The Balance Thread

[Maledictus]

Valid points, thanks.

All this makes me wish we had an easier way to test these things. Say, a way to start a character at level X with all potential skill and stat points (up to level X) unallocated, and some cash for items. I suppose tinkering with XP formulas can allow a fast growing character, I'll look into that. A console approach for this would involve the dev and would allow exploits, which is the reason (I assume) it's not in there now.

[Bluddy]

Originally Posted by Maledictus

Valid points, thanks.

All this makes me wish we had an easier way to test these things. Say, a way to start a character at level X with all potential skill and stat points (up to level X) unallocated, and some cash for items. I suppose tinkering with XP formulas can allow a fast growing character, I'll look into that. A console approach for this would involve the dev and would allow exploits, which is the reason (I assume) it's not in there now.

Yeah Crisses changed the XP required to level in order to make a level 22 mage. I use him all the time for testing. Wish I had higher level mages as well, like a level 50, a level 75 and a level 100.

The nice thing about these high level characters is that they have so much gold that switching skills around is a piece of cake.

[Evander]

you can always overcheat experience and gold for testing purpose.

[Bluddy]

Originally Posted by Evander

For some reason, I always was able to solo a wizzy without problems.
Maybe thats due to the fact that I collect and keep for later a lot of items with mana/hp regen.

It was always possible, but you could only play a very certain way: Shatter was the most powerful spell, and a mastery always had to be brought to a high level.

But game is nearly unplayable in terms of "fun factor", especially in the multiplayer. Let's hope we will be able to help change that =)

I hope so, but it's not my first priority. Maybe multiplayer should have some handicap to make characters need each other. For example, mana regen could be made very low for everyone. In other words, once characters are sufficiently strong on their own, we can think of ways to make them work together for the sake of multiplayer.

Sorry, I meant that he could add logic and commands required only, so that it would not invalidate people mods. Balanced assets would then be able to invoke those commands when needed.

Ah ok. It'll be very hard to not invalidate mods though, what with all the possible changes.

It's just a personal opinion, I don't think that 1023 patch added so many important things, so that I should port all my personal modifications to this new version.

It did add several really important things as far as mage balance is concerned. If you use a comparison program like WinMerge, it's really not too hard to port your mod changes.

And this, this is a bug.

Actually I think it's a way to make the plate armor skill worthwhile. Not only do you get to have plate armor, you also get a boost to your armor. You have to remember that even characters with plate armor won't have all plate, but pieces of different kinds. The boost needs to apply to your total armor score.

I might be missing the point, but why bother changing the formula, instead of simply lowering skill points required to increase a level of that skill.
The result would be similar?

The formula is non-linear. Changing it a little makes a pretty big difference. You can find it being discussed in the Demon War beta forum, or see its effects in my excel book in the page 'DefOld'.

Unless you play Unlucky & Poverty character ;-]

Generally the amount of items dropping is something that also should be balanced at the very end of the process. with U P character I still drop legends from bosses, from time to time.

Yeah I played Poverty characters and didn't notice a huge difference. In my opinion we get way too many items in the game, but I've voiced this before and if Shadow agreed, he could have changed it by now, so I guess he disagrees. I really don't like how many items bosses, champions and elites MUST drop according to the game's rules. Once I create my own personal mod (ie. not a balance mod) I'll modify it to my heart's content.

As for the design choice, I prefer to believe that it was more of an accident
A "Hunter" class, instead of a Druid, with passive skills increasing damage to certain "beast like" monsters would be more in place here.

Perhaps. Regardless, that part of the game's not going to change. You can make a mod that'll replace that subclass if you want -- it's not hard to do.

Well, wasn't that supposed to be a major step forward from DoP? I think I saw that in an advertising...

Anyway, the question here for "I think every sub-class should be viable on its own" is why?

No subclass plays alone. And an attempt to make every subclass able to solo up to the lvl 100 is in my opinion a good way to break the game even further.
You must be very very careful here.

