Sorcery and Wizardry – Cost Comparisons

Sorcerer Hat

Image via Wikipedia

Here we have an inquiry from “Greg”, which came in via Email. Personally, I prefer that questions go in the comments here; that way they don’t get lost in the spam and the answers are available to everyone, just as this answer here will be. Answers to questions of “why” do tend to get kind of long anyway.


I am mostly finished reading thru Eclipse again, and now I remember why I didn’t switch to it earlier. There are a few things that I do not agree with in the point costs. Primarily the spell-casting costs, and some of the 3.5 feats indexed in the back. While I agree that a Sorcerer should cost more in terms of how many spells are cast, there should also be a huge discount based on how few spells they actually can know. In addition, I don’t think the same cost should be applied to all the spell levels for each level equally. After all a 9th level spell is vastly superior to a 1st level spell. At the same time, the number of CP doesn’t change based on character level so in essence those same 24 CP are buying more powerful effects at higher levels. With a little work I think I could find a happy medium for my games.

Now, there are a couple of basic reasons why the Sorcerer progression costs what it does in comparison with the Wizards:

First up is simply back-compatibility; the preliminary base costs for Eclipse were established by breaking down the existing classes. Sorcerers simply didn’t get much except spells – and thus their spell progression was expensive.

This is, however, actually fair enough, as will be discussed below – although I’ll be generally discounting cantrips.

  • A basic Wizard gets (3 + Int Mod + 2 x [Level-1) spells (discounting cantrips). Thus a 20’th level wizard knows a base of (41 + Int Mod at first level) spells PROVIDED that he has never had his or her spell book destroyed or stolen. If that’s happened, a Wizard may not have any spells at all available. Making backup spellbooks is time consuming and expensive, but is a vital precaution for any Wizard.
  • A 20’th level sorcerer gets 34 spells (again, discounting cantrips) – and they can’t be readily taken away.
  • A basic Wizard who hasn’t suffered any spell book related accidents will thus get at least ten more spells to pick from, and quite possibly many more than that – but only actually has a base of thirty-six available spell slots which they must fill with preset spells. Even worse, some of them are likely to be duplicates and others may prove quite useless in any given day, in which case they might as well have not bothered with that spell that day.
  • In actual practice, a Sorcerer will – in any given day – often have access to more different spells than a Wizard, and has the flexibility to use a particularly useful one many times if he or she so desires. Spontaneous casting is an enormous advantage. A Wizard must rummage for spells, carefully protect his or her spellbook, and find out what he or she is likely to be facing in advance to prepare appropriate spells – and STILL often won’t have as many useful spells available as a Sorcerer.

Wizards may be able to obtain old spellbooks and copy them, or purchase scrolls, or trade spells – but that depends on the setting, and on the whims of the game master, and costs both money and time. Sorcerers may be able to purchase devices that convert their spell slots into spells they can’t cast too (In Eclipse these are built using advanced Spell Storing options, WOTC got around to introducing them some years later. You can find them in the Magic Item Compendium as Runestaves) but – like a Wizard – they can’t count on it.

Now Eclipse allows spellcasters to spend character points to get more spells. Wizards can buy them for 1 CP each – but then have that pesky vulnerable spellbook and preparation requirement to worry about. Sorcerers have to pay 2 CP each – but add those spells to their list of spells that are available on demand, rather than just gaining more possible selections to fill a limited number of slots with. That’s a lot more useful, which is why their cost is higher.

Just as importantly, a standard-build Wizard is already using Fast Learner, Specialized in Spells for Double Effect, to get their base of two spells per level past the first. For a mere 6 CP, a first-level Sorcerer can take Fast Learner too, and get one extra spell per level past the first – or reduce the effective cost of his or her spell progression to match the Wizard.

