Skill cost changes

[galehar]

Some of you may have noticed some oddities when training skills in trunk. This is because the skill costs have been changed. If your interested in the technical details, you can get them by reading the commit messages and the wiki, but it's not necessary. The goal of the change is actually to simplify this. Previously, the total cost in XP of training skills could vary depending on the order in which skill were trained, this is no longer the case. The change has a number of side effects:

• Gaining new skills is a bit easier and cheaper.
• In mid-game, low level skills are a bit more expensive, and high level ones a bit cheaper.
• In the early game, magic skills (including spellcasting, invocations and evocations) had a big discount. This has been removed and they cost now the same as other skills.

[snow]

So to be clear would 20K experience gain the same number of fire magic levels regardless of your total experience?

[galehar]

So to be clear would 20K experience gain the same number of fire magic levels regardless of your total experience?

No. Skill costs still depend on your total amount of skill points. It doesn't depend on the skill level anymore. For example: in the old system, a human XL 12 has 7k XP and victory dance it into axes which he has at level 0. The XP will be converted into about 2400 skill points raising the skill to level 9. If on the other hand he trains his long blades skill which is already at level 15, he will gain only 780 skill points, raising the skill by less than a level. In the new system, the 7k XP will be converted into 1750 skill points, regardless of which skill are trained. A level 0 skill would be raised just short of L8, and a level 15 skill would be raised to 16.6 level.

[Nobody]

The early game is still relatively easygoing, but the difference in skill levels seems to make the section between Temple and [Orc/Lair] noticeably harder. Melee combat against snails and similar enemies takes longer, attacks don't hit as often and do less damage, et cetera.

But the main difference is in primary casters. Fewer ranks in the Spellcasting skill means fewer magic points, fewer available spell levels, reduced success rates - they're just generally worse at everything during that part of the game. Especially for characters whose casting aptitudes are already mostly bad, like OgWz. This could be due to the new experience system or the removal of discounts (or just me complaining about new things, which is beginning to seem more and more likely as this post grows longer), but it really does feel like casters took a solid whack with the nerf bat at some point.

[MonorailCat]

But the main difference is in primary casters. Fewer ranks in the Spellcasting skill means fewer magic points, fewer available spell levels, reduced success rates - they're just generally worse at everything during that part of the game. Especially for characters whose casting aptitudes are already mostly bad, like OgWz. This could be due to the new experience system or the removal of discounts (or just me complaining about new things, which is beginning to seem more and more likely as this post grows longer), but it really does feel like casters took a solid whack with the nerf bat at some point.

Yeah, I noticed that too. Transmuter that have a an average Spellcasting Apt (0 oder -1) have a a really hard start. Until XL10 you feel like a weak monk, who sometimes can throw Potions (which will fail in the most Important situations ;>). It takes about 9 or 10 Levels to get Spider Form to very good.
Seems like races with positive Spellcasting Apt got even better in comparison to other races than before .

On the plus side: It forced me to use Sticks to Snake again, which I skipped in 0.8 every time, because it got obsolete too fast. This time I used it for quite a few situations and levels.

[absolutego]

sticks to snakes works well against early jellies and mummies, but yes, you typically want forms before that.
i played (and promptly killed) a transmuter yesterday and i may have found early casting harder than usual, but it's hard to tell and it wasn't too bad. some classes may benefit from higher starting spellcasting, though. should play more.

[Bim]

I've played quite a few games as a Merfolk Skald and found it very, very hard going early on (with a ridiculous amount of deaths). I then went back and played a few games on 0.8 and found it a lot easier (with and without using bezrage). It's nerfed them especially hard with it being harder to get there combat spells reliable or negate the hunger cost with most of them being things you need to cast 'in the moment'/a number of times.

My feeling is its been nerfed too hard, especially earlier on and think it could really do with being toned down a bit, especially for spellcasters. I'm also picking up skills I don't want a lot faster as well, which isn't necessarily a problem as its equally annoying never being able to pick up some skills, but is a bit annoying.

Is the new skill 'you-pick-where-xp-goes' system going to be implemented in 0.9?

[galehar]

Is the new skill 'you-pick-where-xp-goes' system going to be implemented in 0.9?

I hope so, I'm working hard on finish it in time.

[absolutego]

i also played a skald the other day and it took ages to get anything castable at an acceptable spell success (wearing leather armour, no shields), so i concur with Bim on that regard.
it was a demonspawn, XP penalty and all, so i wasn't sure i should put much weight on it, but here it goes. he died at XL10 with no spells at excellent. having to train so many different schools doesn't help.

[Bim]

I'm really looking forward to the new skill system! Should be awesome!

