Stats

Ok I sat on that one long enough.

A player I respect a great deal, and who was my original Armageddon mentor, maintained that not prioritizing was best.

I think you have a better chance at getting a decent roll, if you prioritize. And I think you have a better chance of getting a great roll if you don't.

Certainly, almost all of my AIs have been randomized.

Since it has been discussed recently on the GDB, with a bunch of conflicting and questionable information from people who haven't looked at the code or experimented extensively with different stat setups:

Strength is by far the biggest combat factor of all the stats provided you have high strength. Strength doesn't even do anything until 15 (the last point of 'good' for humans; and into the exceptional range for elves). 10 through 14 strength is almost identical, it adds a tiny bit of carry_weight per point and that's it. 15 gives +1 damage, but then you have to get to 17 to hit +2. After 17, it's another point of damage per point of strength. The bonus accelerates at an absurd rate, making high strength really important.

/*  strength apply  */
/*  to_dam, carry_w, bend_break  */
struct str_app_type str_app[101] = {
   {-9, 1, -1},                /* 0 */
   {-6, 5, 0},
   {-5, 10, 0},
   {-4, 20, 0},                /* 3 */
   {-4, 35, 0},
   {-3, 50, 0},
   {-3, 65, 0},                /* 6 */
   {-2, 80, 0},
   {-2, 95, 0},
   {-1, 105, 1},               /* 9 */
   {0, 110, 2},
   {0, 115, 3},
   {0, 117, 4},                /* 12 */
   {0, 120, 6},
   {0, 125, 9},
   {1, 140, 12},               /* 15 */
   {1, 155, 18},
   {2, 170, 24},
   {3, 195, 32},               /* 18 */
   {4, 230, 40},
   {5, 290, 48},
   {6, 380, 56},               /* 21 */
   {7, 430, 64},

That's why dwarves are so powerful--those two points of extra strength they have over humans can be worth as much as the difference between 10 and 17 strength. In a game where most weapons do damage in the neighborhood of 1d8, having +5/6/7 damage is a ridiculously huge advantage. It more than doubles your damage compared to a 14 strength character, before taking into account the location multipliers. After counting those multipliers into the average, it's more like triple damage. A dwarf with exceptional strength will do *three times as much damage* as a human with good strength. That's fucking absurd.

Strength's damage bonus also "penetrates" armor in the sense that armor is a flat damage reduction value, not a percentage. If you have no strength bonus and you roll 3 on a 1d6 weapon, but your opponent's armor reduces the damage by 2, you did 1 damage. If you had +3 from strength, you did 4 damage--four times as much. This is the reason why dealing meaningful damage against heavy armor without high strength is nearly impossible unless your skills are so much higher than the opponent's that you land every attack and whittle them down with light hits and lucky criticals. With high strength, you don't rely on luck at all and will simply smash the opponent in just a handful of attacks.

In PvE, strength isn't quite as crucial because NPCs just stay put until they're dead. It's okay to take twelve attacks instead of six to kill an animal, eventually you'll be skilled enough that it barely matters. But in PvP, killing someone without big hits can be nearly impossible unless you lock them in an apartment or they have to stay and fight for some other reason (e.g. the arena). In most normal cases, if you're hitting for <10 damage on average, they'll simply run away if it becomes clear that you're more skilled. Without high strength, you won't reel the target and you won't do damage quickly enough to take them out in, say, the duration of a bash.

Now, agility definitely helps with combat. Agility has a reaction substat that gives bonuses to practically every combat roll. Kick, disarm, bash, even base offense and defense. People have long thought that strength affects chance to hit, but it's actually agility, through its influence on offense. Now, that seems pretty good, but there are two reasons why it's not as good as it sounds:

1) Unless you hit the upper ranges of elf agility, the bonus isn't actually that big.

2) It hinders your skill growth. A lot.

For #1, let's look at the numbers:


/*  agility apply  */
/*  react, stealth_manip, carry number, missile bonus c_delay */
struct agl_app_type agl_app[101] = {
   {0, 0, 5, 0, 4},            /* 9 */
   {0, 0, 5, 0, 4},
   {0, 0, 6, 0, 3},
   {0, 0, 6, 0, 3},            /* 12 */
   {0, 5, 7, 0, 3},
   {0, 5, 7, 0, 3},
   {0, 5, 8, 5, 3},            /* 15 */
   {5, 10, 8, 5, 2},
   {5, 15, 8, 5, 2},
   {5, 20, 9, 10, 2},          /* 18 */
   {10, 25, 9, 10, 2},
   {10, 25, 9, 15, 2},
   {10, 25, 10, 15, 2},        /* 21 */
   {15, 25, 10, 20, 2},
   {15, 30, 10, 20, 1},
   {20, 35, 11, 25, 1},        /* 24 */
   {25, 40, 12, 25, 1},        /* 25 */
   {28, 40, 12, 28, 1},

React is the first column. You don't even get a bonus until 16 agility (human 'very good'), and then it's only +5. It reaches +10 at 19 agility, the upper end of exceptional for humans, and then stays the same until 22. Your human never get more than +10 to rolls affected by agility, and you need max human agility to get even that. If you thought EG human agility was a big roll, think again--it's a measly +5 to the combat checks it affects. The stealth bonus is bigger, but we're talking about combat here.