There are other ways of learning more spells of course, such as Paths and Metaspells – which cost the same for both Wizards and Sorcerers. That’s because a closely-linked set of thematic spells seems likely to be just as easy for an intuitive Sorcerer to pick up as for a Wizard to study and record in a tome of spells. For that matter, there are also options for spontaneous casters using the Wizard progression (and getting more spell formula than a Sorcerer but not so many slots) or for changing the Sorcerer progression to use spellbooks and pre-prepared spells.

We can also simply compute the cost by comparison. A level of the Wizard progression costs 14 character points – 11 for the spell progression itself, and 3 for the specialized base caster level to use it with. Now, a Sorcerer gets 50% more magic AND spontaneous spellcasting – obviously “worth” +50% for the extra spells and at least +50% for spontaneity. That gives us 22 points per level.

Now lets knock off 40% for the reduced flexibility. That takes the base cost down to 13.2 points per level. Seems reasonable so far.

That gives us a base cost for a level in the Sorcerer spellcasting progression of (13 points for the spells + 3 points for the specialized base caster level to use them), for a total of 16 points – just what it costs in Eclipse.

Of course, you can also build the progressions by buying Mana as Generic Spell Levels, buying the same Specialized Caster Levels, and buying spell formula.

Lets see…

  • A 20’th level Sorcerer has 273 spell levels worth of spells available. That would – on the average – cost 328 points to buy. That’s pretty pricey – but a Sorcerer can only use Arcane spells (and not all of those) and has to tie those levels up in a set of fixed spell slots. That’s Corrupted, for a total of 219 CP.
  • 20 Specialized Caster Levels cost 60 points.
  • 34 spontaneous spell formulas cost 68 CP.

That’s a total of 347 CP, versus the 320 CP that a Sorcerer is paying out. It costs a little more to build your own custom chart – and you don’t get bonus spell slots for high attributes. The standardized progressions are given a bit of a bonus since the characters are accepting a preset structure and are, quite literally, progressing – they’re following a course of study and building upon what they’ve learned earlier. The same goes for Base Caster Level, Base Attack Bonus, high end skills, and a lot of other things.

That’s important. It takes more effort to learn how things work and to build a proper basis for more advanced studies than it does to learn how to do a few things by rote and rule of thumb. Thus the spellcasting progressions yield better results in the long run than randomly picking up bits of magic – although randomly picking up bits of magic may be faster at first. Go ahead, take some Generic Spell Levels, Corrupted for use in your custom set of slots as prepared arcane spells only, take some Specialized Caster Levels, and buy some spell formula. That will even let you get a bit ahead of a progression-based spellcaster for a bit – but you won’t get bonus spells for high attributes and they, with their well-organized studies, will surpass you at higher levels. If you want, you can build a character with individualized access to particular spells at individualized caster levels, as in this example. That’s even less efficient though.

In effect, you’re paying something extra in advance for access to those high-end effects later on. In effect, low-level spellcasters are banking points – overpaying for minor stuff so as to save up for the major purchases later on.

Now, the argument that the higher-level sections of a spell progression should cost more than the lower-level sections – and not in the part-of-those-points-are-being-banked style of the previous paragraph – is reasonable enough. There are games that work that way, charging a progressive cost for attributes, special talents, and skill levels. In games like that a +1 in a skill might cost one point, a +2 another two, points for a total of three, a +3 another three for a total of six, and so on. This works fairly well, and is believable and natural. It does have some problems though; it greatly encourages every character to dabble in everything; in such a system it’s generally far more sensible to buy two skills at +10 (for a total of 110 points) than one skill at +15 (for a total of 120). One skill at +20 costs 210 points – enough to buy fourteen skills at +5.

Unfortunately, d20 levels do not work that way. The various classes all have an essentially linear construction. For example, a Ranger gets 6 SP per level with unchanging costs per skill level, eight sided hit dice every level, saves in a linear progression, and more – and a fair number of those things are not linearly effective. Base Attack Bonus provides iterative attacks (or simply damage multipliers if combined with modifiers such as Enhanced Strike/Crushing); the difference between +2 and +3 is not at all equivalent to the difference between +5 and +6. Similarly, the feats at the end of prerequisite chains are usually a lot better than the ones at the beginnings – but they’re all one feat slot each.

