Removing Spellcasting as a Skill

[XuaXua]

Due to this thread on cross-school spellcasting, I came up with an idea for eliminating the Spellcasting skill, which currently (like Fighting) becomes a general dump for experience points.

My solution would be to transform Spellcasting into an ability based on the value of all known Spell Skills.

I only had a little in-thread feedback since my post was somewhat off topic, so I posted my idea in full (with some variations) here on the dev wiki. I would really love some feedback.

Do I need to post my proposal in full here, or is everyone OK with going there and posting comments or coming back here to talk about it?

Also, I believe a similar approach could be used to remove Fighting as a Skill.

[KoboldLord]

This has actually come up a couple times before in this forum, but the other threads got buried. I took the liberty of digging them up:

Making fighting and spellcasting into inferred skills first came up here:

viewtopic.php?f=8&t=276

Part of that problem, though, is that Fighting and Spellcasting are God Skills, and obviously so. With the hp and mp boosts on top of all their other benefits, they are extremely valuable even for the most inappropriate possible characters. Even a late-game Trog worshipper could use a little Spellcasting in the lategame to power evocation items that have an mp cost and the Guardian Spirit effect, so a few levels of Spellcasting from a non-penance source are going to be more helpful than a similar amount of xp into the already-high Evocation or Axe skills.

Perhaps Fighting and Spellcasting could be removed as skills and the functionality could be replaced with a sort of inferred skill? Instead of training Spellcasting as a discrete skill, for instance, your Spellcasting score is the average (or the average + a constant or a function of your level) of all your spellcasting school skills. Spellcasting maxes out when you've maxed out ALL your magic skills, and you cannot simply grind Spellcasting with spamming Magic Dart alone. Fighting, similarly, might be the average of all weapon skills, plus Armor, Dodge, and Shield. Or something like that.

The topic came up again in a more refined form here:

viewtopic.php?f=8&t=823

The underlying issue seems to be that the price of just one skill is not enough. So here is my proposal: remove the Fighting skill and do what that skill intended anyway: compute maximal Health from total melee skills (this includes Unarmed Combat). This contains two ideas at once: first, there is no need for an overarching fighting spell: the effect on fighting can very well be governed by the specific skill alone, and the effect on HP -- as well! (This is different to Spellcasting, which domes something beyond MP gain, namely spell hunger.)
Second, this mechanic provides an impetus to train a melee skill beyond its natural limit (when weapon speed is minimised). What is more, it even provides motivation to train more than one melee skill. (I'd think that the effect on HP should have diminishing returns.)
This proposal is not without problems. One is that the system would encourage even more sticking to the weapons of best aptitude. (This could be circumvented by using total xp spend in melee skills rather than actual skill levels, but that is more opaque.) Another is that cross-training suggests certain ladders of melee skills to work up. However, cross-training was invented in order to make attractive the practising of more than one weapon skill. Since this is completely achieved in the above system, cross-training could be removed.
Note that pure fighters generally have experience to burn, so training a second melee skill would not worry them. (With the current system, it would still be inconvenient to do, but I support the idea mentioned in this forum that high-end weapons should use more than one skill.) On the other hand, casters generally don't have superfluous experience to train several melee skills.

Anyway, I'm generally in favor of the idea, but the objection was raised that the idea makes it impossible to have lopsided but mechanically interesting designs such as the ogre's good spellcasting skill and no good schools. It also fairly heavily encourages dabbling in multiple skills that you don't otherwise have any use for, since lower levels are cheaper than higher levels.

[XuaXua]

This has actually come up a couple times before in this forum, but the other threads got buried. I took the liberty of digging them up:

Thank you.

The problem with both threads is that the problem gets explained, but the solution fails refinement process and gets dropped.

It also fairly heavily encourages dabbling in multiple skills that you don't otherwise have any use for, since lower levels are cheaper than higher levels.

Both proposals (an average) fail because they treat all skill levels equally. My system places a greater value on "high" skill levels, which contribute more to the Spellcasting Ability value (read the wiki entry).

the objection was raised that the idea makes it (1) impossible to have lopsided but mechanically interesting designs such as the ogre's good spellcasting skill and no good schools. It also (2) fairly heavily encourages dabbling in multiple skills that you don't otherwise have any use for, since lower levels are cheaper than higher levels.

