the applicability of metamagic feats such as empower, widen, maximize, etc. This was done for the sake of simplicity. Conclusions: As much as I would like to claim that this example shows that the nerf on spontaneous metamagic is unessecary, the test case is still highly limited because it only compares the classes at one level and makes assumptions about a wizard's spending decisions that cannot be universalized or even standardized. My only other conclusion is that I can't believe I spent 3 hours making this post :D Comments on this 'study' are always welcome.
Tony Vargas 04-21-04, 11:01 PM	I think you're missing one very important point. You're looking at spells known. You also need to look at the range of spells available to be cast. That is, what you could really be comparing, is the sorcerers known spells (all of which are available to be cast over the course of the day), with the Wizard's daily alotment of prepared spells (because only those spell prepared are available). I think if you run the baseline (before metamagic) ratio between those two, you'll find that they're pretty close. One reason that wizards and sorcerers are actually pretty balanced, in practice.
Ulden Throatbane 04-22-04, 01:02 AM	Originally posted by Tony Vargas I think you're missing one very important point. You're looking at spells known. You also need to look at the range of spells available to be cast. That is, what you could really be comparing, is the sorcerers known spells (all of which are available to be cast over the course of the day), with the Wizard's daily alotment of prepared spells (because only those spell prepared are available). I think if you run the baseline (before metamagic) ratio between those two, you'll find that they're pretty close. One reason that wizards and sorcerers are actually pretty balanced, in practice. You would indeed get different results, but the focus was put on spells known because it reflects the trade-off between spontaneous and prepared magic. The ratio of spells known should be in the wizards favor (even as much as 10 to 1)-that's the strength of prepared magic. You have a bucket-load of spells, but you need to have the right one ready. The point of this thread was to see if metamagic altered this baseline significantly, but in order to do this it was important to establish what the ratio normally is. In fact, daily allotment would make spontaneous magic seem way too strong because a sorcerer can cast his known spells whenever he wants, while the wizard is stuck with a set list of spells. By measuring spells known on both sides, however, we can establish a 'cost' in terms of spells known for being able to cast spontaneously. In essence, whenever a wizard takes a metamagic feat, he gets a swathload of new spells without having to spend any time or money. Whenever a sorcerer takes a metamagic feat, he gets a bunch of new spells without expending spells known slots. On a per-spell basis, the savings is greater for the sorcerer. But because the number of 'new spells' is based on the number of spells you already have, the total value should come out the same. Basically, what I posted tries to look at the whole metamagic question from a different angle-one that can at least be quantified.
Tony Vargas 04-22-04, 03:53 AM	Originally posted by Ulden Throatbane By measuring spells known on both sides, however, we can establish a 'cost' in terms of spells known for being able to cast spontaneously. Hmm... I don't think it works that way... that is, I don't think there's a clear ration of sorcerer spells known to wiz spells known that 'pays for' the sorcerer's spontaneous casting. Rather, sorcerers are limitted by thier spells known, while wizards are limitted by the number of spells they can prepare. If you compare the number of spells a sorcerer knows the the number a wizard can prep, they're actually pretty close - and that makes sense, since each represents what the character can actually /do/ with thier spells, in a given instance. When a sorcerer acquires metamagic, the variety of spells he can cast is increased. When a wiz gets a metamagic feat, the number of spells he can prepare is unchanged, so the variety of spells that he can have ready to go at any given time isn't improved. Basically, what I posted tries to look at the whole metamagic question from a different angle-one that can at least be quantified. I think you're onto something with that, actually. I hadn't thought about it that way before. The problem is that there's no telling how many (or few) spells a Wiz might know. Some will invest more gold in thier books, some will have fewer opportunities to acquire spells. OTOH, the spells known for a sorcerer are pretty nearly fixed - as are the spells prepped by a wizard. So, they can be compared more easily, and, they're more relevant in a practical, 'what can you actually cast' way.