Right. I didn't mean that every subclass should be viable on its own (though I said it ) What I meant is that every subclass should be a viable choice -- none of them should look like they're useless. Currently, the magician is a very weak choice (just as an example).

Take a look at the warrior example again and it's subclasses.

Weaponsmaster is a damage dealer. Now it's combination of different skills and passive increase in damage.

Gladiator should be all about skills. Either AoE, stuns, defensive stances etc. but otherwise capable of delivering decent damage and avoiding taking so much damage. Right now it's about stances and some passive increase in crushing damage.

Defender is about, well, defence. Focusing on shield and heavy armour, that subclass could be a tank in DC. Say you have an area aggro skill, some shield stun/bash.

So you have a 3 possible ways to develop your warrior. Either go for an awesome offence and damage, but you will have to carry ton's of chickens with you.
...
Now, the point is, you can have all of it in one character (well, almost).
So you focus on, say one weapon type, invest in a single AoE skill, and boost defence for an extra protection.
...
You see the problem here? How are you gonna balance that, if you can play even pure classes differently?

I think the key is to make every skill about as useful as another skill. If a skill is more powerful, it should have a higher mana and skill point cost. Sure you could find some combinations that are more powerful, and that's fine. But there shouldn't be OP skills that dominate. And you need to make sure that a character that really wants to focus on a particular aspect (e.g. defense) will find enough skills to make that work.

As a side note, I believe that a warrior class itself should be redesigned. Weaponmaster should focus on damage/attack increase, rather than on so many activated skills. Gladiator should have them instead. While Defender should be focused on upping up Shield, Plate, Chain stats and have skills related to that (Shield bash, stun resistance etc).
Once I finish with UI mod, I will try to see if I can come up with some decent skills for every subclass so it would have more unique feeling ;-]

Go for it! I do agree that the fighters' skills feel samey to me. So many of them just play with the basic CRIT, CRUSH, Physical damage etc.

There simply must be a reference class. Sort of bottom/upper line for balancing everything else. As shown in an above example with a warrior.
It must be decided that for example "a warrior lvl 25 with no magical items and those skills selected must play that way", and everything else should be modified accordingly.
Otherwise it's just guessing/personal preference. Which is hardly a "balancing" =)

I think if we tried to create such reference characters we'd soon go crazy from the possibilities. Ultimately feedback for this stuff must come from testing that players do. Also, there's really no such thing as no magical items in this game -- there are so many magical items that it's impossible not to have them, and you have to count their effects. They have REALLY BIG effects. I did reduce the maximum effect on CRITs and CRUSHes though. That was out of control.

Well, yes. But using passive skills is a good way to make every class more unique.

Certainly, but so are active skills. And active skills don't disturb the balance of the game as much.

[quote]And what if you make a hybrid with a Defender. A lot of HP, a lot of DEF, polymorph into kodiak and be an unstoppable monster? Seems like possible OP ;-]

You're right. The skill and mana costs should be arranged in such a way that that would be a completely legitimate combo -- but it should cost you a lot. If that's where you choose to put all your points, you'll end up with an unstoppable monster for 1/4 of the time, and a fairly weak character the rest of the time. Again, this is why passives are so dangerous. The kodiak can be controlled using mana consumption and cooldown timers, making it effective only some of the time; but if you give yourself passive boosts, they're effective forever.

Perhaps. But we can afford having one class whose regen is better than others, if only to make it stand out. Of course it must be balanced in other areas.

It seems Shadow's intent was to minimize the use of health regen, and I don't see a need (at this point) to go against his choices.

Like I said. Not a problem for a regular items. For a SET ITEMS we may want them to have more consistent stats, just to make sure they will be awesome.

Oh actually I just checked and set items do have consistent stat changes inside statusEffectSets.gdb. So looks like Shadow thought the same way as you. I just think these effects need to be boosted. If you look in that file, some of those effects are excellent -- relative to the beginning of the game. But by the time you get the full set it'll probably be at least the mid-game. The combined effects are then very unimpressive.