Read the wiki entry I wrote as it addresses / resolves / invalidates both concerns.

After you read it and come back here, you will know the answers to the concerns are

1 - Ogres could be given a high racial-specific point multiplier; potentially x3 - x5 the existing multiplier
2 - Lower levels of various skills contribute much less / are less valuable than higher levels due to the fact that the multiplier increases as the spell school level increases.

The breakdown and balancing for the multiplier(s) is where the most work will come in; I am wondering if low levels (+0.2 Spellcasting Ability) should be decreased further, but I want non-spellcasters with the standard x1 racial multiplier to not have to reach Evocations: 10 just to get Spellcasting Ability: 1 (or maybe I do, now that I think about it). Maybe Evocations: 8, or possibly (yuck) have Invocations add. Having reading Scrolls train Evocations Skill should help increase the avenues of experience gain for Evocations.

Please let me know what additional concerns there are with this sort of system and I will help to resolve them.

[galehar]

What's wrong with the spellcasting skill? Before designing a brand new system, explain what's wrong with the current one and how your proposal would fix it. Personally, I really like the overall skills fighting and spellcasting. They certainly can use some adjustment and balancing, but I don't see how removing them would improve the game.

[KoboldLord]

What's wrong with the spellcasting skill? Before designing a brand new system, explain what's wrong with the current one and how your proposal would fix it. Personally, I really like the overall skills fighting and spellcasting. They certainly can use some adjustment and balancing, but I don't see how removing them would improve the game.

Fighting and spellcasting are unambiguously better than any of the associated weapon or magic skills. Having a god skill is troublesome in a system where a player can heavily influence which skill to improve, because the god skill is the unquestionably obvious choice. Once you have basic competence in the relevant weapon or magic skills, you have great incentive to pump ALL your xp into the god skills because the benefits are overwhelming. This aggravates the victory dancing issue, because getting the xp to go where you want it takes several hundreds of turns of tedious macros after every single significant fight.

XuaXua: While I support the idea of inferred skills in general, I'm not so fond of mechanics that can only be represented in the form of a table. Tables are notoriously hard to represent in non-obfuscatory code, and I wouldn't much like to be on the next generation of devteam and have to sort out some manner of obfuscated mess that probably seemed like a good idea at the time but nobody understands now. Simple is good, as long as it works.

[pratamawirya]

Fighting was something before it became godly. You remove Fighting, you remove the godliness and that something too.

[galehar]

What's wrong with the spellcasting skill? Before designing a brand new system, explain what's wrong with the current one and how your proposal would fix it. Personally, I really like the overall skills fighting and spellcasting. They certainly can use some adjustment and balancing, but I don't see how removing them would improve the game.

Fighting and spellcasting are unambiguously better than any of the associated weapon or magic skills. Having a god skill is troublesome in a system where a player can heavily influence which skill to improve, because the god skill is the unquestionably obvious choice. Once you have basic competence in the relevant weapon or magic skills, you have great incentive to pump ALL your xp into the god skills because the benefits are overwhelming. This aggravates the victory dancing issue, because getting the xp to go where you want it takes several hundreds of turns of tedious macros after every single significant fight.

XuaXua: While I support the idea of inferred skills in general, I'm not so fond of mechanics that can only be represented in the form of a table. Tables are notoriously hard to represent in non-obfuscatory code, and I wouldn't much like to be on the next generation of devteam and have to sort out some manner of obfuscated mess that probably seemed like a good idea at the time but nobody understands now. Simple is good, as long as it works.

They are good, but you still need to invest into your other skills. Having a boatload of HP is great, but you still need decent speed/accuracy/damage output to win. Same with spellcasting; having 50MP and 30 spell slots is great, but you're not going to clear zot with magic dart.
So yes, those skills are good, even too good. But we can balance that by nerfing them and/or increasing the cost instead of removing them. The HP bonus from fighting needs to be reduced (I'm working on that). The problem with spellcasting isn't that bad, but its effect on spell power could be reduced too. You can already VD almost all your XP into spellcasting, and while it's very useful early on, nobody does that the whole game.

