Scores

[jrodman]

As a new player, I feel that 'the deepest dungeon level reached' is already a perfectly fine score.

Counterpoint: I mostly feel that I"m trying to maximize both highest character level and deepest dungeon progress in most Angband games I play, and it leads me to play extremely cautiously. I'm at peace with the strange way I play Angband, but I think this way of thinking encourages playing the game in a way that isn't the most fun.

[Atanvarno]

Providing many statistics so everyone can choose their own scoring method is good, but it would make sense to have a more reasonable default scoring method than the current one.

I like the idea of providing "score multipliers" for different difficulty options, races, classes, conducts, etc, but getting the numbers right would be hard. This idea could be applied to all the methods below.

Personally, I would like the following to be true:
(A) Score never decreases.
(B) Score does not change after winning (so there is no need to keep playing just to get a better score, but there is no penalty either).
(C) If you are ready to kill Morgoth, then delaying it (e.g., to kill more uniques first) does not improve the final score.

B and C mean that comparing winners should be done by a "do less to get a better score" type of method. Some variation of turncount works reasonably well.

Comparing non-winners could be done by a "do more to get a better score" type of method. Deepest level reached is already a pretty good metric. You could add bonuses for killing uniques, or even for the first kill of every monster type, or the first clearing of every special room type (for some definition of "clearing"), and so on. The weight of high-level achievements should be large enough compared to low-level achievements, so that there wouldn't be an incentive to grind for all the low-level ones.

An entirely different approach to the "what is a good scoring method" question would be to consider it from an in-world perspective. The goal is to kill Morgoth, presumably because he is making his minions to do bad things. He should be killed as soon as possible (measured by game turns). Probably the same is true for killing the other uniques. This could lead to the following:
- For the purposes of calculating the score, pretend that every remaining unique dies when Morgoth dies.
- Every unique has a weight, describing how efficient he is at causing bad things.
- Primary score is the sum of weights of dead uniques (how much the player has decreased the rate at which bad things happen in the future).
- If primary scores are equal (which is the case for all winners), then we use a secondary score (lower score is better): Multiply the weight of every dead unique by the time it took to kill him, and take the sum of these numbers (how much bad things they caused before dying).

[brbrbr]

I measure my success as: "quickest time to victory in less number of tries".
So number of turns is good metric (should be compared within same character class combo).
Number of tries to victory is the best one though, how do you measure it?

[Werbaer]

I agree the turn count should affect the score at winning - you should get more points for winning with a low turn turn and less with a high one, factoring in your class.

I disagree.
Score by turncount favours melee playstyle, and punishes use of spells, since you need to rest much more.
My mages never (voluntary) melee or shoot, only use spells (and in some games wands and rods, sometimes not). Playing a mage as a warrior with utility spells gets a better score? No thanks.

I used to look at my scores when I first started playing, but I barely notice them now. I would suggest that the main use of scores is to compare one's own characters' progress, and mainly for newish players. So maybe something simple like character level + max dungeon level + number of different monsters killed.

Seconded.
I especially like the idea of getting points for different monsters. No 'Great Wyrms of Balance aren't worth fighting, much less the Tarrasque' - killing one of these should increase your score.

My idea:
+ (monster level) points for each monster killed for the first time (estimation: about 15.000 max)
+ 3x (monster level) points for each unique killed (estimation: about 15.000 max)
+ (max dungeon level) x (max dungeon level), capped at level 100 (10.000 max)
+ a big value (say another 10.000) when retiring after killing Morgoth.

[Derakon]

I disagree. My idea:
+ (monster level) points for each monster killed for the first time (estimation: about 15.000 max)
+ 3x (monster level) points for each unique killed (estimation: about 15.000 max)
+ (max dungeon level) x (max dungeon level), capped at level 100 (10.000 max)
+ a big value (say another 10.000) when retiring after killing Morgoth.

Be aware that this metric favors "completionist" players who attempt to kill every unique, even if they have to replay levels over and over again to do so. That's a big part of why I tried to factor in character level into the score bonus for killing monsters, because that makes it far harder to grind for score -- and if it's difficult, is it really grinding?

[wobbly]

Here's another view. I'd score on birth options. Both as a multiplier & a metric. Off the top of my head:

Forced Decent - Turn count
Iron Man - ?
Beginner? - Clvl, depth reached
?
?
?

Edit: probably even no need for the multiplier. Either build it in to the metric, or let different play styles compete on equal footing. You can see the options in the dump anyway, if you're looking for something particular.

[Rydel]

I disagree.
Score by turncount favours melee playstyle, and punishes use of spells, since you need to rest much more.
My mages never (voluntary) melee or shoot, only use spells (and in some games wands and rods, sometimes not). Playing a mage as a warrior with utility spells gets a better score? No thanks.

I agree the turn count should affect the score at winning - you should get more points for winning with a low turn turn and less with a high one, factoring in your class.

Emphasis added. I'd put in a bonus for winning, which scales based on how many turns you took, but the point it scales off would be what's considered average for your class (or possibly race/class combo, but I don't think race is causing major differences there). So, while a Mage takes longer, they'd also be judged off a higher target time.

[bio_hazard]

I don't think this should be included in an integrative total score, but I'd be interested in metrics that convey how much exploration occurred.

Total Tiles seen

or

Level dependent
Sum[ DL/100 * Tiles on that DL]

I'm not sure we'll ever get a full replay capability but something like this could help fill out the char dump

e.g.
50' 200 Game Turns, 100 tiles
100' 115 Game Turns, 47 tiles