Maybe Shadow was afraid you'd give the set to a character early in the game, messing up balance. If so, then sets should have minimum levels. You could say that you can't combine 2 pieces of a set before level 30, and that you can't combine 3 pieces before level 50 etc. Then, make the bonuses truly AWESOME.

But, back on the topic. I think useless effects should be removed from Sets as a part of rebalancing.

Yeah money finding should not be in there. Money is pretty much an unbounded resource in the game, and I see no way to change that. You can farm for so long in lower levels, or once the town is saved...

That COULD change if vendors had limited money though. That would completely change the economy of the game, making money much more scarce, and treasure rooms would actually mean something.

Not good. Turtle/Dragon scales for example. They are not metallic.
Better to leave it for now.

Good point! Then maybe just plate? Have a penalty for wearing the strongest armor?

Actually, in my games a lot of arrows/spells do miss. Thats the way I kill delven priests - by dodging their bolts as I run towards them :]

Yes they can miss if they're dodged -- but monsters don't dodge. What I mean is that the spell auto-aims for you -- it doesn't just go where you clicked.

But again, I really believe that player should be able to launch a barrage of missiles over certain area, rather than having to target every imp separately.

Right that would mean the missile going where you click without regard to where the enemy is. I think that's a pretty good idea.

In both cases, I would like to point to Lineage 2. While the game is not really playable now, the way the UI and buffs are handled in this mmo is a very good example to follow. Possibly the best of all mmos.
Perhaps you should take a look, as a part of research :-]

When I get a chance. I already got much deeper into this balancing thing than I originally meant to

I know.
Still as a modder you are only limited by your creativity and what is legal.
I pretty sure that if a new, very good design of classes/skills was made, Shadow would at least took a look. :-]

I think from the perspective of a game designer, there are only so many changes Shadow would make to the core game. I've located imbalances that need correcting, and I'm trying to correct them. I might boost certain skills to make them more worthwhile. But I recognize that if I make changes that aren't strictly necessary to restore balance, from Shadow's perspective it would be more like a mod that people can download if they wish rather than changes that would make it in some form to the core game.

Aanyway, about that evasion on the wizard? Think of it as a ninja-wizard style.
Helps you to get in between monsters to launch AoE nukes, and helps you to successfully accomplish tactical retreat if your mana is out =D

I think it belongs more in a hybrid class. A magician should have some cool way to get that effect. Plus, he already has teleport.

[Bluddy]

Just thinking 'aloud' about mana regen:

I'm wondering if I can find some value to make mana gain from passive abilities useful now. Looking at the different current abilities:

- The weaponmaster gains 1.5 mana from hitting.
- The gladiator gains 1.5 mana when getting hit.
- The defender gains 2 mana when blocking and 2 when parrying
- The assassin gains 3 mana from a crit and 6 from a kill
- The trickster just suffers less penalty from slowdown during combat. He gains an extra 50% during combat. So does the thief.
- The paladin has the same bonus as the defender.
- The archer gains 3 mana on a crit.
- The necromancer gains 6 mana on a kill.
- All casting classes get extra mana regen. Mages also get extra mileage out of mana potions.
- The healer is supposed to generate his mana via prayer.

All of these abilities make up for the fact that during combat, mana regen slows to 1/4 of its normal value. Notice that some classes, such as the hunter or the shaman, end up with no extra mana ability. I don't know why that is -- might be an oversight.

It's not hard to see that these values don't hold up throughout the game. A non-mage class can fairly easily reach 500 mana by level 100. At that level, they're likely to have a mana regen item, but remember that whatever bonus is given must be divided by 4 during combat. A gain of 1.5 mana from being hit, when 6 hits are enough to kill you, means that you're likely to get 9 mana from being hit (before you die), out of 500.

