Which skills are good to train all the way?

[danr]

So I'm learning that some skills are good to turn off after a certain point because there are diminishing returns.

My question now is - which skills, assuming you are using them, are good to train all the way to 27, and leave on permanently. I think the following:

Unarmed Combat
Fighting
Dodging
Armour
Spellcasting
Conjurations

Other possibles:
Stabbing?
Stealth?
Shields?
Throwing?
Evocations?
Invocations (depending on which god?)

Again, this is assuming that you are using the skill in question as a regular part of your game.

[szanth]

What, do the magical schools have diminishing returns?

[danr]

What, do the magical schools have diminishing returns?

Well, once spell power and success are maxed out, what more would they do?

Weapon skills do, as mentioned elsewhere - once you get to a certain point your speed maxes out and the damage bonuses are not that great.

[starless]

What, do the magical schools have diminishing returns?

Well, once spell power and success are maxed out, what more would they do?

Weapon skills do, as mentioned elsewhere - once you get to a certain point your speed maxes out and the damage bonuses are not that great.

That is true, but remember if you switch to heavier armour/shield, the requirements increase so you might end up needing a few more levels in those schools of magic.

[styrix]

So I'm learning that some skills are good to turn off after a certain point because there are diminishing returns.

My question now is - which skills, assuming you are using them, are good to train all the way to 27, and leave on permanently. I think the following:

Unarmed Combat
Fighting
Dodging
Armour
Spellcasting
Conjurations

Other possibles:
Stabbing?
Stealth?
Shields?
Throwing?
Evocations?
Invocations (depending on which god?)

Again, this is assuming that you are using the skill in question as a regular part of your game.

I've noticed that with Stabbing, once near 10, everything dies in one hit. Playing an SpEn, I've had it off since 4, and because a lot of things I kill with stabbing, it is still leveling up pretty quick.

I'm especially curious about Stealth. How effective can it get, especially with multiple items of Stlth++? Can it be invis-like, especially against SInv baddies?

[acvar]

Stealth keeps providing tangible benefits all the way to 27, and there is indeed no limit in practice to how stealthy you can be. It's a pain in the butt to train, though.

But stealth simply is not an effective skill past very low levels. Do you train stealth from 13 to 14 or do you train enchantments up to 12 so you can cast invisibility (and haste). Do you train stealth up from 20 to 21 or do you train translocations all the way up to casting controlled blink. Invisibility and controlled blink are far better for stabbing and the general avoidance of detection then stealth. The experience system is broken because the cost of each new level advances geometrically while the benefits of each new level advances linearly. No skill is "effective" to train up all the way unless you are training all of them up all the way. It is true that some like short blades, enchantments, poison magic, and others have absolutely no use past a certain level, but that does not mean that others skills are "good" to train up to high levels.

[danr]

I don't think it's as simple as that. The XP needed to train even lower level skills increases as your total level of skill levels increases. When you are XL 20, the XP it took to raise short blades from 0 to 15 would not raise another skill to 15.

Also, you have to take into account the fact that at higher levels you are killing monsters that provide 1000X the XP than they did at XL 1. It is not "broken" that it takes more XP to raise levels later in the game. Every RPG is like this. 10-15 XP gets you to XL2. It does not follow that 1000-1500 XP should get you 100x the improvement that XL 1 -> 2 represents.

Something is "broken" if it negatively affects gameplay. While the skill system would be designed differently by different people, it works well in that it is central to the game and lots of people seem to love the game. If you think it's broken, fork the game and do it right as you see fit, or just don't play it. That, or at least give a clear illustration of how it is "broken".

Anyhow, this is all off-topic.

For example re stealth. I have it at level 14, and that makes me "extraordinarily stealthy". Can I improve on that much more? I'm turning it off now.

[evilmike]

For example re stealth. I have it at level 14, and that makes me "extraordinarily stealthy". Can I improve on that much more? I'm turning it off now.

You can improve on that actually. The highest level of stealth that the game reports is "uncannily stealthy", and below that is "incredibly stealthy". And there's really no upper limit on stealth; the game just doesn't have any adjectives for extremely high values past uncannily.

The question is, does your character benefit from being extremely stealthy? Even if you're not a stabber, the answer might still be yes.

[danr]

Well, I just entered vaults, and I've stabbed at least half the monsters in V:1. Giants, dragons, etc - all down in one quick stab! Definitely worthwhile!

[ryak]

Getting back on topic, it is kind of spoilery that ideal play is to turn off certain skills above a particular level but leave others on all the way to 27.

I think weapon skills are clearly the worst offender here. Is there an easy solution to this? Maybe improve the contribution to damage from higher skills in weapons which have hit their speed cap?

[KoboldLord]

So I'm learning that some skills are good to turn off after a certain point because there are diminishing returns.

My question now is - which skills, assuming you are using them, are good to train all the way to 27, and leave on permanently. I think the following:

I think the question is not meaningful without the context of a real-game situation. Most or all skills give *some* benefit all the way up to 27, even if this benefit is increasingly marginal. Some benefit is better than no benefit at all, so it isn't like you'd leave xp in your pool rather than improve your existing skills past the point of diminishing returns.

The question, then, is whether it's worthwhile to continue improving a skill in comparison to some other skill, which will change depending on where you are in the game and what the benefits of that other skill would be.