[XuaXua]

What's wrong with the spellcasting skill? Before designing a brand new system, explain what's wrong with the current one and how your proposal would fix it.

I am answering this post before reading and responses because I read this post and then ran to the wiki to update the page at my earliest opportunity.

The rationale for the need was already on the page, but I added a section at the top which explains two inherent problems

1) that Spellcasting Skill acts as a a skill enhancer, not a directly usable skill

2) that Spellcasting Skill is an experience point dump that you should never turn off unless you don't ever intend to use spells.

3) (just noticed this minor problem it solves) If you use Scrolls, you will always accidentally train Spellcasting to level 1, which counts against you in the level total minigame.

[XuaXua]

XuaXua: While I support the idea of inferred skills in general, I'm not so fond of mechanics that can only be represented in the form of a table. Tables are notoriously hard to represent in non-obfuscatory code, and I wouldn't much like to be on the next generation of devteam and have to sort out some manner of obfuscated mess that probably seemed like a good idea at the time but nobody understands now. Simple is good, as long as it works.

Handing over development only works if you've coded something elegantly and have the appropriate comments in place. Granted, I've never looked at the crawl codebase, but I have seen some of the check-ins of code and cringe when I notice very situation-specific logic in places where core logic sits.

There's also a thing called documentation that always helps break in new entrants to a dev team.

[galehar]

What's wrong with the spellcasting skill? Before designing a brand new system, explain what's wrong with the current one and how your proposal would fix it.

I am answering this post before reading and responses because I read this post and then ran to the wiki to update the page at my earliest opportunity.

The rationale for the need was already on the page, but I added a section at the top which explains two inherent problems

1) that Spellcasting Skill acts as a a skill enhancer, not a directly usable skill

2) that Spellcasting Skill is an experience point dump that you should never turn off unless you don't ever intend to use spells.

3) (just noticed this minor problem it solves) If you use Scrolls, you will always accidentally train Spellcasting to level 1, which counts against you in the level total minigame.

1) Why is it a problem? It also cannot be trained on its own, only while using/training other magic schools. Just like fighting and weapon skills.
2) Evrry skill is an XP dump. That's the point of XP to be dumped into skills. You should never turn off spellcasting because it trains slowly. When we allow players to manually select skills to be trained, we'll see the limit of it.
3) Irrelevant.

[SinsI]

Fighting and Spellcasting skills are too vague, they do everything and nothing specific.
They are more like a "trainable attribute", they are not a "skill" - you can straight on replace them with HP/MP training and no one would really notice(except for spell levels).

It is also very weird that you can have Lvl27 in Ice magic with essentially no mana or ability to memorize spells (via Ash tempering).

I think they should be eliminated from the skill game completely, trained automatically based on XP, with the player able to set their ratio of training (i.e. 90%HP/10%MP).
Spell levels should be replaced with the abovementioned aggregate skill system.

[dpeg]

galehar: I suggest that we outright remove Spellcasting's effect on power and success. It currently does four things: hunger, slots, power, success. The first two are unique (hunger is coupled with Int, slots are coupled with XL), the latter two should be left to the proper skills.

[dolphin]

This should do something toward removing endgame casting uniformity, right? Do you have any idea how much? Or is the answer to be found in playtesting?

[dpeg]

dolphin: hard to answer. But it would move some power (in the sense of a character's ability to kill) from Spellcasting to the other magical skills. Since Spellcasting will still be very useful, it may help a bit. But don't expect too much!

The main reason why I propose this is because it's better, in my opinion, to have few but strong and noticeable effects of something on something else. If feature A also has major effects B and C but also slightly affects D and E, that is (a) hard to explain to players, (b) prone to misjudgement (by players and coders alike -- think of the gross overestimation of the GDR effect until very recently), (c) obfuscates code.

[galehar]

dpeg: ok, good idea. I'll discuss it on IRC and maybe we'll give it a try.