It would be possible to rewrite the system so that steps in a spellcasting progression cost progressively more – but if we do that, we’ll need to give out progressively more points per level at higher levels, since we know that basic spellcasters get one level of their progression at each level and still get their other benefits at a constant rate.

Wait, doesn’t that mean that I could skip taking a my next level of spellcasting at – say – level sixteen, where each new level might be costing three times as much as it did at first level, and spend those points on a huge pile of skills instead? Or perhaps buy first level Clerical, Druidic, Bardic, and Paladin spellcasting? That wouldn’t get me any big spells, but access to even some simple cure spells might be pretty handy at times – and they’d get used every single day. Could I use that slot I could have taken a high-end feat in to take two or three basic feats? Why not?

Wait a minute now, this doesn’t look anything like d20 any longer and I’m going to need some fairly complicated spreadsheets. To make it look like d20 again I’d have to write a huge pile of rules that would… effectively turn it back into a linear progression. That would be a great deal of work to accomplish very little. (There’s an article which covers Linear and Non-Linear skills and progression methods over HERE).

Of course, in other ways, the increased spell levels are almost irrelevant. Which is more effective – Sleep versus a trio of one-hit-die orcs or small monstrous scorpions (A CR of 1.5 or so against a first level caster) or Power Word Stun (a level 8 spell, with a fifteenth level caster) against a trio of 12’th level orc fighters or colossal monstrous scorpions (CR 15)?

Now, yes, the Sleep spell is very good for its level – but quite a lot of low-level spells are very good for their level, and quite a lot of high-level spells are not nearly as effective against opponents of a similar level as they might be. In actual play, a ninth-level spell may not actually be any more effective in solving an eighteenth level characters problems than a first level spell was back when they were level one or two. In fact, it’s often the other way around.

As for the shorthand conversions of 3.5 Feats in the back of Eclipse, I fear I’d need to know which ones Greg disagrees with before I could provide definite answers – and the listing is only there to direct people towards where to look to build exactly what they want anyway. It’s not really a guide to converting characters. Why bother converting characters? Eclipse is back-compatible; you can just use old characters directly.

From prior questions I do know that…

  • A couple of people have found the Armor Proficiency listings confusing since they list a total cost (3 CP for Light Armor, +6 for Medium, and +6 for Heavy) rather than the step-by-step cost.
  • The Great Cleave listing tends to confuse people, since it’s 12 CP cost is for building it directly – whether you have Cleave or not. (There are other ways to build Cleave too).
  • The Augment Summoning listing can be built as listed, but it’s cheaper to build it with a Specialized combination of Metamagic and the Streamline modifier. That took a lot more space to explain though.
  • Point-Blank Shot is simply wrong; It should be +1 Warcraft (Specialized in Missile Weapons Only, only from within 30′ range, 3 CP) and Augment Attack/+1 Damage (2 CP) for a total of 5 CP. That one either got mispasted or mixed up with something else.
  • The costs of many of the listed Feats – anything with an “*” on it – include the costs of the standard prerequisites that build up to them. A lot of people seem to miss that line in the first paragraph for some reason.
  • In general, pretty much any standard feat costs six character points. The ones with higher costs basically all have higher costs because they include prerequisite feats. This is, once again, part of back-compatibility; since standard d20 characters get interchangeable feat slots, virtually all feats must wind up with the same cost – or a few points less, since no one will complain about getting a few extra points.

Now, I hope that helps!

Eclipse: The Codex Persona is available in a Freeware PDF Version, in Print, and in a Paid PDF Version that includes Eclipse II (245 pages of Eclipse races, character and power builds, items, relics, martial arts, and other material) and the web expansion.