The codedump isn't quite clear on all the uses of the 'c_delay' substat, but the one thing it does show is that it affects the delay after drawing a weapon. I'm guessing it might also affect the delay from active combat skills like kick/bash, but the difference isn't very noticeable. Whatever the case, we can see that the granularity of this substat is too small to worry about--there's no difference between 11 and 15, or 16 and 22.

If you have 60 slashing weapons and 75 offense, adding +10 is a shrugworthy affair. It certainly doesn't even come close to strength's doubling or tripling your damage per hit. While agility also affects attack speed (and the codedump doesn't seem to reveal the exact formula), I've never observed a difference anywhere near as big as that of strength's influence on damage rolls. I'd estimate that five points of agility yield maybe 20% more swings. It's a thing, but it's fairly minor. You'll notice a difference between average and exceptional, but probably not between very good and exceptional. Attack speed is governed by a number of things, and agility is just one of them. It's nice to have some but it never matters even half as much as strength's damage bonus.

Elves, of course, can get a much higher react bonus. It goes all the way up to 25 before hitting diminishing returns at the end of the elven agility spectrum. That's a really big bonus to combat rolls, but still, when you take into account your skills and offense/defense, it's only huge in the beginning. When you have 20 parry and 10 defense, +25 is definitely a big boost. Less so when you have 60 parry and 70 defense. Meanwhile, the damage bonus from strength remains immensely valuable throughout a character's career. It's not as if hitting harder becomes less valuable the more often you hit, you know?

And that leads us to #2: skillgains. As we all know, the bottleneck for any fighter is weapon skills and their associated hidden sub-skills (e.g. slashing vs. humanoids). These only go up on a pure miss (dodge) and it takes a lot of misses, easily 10-15 per skill point. The vast majority of characters never max out their weapon/offense skills. Almost nobody ever does. Truly maxing out a weapon skill is something I expect is accomplished by maybe one character per RL year, if that. Most plateau around journeyman/advanced and don't jump through the million hoops it takes to climb further. The more diligent grinders end up somewhere between advanced and the low end of master. It's hard to overstate how improbable it is to actually max out, and until you've maxed out, your agility bonus is working against you by placing the bottleneck at a lower skill level.

So if you start out with +10 offense through agility, that's just how many fewer points you'll have in your weapon skill when your progress comes to a halt. And worse still for elves who can start to struggle with combat skillgains as early as the onset of journeyman. Defense is a bit easier to raise since you can always force yourself to get hit by fighting unarmed/sitting, but offense and weapon skills will be a huge issue for agile elves. People have made it work in places like the notoriously code-savvy Red Fangs where they could spar for months with other agile elves to equalize the agility problem, but for most characters, this is not really an option. Most will be stuck fighting NPCs the 'rinth, hunting animals, or trying to spar shitters who haven't been grinding for months to work up the kind of defense skill required to get a meaningful amount of misses when you have a baseline +25 to offense.

This means that if you have high agility, the combat advantage you get out of it will start to diminish and eventually disappear entirely when you hit the point where access to misses is the bottleneck to further gains. Suddenly you'll find that your +20 offense is worthless because your skills stopped going up at a correspondingly lower level. Your 30-day elf will simply have much lower skills than that 30-day dwarf. It can be a problem for non-weapon skills as well: failing bash and kick, skills crucial for branching disarm on enforcers and raiders respectively, is nearly impossible past journeyman for an elf unless you try something suicidal like bashing salt worms.

Now, agility is quite good if you take a high-agility elf and a high-strength dwarf, give them both the same exact skills, and have them go at it. I'm still not convinced that the elf would win but it would be a real fight. But the reality is that of these two characters, given the same amount of playtime and training habits, the dwarf would have like 20 more points in weaponskill and off/def. Suddenly the elf's agility counts for nothing while the dwarf's massive damage bonus (and ability to wear the heaviest armor and carry any number of backup weapons) is a huge advantage. The elf *may* have won at 5 days, but at 20+, you've hit the point where your agility is hindering your skillgains and thus eroding the stat's benefits.

You could theoretically find ways to continue gaining despite your agility bonus, but this tends to involve a degree of cheating that will make most players (and all staff) recoil in horror. I mean, we're talking about something like kidnapping whatever 'rinth elf NPC has the highest defense and agility, capturing it in a dark room where it can't leave (non-aggro NPCs will run off if you fight them in darkness for a few rounds), and blindfighting it for days and days. Stilt lizards are nerfed, and staff have been known to crack down on people fucking around in gortok dens. There's practically no legitimate way to do it unless you're in the jackpot situation of being clanned with another long-lived, high-agility elf that you can spar with several times a day over the course of months. How many are that lucky?