So as mentioned earlier, one solution is to get rid of this combat slowdown. Mana should refill at whatever regen rate you have. At the same time, these passive abilities are cool. For example, as a gladiator, it means that I have to fight some with my regular fighting skills before I can use my mana. As an assassin, I need to find ways to kill as soon as I can to get more mana with which to work.

Can we think of better values to use for the passive abilities? Well, the first thing to remember is that the first 10 levels of the game are extremely easy. There's no need for balance there. Around level 11, a non-mage will have around 80 mana. Critical hits and kill rates should increase over time, so saying that an assassin should get, say, 40 mana from a kill and maybe 20 mana from a crit seems like it could work. These are just rough numbers -- they need to be tested. It's harder to come up with absolute numbers for the warrior. As the levels go up, you need to hit LESS because your DPS quickly rises. You also can afford to get hit less. This means you'll be getting less and less mana with each passing level. Blocking and parrying will increase just as crits do though, so that one's ok.

There's a problem here -- the warrior benefits from all 3 passive abilities, even though he has just as much mana as a gladiator or a defender will. This means he'll gain mana much faster, which really isn't fair.

Now that I think about it, I think I'm pretty close to just ditching (ie. ignoring) these passive skills and simply making combat regen slowdown much less of a factor (say 50% or 75% of regular regen). I like the fact that between combat, mana regens faster so you don't have to wait around too much. But do we really need these passive abilities?

Take the warrior's rage gain on being hit. It's supposed to make sure that he doesn't just use his skills -- he has to fight some to get mana. But the same thing can be done (to an extent) by giving him a certain mana regen rate. The only difference is that a regen rate means he can use his skills, then wait for a while, then use his skills again -- there's no need to actually fight with his regular fighting ability. But then he'd be wasting time, which is probably the most important resource in the game. From a narrative perspective, we could say a warrior constantly builds up rage (the name of his mana type from DoP -- just like the Hulk). He lets it out when he fights.

What about the assassin's crit/kill bonuses? These don't really make sense in the first place. While they make sense thematically -- the assassin wants to kill and get crits, or he builds up 'momentum' (from DoP again), this doesn't really contribute in a gameplay sense. The assassin has to kill to get mana ie. to use his skills. But his skills are used... to kill and get crits. In DoP, I often found myself unable to kill enemies and stuck with almost no mana as a thief because I couldn't kill my enemies, so then I couldn't use my skills, which meant I couldn't kill my enemies, and so on. To make matters worse, mana constantly decreased in DoP.

Of course, you may be thinking 'what is he talking about here? I don't feel a need for these mana regen things in DC'. That's true, but it's true because the mana costs are so low. Once I adjust them, all these issues will crop up. And the mana costs need to be adjusted, because otherwise mana is just another useless resource we have tons of, and INT and SPR are unnecessary, and skills have much less to distinguish them from each other. Also, there's way too much spamming of skills going on, especially in the warrior classes, and not enough mixing of regular combat and skills. This again is because the cost of skills is too low -- it wasn't possible to spam as much in DoP.

OK so my conclusion after all that is that these abilities are probably not worth it. I'll adjust mana regen rates without considering these abilities (which are currently too minor to have any impact anyway) and I'll reduce slowdown during combat, which as of right now kills regen rates. I'd like to tune it so that mana potions are very helpful to mages during combat, but not so necessary for the other classes.

[Bluddy]

I've now added magician balancing. I'd really appreciate some testing feedback. Not sure if arcane drain is working properly -- I have to test it. It's supposed to make a weakness aura that only affects monsters.

[Evander]

Originally Posted by Bluddy

It was always possible, but you could only play a very certain way: Shatter was the most powerful spell, and a mastery always had to be brought to a high level.

Actually, I played some mmos in the past. There I learned that a combination of skills/spells works best.

Say, stun + power strike for a warrior, or a debuff+nuke for a wizard.