The Practical Enchanter can be found in a Print Edition (Lulu), an Electronic Edition (RPGNow), and a Shareware Edition (RPGNow).  There’s an RPGNow Staff Review too.

4 Responses

  1. On Behalf of “Greg”:

    NOTE: The ‘?’ are either ‘ or “, I tried to correct those before posting.

    I’ll break this up a little so it’s easier to read.

    Starting with spell caster levels and the cost disparity between clerics, druids, wizards and sorcerers, with primary concern on the Sorcerer and Wizard.
    First I agree that back compatibility is necessary when the aim is to stay within the rules boundaries of the d20 systems, otherwise you might as well design a whole new game system. I also agree with hand-waving 0-level spells as they are largely useless in combat beyond 2nd level.

    -Number of Spells known. From your response it seems as though you are saying a wizard will only ever have a maximum of 41+Int number of spells (not counting cantrips). This is simply not true. This is in fact the *minimum* number of spells a 20th level wizard will have recorded in his spellbook. Pages 178-179 in the 3.5 Player?s Handbook confirm this. In fact a Wizard, not a Sorcerer, is free to develop spells through research, and thus has no true limit to the number spells they can record. The only limit they face at this endeavor is money and time.

    -Practical Spell casting. Sorcerers can cast more spells during one day than a Wizard. The table clearly shows this. In many cases the Sorcerer does have some degree of flexibility over the Wizard in that he can cast the same spell numerous times with the need to prepare ahead of time. However the Wizard has some options that a Sorcerer does not. Primary among them is what I term Utility spells. Wizards by and large will have far more mundane Utility spells than a Sorcerer because they can afford to have them. With the horribly low number of spells Sorcerer can know, no Sorcerer will expend his precious spell slots on Utility spells that don?t have a historically large use in their GM’s campaigns. A Wizard on the other hand is free to scribe that “useless” spell into his book on the off chance it may come in handy one day. Also it seems that you missed that a Wizard can leave any number of spell slots ‘open’ when preparing his spells (pg. 178 PHB, Spell Selection and Preparation). This allows the Wizard to look-up his handy Utility spell and cast it when needed. Granted it will take him 15 minutes to prepare the spell and is not an option that will always be available, but it is still there. As an aside, I have never been in a campaign where a wizards spell book (mine or another players’) was ever destroyed or taken without some means of recovering it. GMs that do this are, in my opinion, simply being spiteful and not creative or mature enough to deal with a problem player/spell in a rational manner.

    So what does this mean from my point of view? I do feel that Sorcerers spell-casting is better than a Wizards in terms of availability and instant access to useful spells in combat. As, I said before however, I feel that Sorcerers should get a break in terms of cost due to the limited number of spells they can have. From my point of view, you are trying to justify the costs of the spell progression with a rule within your own book, namely that spell casters can by more spells. If the GM is using the point-buy system only to allow for a classless system but is sticking to the standard spell tables, then this justification falls flat. You also suggest Sorcerers take Fast Learner to gain more spells, but again that is an option that will not always be available and should not be used to justify a higher cost. The basic spell casting should be designed to reflect that. Options are exactly that, options. The core mechanic should be built on the assumption that no options will be used, and the options should be designed to integrate with that core mechanic.

    What I would like to see, and feel would be more in line with the core d20 mechanic, is the basic spell casting broken down based on total number of spells cast per day and total number of spells available to that caster. In the 3.5 PHB alone, not including cantrips) there are 352 spells on the Sorc/Wiz list (yes I counted), of which a Sorcerer can know less than 10%. If other WotC books are used, the arcane spell list soars over 2,000 spells. What I think would be a fair distribution is one CP cost for the Spells/Day progression of each list independent of the spell list itself. Then a different CP cost for the Caster level that includes the spell list, spontaneous vs. prepared, and in the case of limited casters (Sorcerers and Favored Souls) total spells known. That would allow for wizards, clerics, and druids to cast as many spells per day as a sorcerer, but at an increased cost. I do not like that you seem to be forcing the concept of Paths/Metaspells and the Mana system as a “better” alternative to spell casting. It may suit some groups, but not everyone will want it.