So, in conclusion, strength is just a vastly superior combat stat on all fronts. It helps you offensively through damage, defensively through armor, and doesn't hinder your skillgains. Please don't buy into the X-D "I know everything about combat" meme. I'm tired of seeing him disagree with every opinion anyone ever posts about the code because it makes him feel smart, and others backing him up purely from reputation. There's a reason the combat vets (including him) roll dwarves ten times more than they roll elves. Don't listen to the bullshit of people who just want to sound like they know better, without ever providing any kind of evidence. If you think +10 to combat rolls that easily end up with an aggregate modifier of over 100 just from the skills alone is better than doubling/tripling your damage per swing, you're wrong.

I already talked about all of that. The agility bonus would be fine if you could expect to max out your skills, but 99.9% of characters don't, no matter how long they live. You get stuck around advanced in weapon skills and a similar level in offense, probably a bit more in defense, and then you never get the last 20-40 points. In those cases, your agility bonus simply causes weapon skills, hidden weapon skills, offense and defense to stop progressing at a correspondingly lower point. If you had average agility, you might land at 75 slashing. If you had +20 offense from agility, you might instead get stuck on 60.

All due respect, I don't believe you have a whole lot of combat experience (if you are who I think you are). You can't just look at numbers and suppose agility is better because it sounds better. Take any particular agility score and the same strength score, and the strength will be vastly better. If you take 26 elven agility and compare it to 19 human strength, the difference is less, but we're talking about a gulf of almost 50% in raw stats there; and I'd still venture to claim that 19 strength would yield a better fighter than 26 agility. When you compare 19 agility to 19 strength, however, or even 26 elven agility to 21-22 dwarven strength, there's no contest. The strength increases your damage dealt by over a hundred percent. No flat bonus to offense/defense/parry/etc. comes even remotely close to this.

Remember that the strength not only adds to your damage but also lets you wear the best armor in the game. You can't wear this armor plus the other combat trappings and general necessities without very high strength. The difference in survivability between light armor and silt-horror plate is much, much greater than the difference that +10 or +20 defense and parry makes. High agility is nice if you can get it, but if you can't, you'll still be a very good fighter as long as your strength is high. You just don't get that much combat mileage out of the agility bonuses. It's great for stealth and things like climb and steal, but that's it.

And you can't overstate the the impact strength has on damage, you really can't. Most weapons have a base damage of about 1-8, or less for smaller weapons like daggers and shortswords. Adding 4 to an average of 3 or 4 is double damage, and that's possible even for humans. You'll kill things twice as fast as someone without that strength bonus. No amount of agility even approaches that kind of impact. +20 offense does not double your damage. It probably increases your accuracy by something like 5% once you're past the brief early stage where that amount of offense is a lot compared to your actual skills. Between that and the attack speed, it's still below the benefits of max strength for any race besides elves whose strength caps too low to matter.

Why do you think people are so scared of dwarves and muls? Why do you think dwarves are the go-to race for twinks and griefers? Why do you think 95% of human warriors prioritize strength? It's not because agility is secretly better and people just haven't realized it after twenty years. It's because high strength and middling agility is better than middling strength and high agility by orders of magnitude, even if you were to discount the fact that agility is actually detrimental to your skill growth. The only notable elven fighters have been a small handful of individual d-elves who played in the right tribe at the right time to spar for months with the likes of Sirra and X-D, and simply outskilled anyone they ended up fighting. They would probably still lose to Abuzer in a melee fight.

Just to highlight how silly strength gets, here are the damage bonuses including two handed:

/* Add up all the damage bonuses */

           dam += str_app[GET_STR(ch)].todam;

           // plus half str bonus if using etwo
           wp_pri = get_weapon(ch, &n);
           if( n == ETWO )
               dam += str_app[GET_STR(ch)].todam / 2;

           // two handed skill can add skill% worth of str bonus
           if( n == ETWO && has_skill(ch, SKILL_TWO_HANDED))
              dam += str_app[GET_STR(ch)].todam * get_skill_percentage(ch, SKILL_TWO_HANDED) / 100;

So let's say you have high exceptional strength on a human, from which you get +4 damage. You're two-handing a weapon, netting another +2. And you have 75 in the two handed skill for an additional +3 (75% of your strength bonus).

Nine extra damage per swing just from strength. Six or seven even if your two handed skill is low. People often say that two-handing is a way that weaker characters can compensate for their low damage, but most of the etwo bonuses pertain to your strength bonus and thus do nothing if you have none. There may be other two handed damage bonuses unrelated to strength, but strong characters benefit vastly more from the skill. 

Like I noted in a previous post, most weapons have damage in the neighborhood of 1d6 or 1d8. Adding a flat 9 to that roll makes a huge difference. And that's just for humans. A dwarf with a similar strength roll (upper exceptional) would get 13-14 added damage from the aforementioned bonuses. A bit more if they actually maxed out two handed--I went with 75 because it's a bit more reasonable. That's an increase of several hundred percent baseline damage. Who gives a shit about agility at that point?