Ulden Throatbane 04-22-04, 09:31 AM	If a wizard leaves a spell slot open, though, he only needs to take 15 minutes to prepare a spell for it. Also, spells known reflect what items he can make, especially with regards to scrolls and wands. Here, the wizard has a clear advantage in having more spells known. The 'no clear ratio' of spells known is a problem, but comparing the spells a wizard has prepped vs. the spells a sorcerer knows would be problematic because they are two different things. The spells the wizard knows only reflects what he's prepared for that day, while the sorcerer is stuck with what he's selected for the next two levels. Spells known reflects the tradeoff in overall versatility over the course of the entire campaign. That is why it is more appropriate to compare the two. Comparing spells known for a sorcerer and spells prepared for a wizard produces a skewed picture of the trade-off in the casting systems that is only exacerbated when metamagic is taken into account. In fact, if one is comparing the ratio of spells known to spells prepared, the inclusion of metamagic into the system would so skew the ratio in the sorcerers favor that it would suggest that metamagic for spontaneous casters should not be allowed. Ever. But such a conclusion is faulty because it only reflects "what can be cast" for that day only. OTOH, the spells known for a sorcerer are pretty nearly fixed - as are the spells prepped by a wizard. So, they can be compared more easily, and, they're more relevant in a practical, 'what can you actually cast' way. Spells known fixed for a sorcerer: yes Spells prepared fixed for a wizard: most certainly not. The wizard can easily change what spells he has available. AND unlike the sorcerer's spells known, a wizard's spell slots go up as his intelligence increases.
Tony Vargas 04-22-04, 01:45 PM	BTW, thanks for starting this discussion - it's amazing that, even though this sorcerer stuff has been bandied about for years, we can still find new ways of looking at it. Originally posted by Ulden Throatbane Spells known fixed for a sorcerer: yes Spells prepared fixed for a wizard: most certainly not. The wizard can easily change what spells he has available. Compared to spells known for a wiz, the /number/ of spells prepped is pretty stable: 4, 5, o 6 depending on INT & specialization. I don't think we want to compare specialists, and, if you assume a really high INT, 5 would be a pretty good estimate. The point is that, at a given moment in the course of an adventure, the spells the sorcerer has available are the ones he knows, whereas the spells the wiz has available are the ones he has prepped. That's something you can actually compare, and, it illustrates the impact of metamagic: metamagic increases the number of spells a sorcerer has available at any given moment - it doesn't do that for wizards. Now, you could say that that amounts to comparing the strength of the sorcerer class to the weakness of the wizard - but it's not, it's comparing thier weaknesses: the sorcerer's few spells known to the wizards limitted set of prepped spells. You could try to compare thier strengths: the sorcerers ability to cast the same spell many times vs the wizard's vast potential pool of known spells - but, that's much harder. It's hard to say when being able to cast Melf's Acid Arrow 7 times in a row will be really useful - it's situational. It's also very hard to quantify the advantage of being able to prep just the right spell for the occassion - because it's hard to judge how dependably a player will /know/ what that right spell really is. And, really, it's the weaknesses of thier spellcasting abilities that define and balance the two classes. ... So, you can see why metamagic has been limitted for sorcerers - it's because it directly addresses thier major weakness. The extra time at least slightly limits the availability of metamagicked spells to the sorcerer, limiting somewhat the increased number of spell available ('known') provide by that metamagic. (Quicken is even more pronounced: it effectively provides more spells known, /and/ it plays directly to the many spells/day strength of the sorcerer, as well) The reverse is true of item creation. Item creation directly addresses the wizard's weakness: the limitted number of spells/day that must be preppared. Making items makes more spells available for the wizard. Item creation is limitted by the need to expend gold, and, moreso, since you can buy items on the open market, anyway, by the expenditure of experience. The main difference is that, if a wizard takes metamagic, he benefits less from it than the sorcerer, but is not hit by the balancing factor (extra time) that the sorcerer faces. In contrast, if a sorcerer takes item creation, he's still hit by the balancing factor - exp expenditure - that is there to limit the wizard's (or any other prepped caster's) greater advantage in taking it.
Haldrik 04-22-04, 02:48 PM	Tony Vargas: Comparing thier weaknesses: the sorcerer's few spells known to the wizards limitted set of prepped spells. The statement juxtaposes few spells known versus few spells per day. I don't see the tradeoff occurring here. First Tradeoff: Many spells known with preparaton versus few spells known with spontenaity. Second Tradeoff: Few spells per day with normal spell levels versus many spells per day with staggered spell levels. (This second tradeoff proved disasterous for the Sorcerer, IMO. I think its long term negative consequences were more than intended because the staggered spell levels effectively denied the Sorcerer status as a primary spellcaster.) Anyway, keeping track of where the mechanics is happening is crucial for the structural analysis. As Ulden maintains: The 'no clear ratio' of spells known is a problem, but comparing the spells a wizard has prepped (per day) vs. the spells a sorcerer knows would be problematic because they are two different things. The reason I'm pointing this out is because these two weaknesses are not equal: the Wizard's few spells per day and the Sorcerer's few spells known. The Wizard's nerf per day is potent near 1st level but is softened by Intelligence's spell bonuses. The nerf dissipates by high levels because Wizards tend not to run out of spells anyway. By contrast, the Sorcerers few spells known grows severe. It starts low at 1st level, which is appropriate and not a problem. However, while the Wizard's spells known are growing (sometimes geometrically) each level, the Sorcerer's few spells known are actually tapering off, becoming acutely painful by high levels. The two weaknesses don't really compare. I agree there is a hiccup between what the Wizard knows and what the Wizard can cast, but scrolls, wands and magic items assauge this, and Ulden's premise seems reasonable to me: By measuring spells known on both sides, however, we can establish a 'cost' in terms of spells known for being able to cast spontaneously.