I hope so, but it's not my first priority. Maybe multiplayer should have some handicap to make characters need each other. For example, mana regen could be made very low for everyone. In other words, once characters are sufficiently strong on their own, we can think of ways to make them work together for the sake of multiplayer.

Let me put this in different words then.

There is a company, called Blizzard. Even after they released their Warcraft 1, people didn't cared that much about them.
Then there were Diablo 1 and Starcraft. Both games having awesome graphics, interesting story, artistic music. But what really made them word-renowned titles, was their multi-player capability.
Starcraft 1 is still played, 13 years after release it has thousands of fans worldwide, there are even public matches. So is Diablo, even tho Diablo 2 is now more popular.

Not to mention, that Blizzards biggest, most profitable game - World of Warcraft is multiplayer only :-]

Here is how it works:
A guy named, say George, finds a cool looking dungeon crawler. It has well-balanced skills, nice GUI, and good gameplay. It talks about that game with his friends and he manages to convince 3 of them to buy the game so they could slay hordes of monsters together.

In terms of game developer, this increased popularity equals increased profit, which they can use to develop more and better games :-]

You get my point? :-]

It did add several really important things as far as mage balance is concerned. If you use a comparison program like WinMerge, it's really not too hard to port your mod changes.

One day I'm going to merge my mods into next version, it's just not that day yet.

Actually I think it's a way to make the plate armor skill worthwhile. Not only do you get to have plate armor, you also get a boost to your armor. You have to remember that even characters with plate armor won't have all plate, but pieces of different kinds. The boost needs to apply to your total armor score.

But if there were skills that apply only to certain types of armours that would cause players to invest time and effort in obtaining them, rather than going with different pieces of equipment. Plus, skills related to them could be implemented, like increased movement speed and evasion for chain armours, resistances to crits/stuns for plates etc.

Yeah I played Poverty characters and didn't notice a huge difference. In my opinion we get way too many items in the game, but I've voiced this before and if Shadow agreed, he could have changed it by now, so I guess he disagrees. I really don't like how many items bosses, champions and elites MUST drop according to the game's rules. Once I create my own personal mod (ie. not a balance mod) I'll modify it to my heart's content.

To be honest, I think that U+P is how it should be in regular game. Given the quests to obtain sets and epics, players would be more inclined to actually go after these quests, if they wanted to have a set or an awesome weapon.

Perhaps. Regardless, that part of the game's not going to change. You can make a mod that'll replace that subclass if you want -- it's not hard to do.

I plan to try it in a future.

Right. I didn't mean that every subclass should be viable on its own (though I said it ) What I meant is that every subclass should be a viable choice -- none of them should look like they're useless. Currently, the magician is a very weak choice (just as an example).

Maybe. If the evasion was a buff, and armour melt was an AoE, and Teleport would be a DC version of Blink, then this class would be very good to support a necromancer. Have your minions enjoy Lich bonus as well as extra evasion and unleash their fury on a weakened enemies (despair and armour melt), with a necro safely nuking from behind and capable of quick retreat via teleport when in danger, that would make quite interesting hybrid.

I think the key is to make every skill about as useful as another skill. If a skill is more powerful, it should have a higher mana and skill point cost. Sure you could find some combinations that are more powerful, and that's fine. But there shouldn't be OP skills that dominate. And you need to make sure that a character that really wants to focus on a particular aspect (e.g. defense) will find enough skills to make that work.

Have you played Baldur's Gate, or any good MMO?
In BG2 especially, I enjoyed my wizzy casting web to incapacitate enemies and then set them on fire with fireballs - a moment before I unleashed a poison cloud on them.

Some skills could be meant to be used before others. In Lineage 2 necromancer class had a Gloom skill - one that decreased magic resistance of target by some incredible amounts - therefore, a necro was always a welcome addition to a party of nukers, as it helped them to kill mobs faster.

Or in a more recent example - Dragon Age 1 had it, so that you could cast Grease, covering area with oil and then set it ablaze with fire spells.