[absolutego]

you wouldn't be able to cast low-level spells of schools you have not trained relying on spellcasting alone. which makes sense. i don't think it'd change the endgame much, but it'd take longer to get there. say you want to grind for haste at excellent: you may need two or three more levels of charms. (there's one less skill to train, so more experience to share between the rest, but it's still a nerf.)

but the mid-game might be more problematic, as training school from scratch using a level-3 spell you've found in a book might be a pain. dunno. others may infer from experience alone; i'd have to playtest. but it sounds interesting.

[smock]

galehar: I suggest that we outright remove Spellcasting's effect on power and success. It currently does four things: hunger, slots, power, success. The first two are unique (hunger is coupled with Int, slots are coupled with XL), the latter two should be left to the proper skills.

You might consider moving hunger to depend on the spell's school skills as well. This would serve to discourage off-school high-level spells. Meaning that it would be harder for conjurers to cast haste, which would be a good thing I think.

[XuaXua]

Sticking with the theme of eliminating Spellcasting and based on feedback in this thread, I have consolidated some ideas together:

Integrate the effects of Spellcasting into each Spell Skill and make the concept on a per-cast basis, dependent on what spell is being cast.

Example:
My character has Conjurations: 4, Fire: 8.

POWER / SUCCESS (from dpeg)
Any Fire spell cast is treated as using Spellcasting: 8.
Any Conjurations spell cast is treated as using Spellcasting: 4.
Any Conjurations/Fire spell cast is treated as using Spellcasting: 6 (average the spells levels and round down fractions)

HUNGER
Hunger reduction should be based on Intelligence combined with the Spellcasting level specifically used for the spell that was cast.

MEMORIZATION SLOTS
Memorization Slots would be based on Intelligence combined with... a value based on the sum of a series of each Spell Skills divided by some number that is modified by the Racial Modifier (below)

example:
My character has Conjurations: 4, Fire: 8
Memorization Slots is based on
(Conjurations(1 + 2 + 3 + 4) + Fire(1 + 2 + 3 + 4 + 5 + 6 + 7 + 8))/ some number = 46/some number = number of slots

In this fashion, Spell Schools at level 1 contribute, but not as much as Spell Schools at higher levels.

- - - -
RACIAL MODIFIERS
To compensate on a per-race basis, a racial modifier can be added which enhances the level of Spellcasting. Racial Modifiers can only increase the Spellcasting value to 27, maximum.

For example (arbitrary values that are not set-in-stone are used in this example):

Ogres (to use KobolLord's point about them having high Spellcasting) can get a x2 modifier, so if he had Conjurations: 4, he'd use Spellcasting: 8 for any Conjurations spell cast.

- - -
NON-SPELLCASTERS
To compensate for non-spellcasters
1 - change Scrolls to allow reading a Scroll to train Evocations at any level (as opposed to just training Spellcasting to Level 1)
2 - Have every level of Evocations provide 10%-20% of Spellcasting ability, adjusted based on Racial Modifiers. Once a certain level of Evocations is reached, Spellcasting ability is activated and the character can read from books.
EDIT: 3 - Allow Intelligence to apply to Spellcasting qualification if it doesn't already.

Optionally, alter point 1 to
1 - change Scrolls to allow reading a Scroll to train the appropriate Spell School (likely Charms for most scrolls).

Questions I need answered to be able to clarify/modify the Evocations/Non-Spellcasters rationale:
1 - Does reading Scrolls currently train Spellcasting past level 1?
If no, for this proposal to work, we would need to change it to actively train Evocations at all levels.
2 - Does Spellcasting Skill affect Evocations Skill? I would think not.

[galehar]

This would serve to discourage off-school high-level spells.

What does that mean? You can't cast high-level spells if you don't train the relevant schools regardless of int and spellcasting. Casting without any school skill is only for low level spells. And you would need to train school skills even higher if spellcasting doesn't help anymore. No need to tweak the hunger cost.

[XuaXua]

edited my post (two above) to solve the memorization slots and hunger issues.

[smock]

This would serve to discourage off-school high-level spells.

