Yet another Gnoll rework

[Rast]

I'll start off by saying that I'm not sure Crawl needs a species with boosted skills in early game but almost no skill increase past lair. But here's my take on it anyway.

* Gnolls have N/A skill aptitudes and don't gain skill xp when gaining xp.
* Instead, their ranks in all skills increase to fixed amounts at every level up.
* They can view the skill screen ('m') but don't have any options there.
* They don't benefit from manuals and can't take the Ru -1 apts sacrifice.
* They can't worship Ash, for flavor and balance reasons.
* But Oka is OK.

Their starting skills at XL 1 are as normal for their background, but as if they had +1 apts across the board.
At XL 2, all skills below 1.0 increase to 1.0. Skills already at or above 1.0 remain unchanged
XL3: skills below 2.0 increase to 2.0
XL4: 3.0
XL5: 4.0
XL6: 4.8. This is more than any background could start with, so from here out all Gnomes have identical skills regardless of background.
XL7 through XL 15: All skills increase by .8 at each levelup.
XL10: 8.0
XL15: 12.0
XL 16 onwards: All skills increase by .2 at every levelup
XL 20: 13.0
XL 25: 14.0
XL 27: 14.4


Stats and stat gain, HP, MP, XP apts, MR: Current trunk versions of these are fine. In particular the bad MR apt is neat.

[Leszczynek]

Even if it's functionally similar to what current Gnolls do, taking away the choice of skilling from a player and setting it to some arbitrary levels doesn't seem particularly exciting to me. How is it better than what's currently around?

[duvessa]

It's better than the current system because it 1. accomplishes the intended goal better, 2. involves much less micromanagement on the player's side.
That said, I'm not a fan of the maze of breakpoints. You could create a practically similar distribution with just one breakpoint, e.g. +1 each level until XL5 and +0.5 thereafter (0, 1, 2, 3, 4, 4.5, 5, [...] 15 at XL27).
It would also be cleaner to measure it by skill points rather than skill levels.

[watertreatmentRL]

I like the OP's concept of skilling being controlled purely by the species so the player never has to enter the skill menu except for their own reference.

Rather than washing out the effect of the background after a few levels, though, it would add more to the game to go the opposite direction and make the effect of the background more permanent for the weird skill species. To that end, skills would start at whatever levels and further skill development would follow a schedule like the one described in the OP with a straight skill level bonus, a la Heroism, maxing out at 14, say. At the maximum boost, gnolls with different backgrounds would have a broad set of skills, but maintain their specialization, be able to hit different weapon and shield breakpoints, cast different spells effectively and so on. The effect could be accentuated by reducing the max skill bonus and increasing effective aptitude used to set initial skills.

In my opinion, it would be best to make the bonus something extremely simple, like .5 skill levels per experience level and compensate the loss in early game skill with higher initial aptitudes for computing the background contribution.

[ion_frigate]

So the stated goal of gnolls, in the recent commit disabling them in stable, is to encourage hybridization. A simple way to do that would be to force them to always be training more than one skill. They would get decent non-decaying aptitudes in return (maybe something like +1 across the board, but forced to train four skills at once). If this is too cruel early on, the forced diversification could come on level-ups.

[Siegurt]

So the stated goal of gnolls, in the recent commit disabling them in stable, is to encourage hybridization. A simple way to do that would be to force them to always be training more than one skill. They would get decent non-decaying aptitudes in return (maybe something like +1 across the board, but forced to train four skills at once). If this is too cruel early on, the forced diversification could come on level-ups.

I often train four skills at once when not hybridizatizing, for berserkers fighting/weapon/dodge/armour and swap out evocation(or ranged weapon) once weapon is at min delay, for FE spellcasting/fire/conjurations/dodge will spellcasting swapping out with fighting when i don't need it.

Four skills at once does nothing to encourage hybridization, and is only a penalty really in the very early game (where getting an offense online quickly is important, and defense isn't yet)

It is slightly more optimal to train one at a time, but the overhead of micromanaging your skills doesn't make enough difference in your practical results to me.