Perhaps this is something to keep in mind?

I think if we tried to create such reference characters we'd soon go crazy from the possibilities. Ultimately feedback for this stuff must come from testing that players do. Also, there's really no such thing as no magical items in this game -- there are so many magical items that it's impossible not to have them, and you have to count their effects. They have REALLY BIG effects. I did reduce the maximum effect on CRITs and CRUSHes though. That was out of control.

Not at all. Said warrior with nonmagical items bought from shops or picked up as a bottom line. A warrior with sets/artifacts as an upper line.
Same for wizzy/priest/thief.
8 cases, and all quite manageable :-]

Certainly, but so are active skills. And active skills don't disturb the balance of the game as much.

Well, that depends on how you will balance them, eh? :-]

Oh actually I just checked and set items do have consistent stat changes inside statusEffectSets.gdb. So looks like Shadow thought the same way as you. I just think these effects need to be boosted. If you look in that file, some of those effects are excellent -- relative to the beginning of the game. But by the time you get the full set it'll probably be at least the mid-game. The combined effects are then very unimpressive.

While their magical stats are always the same, the actual quality of an item varies. I happened to have a 2 pieces of a set chainmail, where difference in armor was more than 20.

Maybe Shadow was afraid you'd give the set to a character early in the game, messing up balance. If so, then sets should have minimum levels. You could say that you can't combine 2 pieces of a set before level 30, and that you can't combine 3 pieces before level 50 etc. Then, make the bonuses truly AWESOME.

Sets do have minimum levels, as well as other stats requirements. They do also have many useless effects, like find more money. And very few give you regen bonuses. I always skip set items over items with mana/hp regen. In fact, my shared stash is full of all types of armours, all of them have regen effect on them.

Also, different sets drop on different levels. On low levels you drop Ciglio's set (not bad for a low lvl wizzy), on higher an Aversion leather.
And because there are multiple sets on close levels, then you keep dropping different pieces from different sets. And of course they have different quality of an item, so you may end up with same pieces of the same set, which are becoming useless on your level.

Yeah money finding should not be in there. Money is pretty much an unbounded resource in the game, and I see no way to change that. You can farm for so long in lower levels, or once the town is saved...

That COULD change if vendors had limited money though. That would completely change the economy of the game, making money much more scarce, and treasure rooms would actually mean something.

Since the money get spent on skills and potions, there are ways to get rid of them, if you feel you are too rich for your own good.
The problem is, with so many items dropping everywhere, you don't actually get poorer. So the problem with money is the result of a problem with item drops. Solve the later and the former will solve itself.

Good point! Then maybe just plate? Have a penalty for wearing the strongest armor?

I would say that limiting evasion/attack speed/movement speed.
One might say, that wearing a cloth armour does not impede your movement, attack and doge ability; same as leather if you are trained.
But a chain armour may be heavy enough to slow you down a bit, and limit you ability to avoid attacks. And plate armour would do this even more.

So you can be a walking wall of steel, but not a jogging one; or you could decide to move fast as a wind, but would be cut to ribbons should the escape would not be possible

Yes they can miss if they're dodged -- but monsters don't dodge. What I mean is that the spell auto-aims for you -- it doesn't just go where you clicked.

Some do, as this demon fire spell, which is a bug imo. Other spells, like those from delven priests just go in straight line and you can avoid them.

Right that would mean the missile going where you click without regard to where the enemy is. I think that's a pretty good idea.

It would work well with the above example. An archer could run around, avoiding spells and launch arrows in the general direction of his enemies.

I think from the perspective of a game designer, there are only so many changes Shadow would make to the core game. I've located imbalances that need correcting, and I'm trying to correct them. I might boost certain skills to make them more worthwhile. But I recognize that if I make changes that aren't strictly necessary to restore balance, from Shadow's perspective it would be more like a mod that people can download if they wish rather than changes that would make it in some form to the core game.