What does that mean? You can't cast high-level spells if you don't train the relevant schools regardless of int and spellcasting. Casting without any school skill is only for low level spells. And you would need to train school skills even higher if spellcasting doesn't help anymore. No need to tweak the hunger cost.

Maybe I'm misunderstanding how spell hunger works. Currently it depends only on spellcasting, right? Assuming this is so...

Take the case of an EE that wants to cast Deflect Missiles in the late game. The EE will train up spellcasting, charms and air in some combination. At present, spellcasting does so much that he'll likely train spellcasting rather high instead of focusing on charms or air. This means that DM will have no or little hunger cost, because spellcasting is so high. If hunger depended school skills only, then he'd face higher hunger penalties. He might choose to sacrifice a level of spellcasting to gain a couple levels in charms and air.

You're right that there no need to tweak the hunger cost. But I think is would have an effect similar to moving power and success to school skills, in that it would make spellcasting a bit weaker. (I'd bet that leaving power and success as they are and moving hunger to school skills would help somewhat, but likely would not be as effective.)

So, I think it would discourage off-school high-level spells by keeping them expensive in terms of hunger. Reducing the hunger costs would require additional costs in terms of XP spent in the school skills.

I hope that's clearer. Let me know if not.

[XuaXua]

This would serve to discourage off-school high-level spells.

What does that mean? You can't cast high-level spells if you don't train the relevant schools regardless of int and spellcasting. Casting without any school skill is only for low level spells. And you would need to train school skills even higher if spellcasting doesn't help anymore. No need to tweak the hunger cost.

Maybe I'm misunderstanding how spell hunger works. Currently it depends only on spellcasting, right? Assuming this is so...

I believe it is a factor that is based on Spellcasting plus Intelligence.

In my opinion it should be based on some factor of the Spell School level(s) of the spell cast and Intelligence.

[vintermann]

Let's look at the factors one by one:

Slots: Currently matters little. Especially if you are not a caster - high-level warrior types can have all the nifty little utility spells.

Success rate: Currently matters for the very high-level spells, and for armored hybrids. Is effectively what keeps hybrids out of the top spells. It works well as it is today, really - although miscasting a spell you have at excellent and getting a super-nasty miscast effect (depends a lot on school) is annoying as there's so little you can do to avoid it. Like eating brown chunks, you just have to do it anyway. This leads to...

Miscast severity: Currently, it only depends on spell level and divine miscast protection - not on success rate, which can be a bit counterintuitive.

Hunger: Currently matters most for dedicated casters which cast in every combat. Matters less for transmuters, warpers, all hybrids except maybe reavers. Effectively keeps dedicated mages from spamming their top-level spells, something we want to keep. Impact of spell hunger varies hugely by race.

Power: Currently has highly unequal impact on the various schools. Absolutely essential for hex-casters, insignificant for warpers.

Then it's what you have to help with it:

Intelligence: Currently extremely useful. Helps with everything except miscast severity.

Spellcasting skill: Helps with hunger, which is extremely useful for some builds. Also helps with power (but power often doesn't matter) and success rate.

Spell school skills: Currently have dramatic effect on success rate, and are important for power (but power often doesn't matter).

Level: Crucial for MP. Matters for spell slots.

How can this be rearranged for a better game?

I suggest taking away the effect of level on spell slots, making it rely on intelligence and spellcasting instead. That way, a fighter who picks up spellcasting late might learn a few useful low-level spells - but not all of them. Spell power should depend on spell school skills, and the modest contribution from spellcasting should be removed to make that skill less universally useful. Success rate should depend on spellcasting skill and school skills, and miscast severity should use success rate as a factor along with level. Hunger I tink has to stay as it is, much as I would like to get rid of int from it.

[galehar]

Miscast severity: Currently, it only depends on spell level and divine miscast protection - not on success rate, which can be a bit counterintuitive.

This isn't true. The miscast severity formula is:

  Code:

severity = (pow * fail * (10 + pow) * 3 / 1400;

I have tried to simplify the formula so it's possible I've made a mistake. But the factors are pow and fail.
pow is the spell level and fail is the amount by which you missed your success roll. So by having a high success chance, you reduce the possible range of fail and the probability of it being high. Spell level weight much more in the formula of course.