For instance, suppose you're a conjurations-focused spellcaster. You're going to have conjurations pretty high, and it gets slowly better as it gets higher. You've got translocations at 4 because you found Blink earlier and you already raised it to Excellent. You're hoping for Controlled Blink eventually, but you haven't found it yet. In this situation, more translocations give zero benefit so you might as well stick with conjurations even all the way up to 27. Putting more into translocations is a gamble that the desired Controlled Blink spell will indeed show up. If it does, you're already done with some of the work. If it doesn't, that extra work is wasted and you don't have the marginal benefits of the extra conjuration skill.

Even Fighting has marginal benefits toward the end when you consider the hp boost as a percentage function of your previous hp. A Trog-worshipper thinking about raising fighting from 24 to 25 for Zot should seriously consider bootstrapping Traps or Evocations or Throwing/Bows/Crossbows instead if they haven't already done so, because those are seriously valuable utility effects.

A real-game situation is, of course, going to be more complicated because you rarely have a situation where Controlled Blink is the only utility effect you might want that you don't already have the ability to use near-perfectly. The caster above is going to spend most of the game greedily trying to gather spells from many schools, like Abjuration and Controlled Blink and Haste and Revivication, and until the caster decides that Fire or Ice Storm is the next effect they want to add to their arsenal Conjurations is probably going to have to wait in the back of the line.

[lucy_ferre]

For a HaBe? Fighting, Dodging, Shields, and maybe Evocations.

[ryak]

I recommend leaving Stealth on too. Not only does Ha have a good aptitude for it, but if youre going dodging focused, youre going to be in reasonably light armour anyways.

Being able to pick your battles will save your life.

[danr]

Thanks.

Yeah, I'm enjoying the stealth - I was able to stab most of the denizens of Vaults:1.

What about throwing and slings? Do those keep getting better / doing more damage or is more skill mainly about speed there as well, and it tops out?

[lucy_ferre]

A sling of velocity with highly enchanted steel ammo deals a lot of damage. And halflings are great with slings. I saw one guy during the tournament paralyze Cerebov with a needle, so throwing seems to remain useful for the entire game.

[danr]

Does damage output increase significantly with slings skill? And does throwing skill help sling skill? I'm only using slings, and it's training throwing, should I turn throwing off?

[szanth]

Throwing -is- slings, isn't it?

[Un67]

No, slings is it's own skill. It does crosstrain with throwing, though.

[LunarHarp]

What, do the magical schools have diminishing returns?

Well, once spell power and success are maxed out, what more would they do?

Weapon skills do, as mentioned elsewhere - once you get to a certain point your speed maxes out and the damage bonuses are not that great.

Magical Schools do in fact have diminishing returns. After a certain amount of power (I don't remember exactly where sorry) spell power is recalculated by scaling back your power after a certain point. In order to max out a spell with a max of 200 power you have to effectively get more then 700 spell power which is very hard to do. That's not to say that training magic schools don't keep providing tangible returns. However, some builds, especially those with lower int, there are better things to train once you can reliably cast those mid-level spells.

I personally like keeping the stealth skill on. Invisible is awesome but the biggest problem enemies in zot can all see invisible.

[Kate]

Stepdown starts being a factor as soon as you hit 50 power - learnDB says: "spell power[1/5]: (Spellcasting/2 + 2*avg school skill) * INT divided by 10 * (1.5 ^ enhancer count). Now halve the part over 50, halve the part over 100, halve the part over 150, and plain cap at 200."

To actually max spell power you need 27 spellcasting and schools, 35 int or so and three enhancers, if I remember right. Luckily this doesn't really matter because for plenty of spells (Enchantments at least) the spell power doesn't really matter past a certain point (such as the point where you can reliably cast Haste). For most Translocations power doesn't even have any effect at all so again, it's just a matter of getting your highest level spell (Controlled Blink) reliably castable. With straight-up blasty conjurations it's definitely fine to keep training all the way to 27 for more damage, though.

Also, acvar's post is hilarious, good job.

[danr]

No, slings is it's own skill. It does crosstrain with throwing, though.

Well, I have hardly thrown anything. I've been using a sling all game, and now my throwing is at level 7 while slings is at 15. It doesn't just cross-train, it actively trains when you use slings.

My question is - since Throwing trains when you use slings, does it enhance your ability with slings, how far, and is it worth keeping on (I actually have a manual of Throwing lying around somewhere too yet).

[galehar]

No, slings is it's own skill. It does crosstrain with throwing, though.

Well, I have hardly thrown anything. I've been using a sling all game, and now my throwing is at level 7 while slings is at 15. It doesn't just cross-train, it actively trains when you use slings.

My question is - since Throwing trains when you use slings, does it enhance your ability with slings, how far, and is it worth keeping on (I actually have a manual of Throwing lying around somewhere too yet).

No really, using sling doesn't train throwing. Just checked in debug mode. What version are you playing? Have you used evaporate?

[jackalKnight]

If you are a caster, you pretty much have to train Spellpower as high as possible for the sake of reducing hunger cost. The other casting skills you need train as high as necessary as possible for the casting rate to become "excellent". But Hunger is a huge threat, since it can prevent casting at the worst possible times.

[danr]

No really, using sling doesn't train throwing. Just checked in debug mode. What version are you playing? Have you used evaporate?

Okay, maybe I'm just really confused.

Will slings bring throwing to L1 at least, like reading scrolls does for spellcasting?

I'm really confused. Maybe it was ALL from that manual of Throwing I found.

[galehar]

Will slings bring throwing to L1 at least, like reading scrolls does for spellcasting?

No.

[danr]

Vastin, I split your post off into its own topic in the game design forum. It was a bit OT here.