You are somewhat right here.
But let me tell you this - in a large company, developing a game is a process involving sometimes a hundred of people. A project manager in such company has many minions under his command, and many of these are often players. Should he decide that opinions are needed on a certain game element, he would receive many, often contrary, but made from different points of view.

And there are usually testers, both within a company and working for a publisher, sometimes even from external testing labs. All of them report possible improvements, bugs, ideas.

For a small indie dev, a number of people working on a title is limited to something like dozen. Often having a similar view on things, giving their games to friends and families for testing, the feedback they receive may be somewhat limited.
Also, with only a handful of programmers indie must code in only what is possible for them, and not always what they would like to have in the end.

And there is problem with routine. For a devs, who know everything about their game, it is easy to miss the obvious because they simply won't pay attention. An example in DC is gui, which is something many people try to mod as it is simply impractical in vanilla.

Really, there is no better thing for an indie dev than a group of dedicated fans - like you - who do a lot of work that would not be done otherwise and who help improve the game further - as well as point out problems and good ideas.

I'm pretty sure that If someone came up with ideas/mods that would greatly improve game and gameplay experience, devs would took a look at them - because, hey, why not?

I think it belongs more in a hybrid class. A magician should have some cool way to get that effect. Plus, he already has teleport.

..which has a long reuse :-]
So you port out of group of monsters into the waiting claws of another one - the one which you may hope to evade.

[Bluddy]

Just updated with a new version of my mod. I finished the sorcerer.
I also increased regen during combat to 50% of regular regen. Let's check this out -- I may have to increase it more.

Next is necromancer, I think. Time to make those skeletons more worthwhile. I'll also have to nerf raise dead a little since it's too good as is.

[alstein]

Idea for raise dead, give it a timer? Timer would decrease with level.

Also limit it to 1, but have lich increase the limit for every 2 levels.

Might be too powerful of a nerf though.

[Bluddy]

The timer idea is interesting, but I like my dead to stay around, and I think that's what most people expect. Actually, what would ideally be there, is a constant drain on your mana the more creatures you have. Once you can't sustain the creatures, they disappear (or you just have no mana). This would balance out the problem with increasing numbers of summoned creatures: the creatures get more buff with levels, but there's also more of them, so they become very strong in terms of power. The problem then becomes that there's no way in the game to get rid of a pet who's draining your mana, so right now this is not practical.

As a simpler idea, raise dead should just raise really weakened creatures. Makes sense. It takes mastery of the spell to get undead who are really powerful.

Also, maybe there should be more or less separate controls of undead power and number of pets. Maybe only liche should increase their power, whereas buying points in the skill will make more of those undead but not any stronger (or just a tad stronger). This would balance out the effect. Also, if I make skill costs more expensive, you won't need to buy 4 levels of raise dead to get one more pet.

EDIT: alstein, I now see that you said "also limit it to 1..." which may be this idea except reversed. Sorry if I didn't get it right away.

I also need to think of a way to make demons better than the undead, because they can't co-exist with each other.

Now onto the subject of mana regen again. I thought about it some more, and it DOES make sense that a warrior will get regen from all the different abilities. A warrior/mage will have extra regen and mana from the mage part, so it makes sense to give something extra to a full warrior. So we're back onto the passive mana ability track.

Even cooler is that I tried out the idea of strengthening the passive mana abilities with each skill level in their particular tree, and it seems to work! This is truly awesome. I can now make those passive abilities matter! It's really exciting.

Another thing I realized from another thread, is that it's going to be very hard to make mage spells (ie non-scaling skills) worthwhile to invest in when you could invest your points in scaling skills instead. Non-mage characters can get high DPS without even using any skills -- just based on weapons. Their strength amplifies the damage, and their DEX makes them hard to hit and gives them crits. I think I'll have to nerf almost all non-mage skills somewhat to make the mage skills equivalent by comparison. Skills that give 0.4 crit per level are just way too powerful for example.