[Stormfox]

Miscast severity: Currently, it only depends on spell level and divine miscast protection - not on success rate, which can be a bit counterintuitive.

This isn't true. The miscast severity formula is:

  Code:

severity = (pow * fail * (10 + pow) * 3 / 1400;

I have tried to simplify the formula so it's possible I've made a mistake. But the factors are pow and fail.
pow is the spell level and fail is the amount by which you missed your success roll. So by having a high success chance, you reduce the possible range of fail and the probability of it being high. Spell level weight much more in the formula of course.

Thanks for clarifying that. I've always felt intuitively that miscast effects seemed to be worse (or had a better chance of being worse, anyhow) if you had a greater chance of failure, but have been too lazy to source-dive.

[XuaXua]

Let's look at the factors one by one:
...
How can this be rearranged for a better game?

Thank you. I didn't consider all the factors. I incorporated your suggestions (with XP not affecting slots and others).

I am now leaning towards the second proposal, which transforms Spellcasting Ability (formerly trainable Spellcasting Skill) into a situational value (meaning: primarily based on Spell School of the spell being cast), and can be applied as follows:

Racial Aptitude:
This value increases Spellcasting Ability for a race with an affinity for magic. It was formerly the factor which modified experience for Spellcasting Skill.

Spellcasting Ability:
This could be a situational value based on an average of the Spell School levels used to cast a given spell, modified by Racial Aptitude, modified by Intelligence, modified negatively by knowledge of any opposing Spell School levels (remove the opposing spell school learning penalty).

Success rate:
This could be a value based on Spellcasting Ability compared to Spell Level.

Miscast severity:
This could be a value based as it was; on the failure value, increased by Spell Level and reduced by Miscast Protection

Hunger:
This could be a value based on the Spell Level and Spellcasting Ability. Could potentially / optionally be altered by spell failure.

Power:
This could be a value based on the Spell Level and Spellcasting Ability.

Slots:
This could be a value based on the sum of the sequences (is that the correct mathematical term?) of all learned Spell School levels, plus some factor of Intelligence, and/or modified by Racial Aptitude. By using a sequences (level 2 = 1 + 2 = 3, Level 5 value = 1 + 2 + 3 + 4 + 5 = 15), we increase the contribution value of higher level Spell Schools.

I don't know the current formula for determining Slots, so I can't finish this equation.

Magic Points
This could be based using the same mechanics to determine Slots, also modified by Experience Level.

I don't know the current formula for determining Magic Points, so I can't finish this equation.

Intelligence
As noted, could become part of determining the Spellcasting Ability and is used in the formula to determine Slots and Magic Points. Contribution to Spellcasting Ability could be adjusted on a per-school basis if desired.

[Night2o1]

I have a proposal for the removal of Spellcasting as a standalone skill somewhere in the dev wiki.

I hope it doesn't get picked up.

Spellcasting vs spell schools is an actually interesting mechanic and doesn't need to be dumbed down. Honestly I cannot even comprehend a way in which removing spellcasting would improve the variety, depth, or fun of the game. If you haven't noticed, different species have very different routes available to them because of the current design which actually differentiates the species from one another. Why do so many people make game suggestions that only benefit the particular playstyle that they enjoy..? Why don't you consider the implications to other playstyles and the overall objectives of the game design (replayability, player immersion)?

[XuaXua]

I have a proposal for the removal of Spellcasting as a standalone skill somewhere in the dev wiki.

I hope it doesn't get picked up.

Spellcasting vs spell schools is an actually interesting mechanic and doesn't need to be dumbed down. Honestly I cannot even comprehend a way in which removing spellcasting would improve the variety, depth, or fun of the game. If you haven't noticed, different species have very different routes available to them because of the current design which actually differentiates the species from one another. Why do so many people make game suggestions that only benefit the particular playstyle that they enjoy..? Why don't you consider the implications to other playstyles and the overall objectives of the game design (replayability, player immersion)?

Yes, I proposed specifically for the playstyle that I enjoy. That's exactly and precisely what I did.