Tony Vargas 04-22-04, 04:17 PM	Originally posted by Haldrik You juxtapose few spells known versus few spells per day. I don't see the tradeoff occurring here. I don't think there is a direct 1:1 tradeoff going on at all. What the comparison between few known spells, and few prepped spells/day available does, though, is compare the most critical limiting factor on each class's casting ability. And, that strikes me as quite relevant. First Tradeoff: Many spells known with preparaton versus few spells known with spontenaity. Second Tradeoff: Few spells per day with normal spell levels versus many spells per day with staggered spell levels. I don't think you can sepparate it out like that. The slower progression of new spell levels impacts spells known, while the fewer spells per day impacts preparation. They're really one big tradeoff: Many potential spells known and few spells/day that must be prepared vs few spells known with many spells/day cast spontaneously and slower acquisition of new spell levels. As an Aside: A sepparate tradeoff might be: spells known dependent on expensive spellbook that can be lost or destroyed, but gains 5 bonus feats (up 4 of which can be spent to reduce that dependency) vs, well, nothing. But, that's not really relevant. The major limmitting factor on the spell power of the wizard's spellcasting is the comparatively small number of spell he can have prepared at one time. He might know a huge number of spells, but, at a given moment, he can only cast one of those prepared. The major limitting factor on the spell power of the Sorcerer is the relatively small number of spells known - he may be able to cast many spells per day, but, at any given moment, he can only cast one of those he knows. The reason I'm pointing this out is because these two weaknesses are not equal: the Wizard's few spells per day and the Sorcerer's few spells known. One of the insights I'm getting out of this discussion is that they may very well be very nearly equal - in balancing or limiting spellcasting ability, though, probably, not in some sense of absolute value, by themselves. They may be two differnt things, but they determine the same thing: which spells an arcanist has available to cast at a given moment. The Wizard's nerf per day is potent near 1st level but is softened by Intelligence's spell bonuses. The nerf dissipates by high levels because Wizards tend not to run out of spells anyway. I haven't played 3E wizards that much, but, I played 1 & 2E wizards extensively, and, I don't think this particular observation will have changed much: Wizards at high level rarely use up all thier spells in one day, but they very frequently run out of spells that are useful for the situation at hand. If you don't prepare a variety of spells (or keep a number of slots open), you may find yourself in a situation where you have no useful spells to cast - conversely, if you do, you will almost certainly find yourself in a situation where you've cast your best spells for the challenge at hand, and are really wishing you could cast them, /again/.
Haldrik 04-22-04, 05:18 PM	Tony Vargas: I haven't played 3E wizards that much, but I played 1 & 2E wizards extensively, and, I don't think this particular observation will have changed much: Wizards at high level rarely use up all thier spells in one day, but they very frequently run out of spells that are useful for the situation at hand. Running out of needed spells is still true in 3.5e as it was in 2e, but not nearly as bad. The new Wizard, mercifully, gets spell bonuses because of Intelligence. These extra spells are life-or-death at levels 1 to 4. (The bonuses make the difference between the Wizard's and the Sorcerer's spells per day less significant, IMO.) Plus, now the open slots allow the Wizard to actually use the entire spellbook, tho this applies to utility spells really. It depends on the circumstances but I leave about 1 out of 4 spell slots open for utilities and only prepare combat spells. (Dimension Door is a combat spell. ;) ) Again, it depends on circumstances. Finally, the concept of magic item creation is much, much better than it was in 2e. Wizard's are assumed to produce and amass magic items. When I make a Wand of Fireball, I don't need to prepare Fireballs and can take other combat spells in their place. Wand of Fireball doesn't really "run out". (In 2e, I normally had to *find* a Wand of Fireball in our campaigns.) Also, 3.5e hastens the rate of level advancement by reducing the xp reqs. This allows Wizards to enter *real* power soon without the painful "hazing" process that 2e forced on the Wizards who waited forever in excruciating vulnerability at low levels. "Wizard hazing" still exists but it's not quite as extreme. Now, life for the Wizard is much better and richer. Finally, writing my own scrolls really does compete with spontaneous spellcasting. By the way, I'm questing for ways to quantify the Wizard's power, and this is an especially interesting thread!