I really, really found it hard to get going with a skald, the cloak doesn't seem to do much at all, and with not being able to get much up to a reliably castable level I found it hard to get by, even when I was lucky with equipment. I wouldn't say I'm a brilliant player, but I have played for a good few years and seem to be doing noticeable worse now than before after a lot of goes. Possibly a boost to spell success (either through boosting spellpower or making spell success easier) would even the playing field? I spent ages victory dancing spellcasting and didn't seem to get far with it at all.

[galehar]

i also played a skald the other day and it took ages to get anything castable at an acceptable spell success (wearing leather armour, no shields), so i concur with Bim on that regard.
it was a demonspawn, XP penalty and all, so i wasn't sure i should put much weight on it, but here it goes. he died at XL10 with no spells at excellent. having to train so many different schools doesn't help.

I think skalds should focus on charms and melee in the early game and not try to train all the schools at once. The change makes aptitudes much more relevant, so not recommended combos are going to be harder. Species with neutral or positive aptitudes don't suffer as much.

[absolutego]

i know, i'm not complaining, just offering it as feedback. some combinations may benefit from slight adjustment of starting skills, and skalds are new in their current incarnation, so it never hurts to know the players' say. but i haven't played them enough to offer significant insight.

again, for the record, i trained elemental schools to 1 (and later 2, because it felt cheaper than considerably higher levels of charms, where i might be wrong) and otherwise focused on melee, and this particular combination still wasn't easy to get off the ground (unless you get good early racial mutations, i guess). i was lucky enough to find an early buckler and i didn't dare use it because spell success went down the proverbial drain.

[Bim]

The change makes aptitudes much more relevant, so not recommended combos are going to be harder. Species with neutral or positive aptitudes don't suffer as much.

I agree that this is good, but my only complaint is that the combo's me and absolutego were playing are recommended as Skalds and also have pretty positive (ok, demonspawn don't (though they're no better off than any other combo) but merfolk are the 'classic') attributes and I found them a lot harder.

If the aim of this is to make the game harder, then that's fine (although I don't think it really needs to be), but it certainly seems to harm hybrids a lot. It means focussing on anything other than one or two skills for the early/earlymid is hard going, and I think that harms playability, giving you less options and less flexibility. Thinking of newer players, these changes make it even harder to see the more fun mid-game, especially with hybrid chars. which are probably the most fun.

As mentioned, making spell success a bit easier (or having a slower slope to start with) would be good, even if it gets harder to raise it later. I think Giving the player a few workable spells to survive in the beginning is really necessary, especially as the most successful MeSk I got I did purely as melee, VERY rarely casting ice/fire brand, which took alot of fun out of it.

[Bim]

Would have edited if I could, but I'd just like to say that I've tried a lot of other combo's with latest trunk and really notice it's got a lot harder. As said, If making it harder is the aim, than that's fine, but I really do think a lot of things (especially skalds and other hybrids) have been nerfed pretty damn hard.

[galehar]

Well, it's certainly a nerf to casters and hybrids. Magic is powerful, but it's supposed to be balanced out by the fact that you have more skills to train. If said skills are cheaper, then it makes casters overpowered, which is what this change is trying to fix.

[Bim]

Ahh, well you've definitely achieved that, although I would suggest that spellcasting is maybe made a bit easier, or atleast the hunger/success rate is slightly boosted. As I've said, I've found playing hybrids far too difficult in the beginning compared to being a 'pure' class. If not then perhaps a boost in some other way would make hybrids a more acceptable choice, as at the moment they're pretty dreadful.

[dpeg]

Far too early for any boosts. Magic is considered much stronger than melee; Crawl is known to favour generalists over specialists, hence the change. We will play 0.9 with it and see how it turns out. It may very well be that some currently-recommended combinations should be unrecommended; we're grateful for specific hints on which combinations to toggle (both ways).

[galehar]

You're right, I've been too mean. Let's increase base HP for everyone to compensate! No wait...

[galehar]

As I've said, I've found playing hybrids far too difficult in the beginning compared to being a 'pure' class.

I think it's how it should be. You're free to hybridize later, but if you want to do it right from the start, then you're spreading thin, it should be harder. The old crusader went around this with a combination of overpowered spells (berserk rage, haste) and skewed skill system. Why should the skill system have a hidden special discount to make hybrids' life easier? The overpowered crusader has been replaced by a challenging, but IMO more interesting skald. And the overall balance is better.

If it's optimal to start with a pure class and hybridize mid-game I think it's fine. Starting with a skald or playing a DEFE without ever touching a weapon being harder is ok too.

[Bim]

Well it's fine if that the route you want to go down, as I said, I was just pointing out that it had got significantly harder compared to non-hybrid starts. However, I think Skalds have been particularly nerfed over all the other hybrid classes, with a much weaker spell book and a harder start. Granted, if that's how its wanted, then that's fine. But I still think it's a bit out of step with most other classes.