The proposal is to distribute the concept of Spellcasting across the various spell schools.

[Night2o1]

I have a proposal for the removal of Spellcasting as a standalone skill somewhere in the dev wiki.

I hope it doesn't get picked up.

Spellcasting vs spell schools is an actually interesting mechanic and doesn't need to be dumbed down. Honestly I cannot even comprehend a way in which removing spellcasting would improve the variety, depth, or fun of the game. If you haven't noticed, different species have very different routes available to them because of the current design which actually differentiates the species from one another. Why do so many people make game suggestions that only benefit the particular playstyle that they enjoy..? Why don't you consider the implications to other playstyles and the overall objectives of the game design (replayability, player immersion)?

Yes, I proposed specifically for the playstyle that I enjoy. That's exactly and precisely what I did.

The proposal is to distribute the concept of Spellcasting across the various spell schools.

I did respond without having looked up/read your suggestion, sorry about that, but it turns out my point was correct. The change you suggest would infact reduce species differentiation and design flexibility. Using the assumption you make (that it is optimal to always have spellcasting on as a magic user), let us contrast the playstyles of an Ogre and a Human of the Fire Elementalist background. I'm going to make up the level differences, but the point will stand:
Levels 1-5
Ogre is spamming flame tongue on everyone
Human is spamming flame tongue, but gets throw flame online at a good cast % 2 or 3 levels earlier

Levels 6-9
Ogre is spamming throw flame, and has to use a lot of conjure flame in order to control the battlefield. They have more mana than the human. The ogre can't use sticky flame at a reasonable % yet, and his damage is way less mana efficient. Ogre has more spell levels available and might of lucked into one or two other low-level utility spells which are helping him along, such as repel missiles.
Human is using sticky flame and raping everything by level 9. He doesn't have any spare spell levels.

Levels 10-15
Ogre finally gets sticky flame online at a good spell success. He is doing well, but he is probably progressing slower because he has to retreat from groups of monsters more often than the human..

Human has fireball online and is much more easily aoe nuking whenever he encounters a group.

Your proposed change would tie the Ogre into being a "Shitty at magic" character who played exactly like the human except weaker in every way. With the Spellcasting skill there, differentiating the species, ogre actually has the opportunity to make up for his weaker spells by adjusting his playstyle and having room to pick up more utility spells/learn the whole spell book.

Simply, there is no harm in spellcasting being "a skill that is always optimal to 'waste' exp in". Your character is not in competition with others. It is a single player game. What the spellcasting skill does is provide a mechanic which can be adjusted in order to differentiate species and play styles.

I hope this sort of explains what I mean when I accuse people of blindly making suggestions to tune the game to their own playstyle & perceptions without first considering game development.

[XuaXua]

I did respond without having looked up/read your suggestion, sorry about that, but it turns out my point was correct. The change you suggest would infact reduce species differentiation and design flexibility.

Did you skim it and disregard the section on "Racial Modifiers / Aptitudes" adjusting the "Spellcasting Ability Point Multiplier"?

[varsovie]

I'm not warn nor cold to this idea.
A problem I do see actually, is that a caster will have the incentive to learn a spell he don't need only to get a low cost boost in his spellcasting.
The only things I hate with magic system actually is that it is too powerfull compared on melee (but still can be tweaked).
And also the "spell slot" that is nearly never a concern past XL4 and anyway capped at 20, and you can reach that before finishing the lair.

[crate]

I'm not warn nor cold to this idea.
A problem I do see actually, is that a caster will have the incentive to learn a spell he don't need only to get a low cost boost in his spellcasting.
The only things I hate with magic system actually is that it is too powerfull compared on melee (but still can be tweaked).
And also the "spell slot" that is nearly never a concern past XL4 and anyway capped at 20, and you can reach that before finishing the lair.

You can cast spells and hit things with a weapon also, unless you worship Trog.

Melee is not underpowered at all (personally I think killing stuff with tab is actually more powerful than spells, since it is much harder to end up in a situation where you are suddenly in trouble that you did not expect when you can kill stuff while spending 0 MP), it is just significantly harder to learn to play well than just killing everything with fire is.