    I agree that trying to make a non-linear system for spells would not mesh with the rest of the d20 system. Its an unfortunate effect of the level scaling in general. That’s why a stated previously:
    “At the same time, the number of CP doesn’t change based on character level so in essence those same 24 CP are buying more powerful effects at higher levels.”
    I have played a variety of point-buy systems (HERO games), level systems (1st thru 3.5 D&D, Star Wars d20 Mutants & Masterminds 1st ed), and increased cost systems (L5R and 7th Sea). I have seen some flaws in all of them, I have also seen numerous advantages to each system within its own scope and genre. Overall, the level system works best for me in fantasy games. I love the flexibility of point-buy however which is why I picked up Eclipse in the first place.

    As for the Feats I’ll start with one that stuck out the most to me as being the most overpriced and is easiest to show why; Weapon Focus.
    In the book, Weapon Focus is listed as being built with Augment Attack. At its base it’s a 3CP ability. But for some reason there is a note saying the cost is increased to 6CP because it is assumed the character will use other weapons as well. This is way overpriced. Weapon Focus allows the character to get a +1 bonus to attack rolls with a single type of weapon. You could achieve the same result by purchasing Warcraft (Specialized: One weapon type, Corrupted: does not add to Base Attack) for 2CP.
    Weapon Specialization is another one that is easy to show. Again it notes the feat was built Augment Attack with the same note about using other weapons. Weapon Specialization grants a +2 bonus to damage with a single weapon type. Under Augment Attack it states that a +1 damage bonus only costs 1CP and a full d6 (average bonus +3.5) costs 3CP. By that logic shouldn’t Weapon Spec cost 2CP?
    Yes these are two minor points, but it implies that I would need to go through each and every feat and determine of the CP cost is correct. That’s a lot of work, especially once I start including other official D&D sourcebooks.

    This brings me to probably my biggest hang-up of the whole system. Almost everything in the book feels as if it is based off of a 6CP ideal. I do not like that. From my gaming experience of over 20+ years with D&D alone, I can say for fact that not all elements of a game system are equal. While reading through the book there were a few points (can’t remember which ones now) where I felt that you were forcing something to fit into the 6CP mold. I would much rather see the system designed from a variable point cost based on how important it is to a central mechanic. To me HERO is a good (if overly complex at times) example of this. In HERO buying one point of extra attack is worth more CP than buying an extra point of Defense. I agree with this idea to an extent. It does make it possible for player’s to achieve truly horrendous Defense scores without GM control, but any point-buy system by default must include GM control to prevent abuse, it’s the nature of the beast. I would be much happier to Eclipse use a similar approach. It does make for a more book keeping, but in my experience it also achieves greater play balance.

    Thank you for your time, Greg.

  2. A couple of those issues are my fault anyhow. Thoth wrote the basic rulesset, but I turned it into a comprehensive, and comprehensible, book for anyone to use. Along the way I made some assumptions which seemed justified.

    I assumed, for example, that people would be buying their armor skills right from the start. You can have whatever class you like or can imagine (or can bribe the GM into handing you), so it seemed certain that people would simply buy their armor skills right from the beginning. This broke the standard presentation format and was a mistake.

    We hemmed and hawed on how to present the feat conversions. On the one hand, we wanted to break down the entirety of advanced feats in such a ways that you could just double-check one, and then take it, even if it were an advanced feat. In Eclipse, there are no specific prerequisites, after all.

    But on the other hand, we recognized it could confuse people, hence the note saying that the more “expensive” feats were including previous functionality. So Great Cleave is 12 points, but you can just buy that straightaway – no pre-purchasing Cleave. I didn’t entirely like putting things down this way, but I do think that it remains the better option, whereas the armor listings were my screwup.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: