Revamping combat

[Magnate]

Thank you - that's most enlightening. Really good to know that you can do it the incremental way and still end up with a good game. Definitely time to give Un another outing and see how it plays. Last time was a long time ago, and I couldn't get past my OCD which required me to open and inspect every empty cupboard - am I imagining this or did you add a "check all cupboards in the room" command at some point?

[andrewdoull]

With hindsight, how much would you say that you set out right at the start of developing Un, and how much occurred incrementally as you went along (change one thing, realise that something else would be better slightly different, etc.)? Is there a particular blog post (or three) that sets out your vision for melee combat? And if so, how close to that vision did you end up?

Pretty much every change to Unangband has been an incremental one with no design intentions before hand.

My melee combat fixes pretty much amount to:

a) Limit weapon damage bonus to (sides * dice) + 5 (not strictly correct, but the object power evaluation uses this metric).
b) Reduce to-hit chances of classes starting out so that early monster AC has a meaningful impact.
c) Give some monsters resistance or immunity to edged weapons, some resistance or immunity to blunt and make some evasive or huge (hard to hit in interesting ways).

And by extension:

d) Make throwing / archery / magic / wands interesting and viable alternatives at even low level.
e) Make monsters more interesting.
f) Make attributes (str, int etc.) have more in-game effects.

There's other stuff (charging, applying potions to weapons etc.) but the above is the main mechanical changes.

[Derakon]

Dual-wielding as implemented in the vast majority of RPGs is a silly and stupid concept. Having a sword in each hand doesn't make you any more effective offensively than having a sword in one hand; it just gives you a different form of defense than having a shield in the off-hand.

(Well okay, the off-hand weapon could be used for stabbing when your opponent's weapon was brought out of line by your main weapon, but if anything that'd be more like getting a bonus critical hit. Certainly not free blows.)

[Djabanete]

While I'm ambivalent about the whole 2h/dual wield thing (I see it as nice flavour but far from important)

Two-handed wielding adds some depth to early/mid game weapon selection and rewards heavier weapons, so it's definitely not just flavor.

Dual wielding IMO doesn't belong in Angband, not in V anyway, because it rewards bizarre behaviors like wizards dual-wielding for the stat boosts and resists.

[Magnate]

It does require quite a big redesign.

With hindsight, how much would you say that you set out right at the start of developing Un, and how much occurred incrementally as you went along (change one thing, realise that something else would be better slightly different, etc.)? Is there a particular blog post (or three) that sets out your vision for melee combat? And if so, how close to that vision did you end up?

[andrewdoull]

Easyest way will be to implement stats like in Unangband. So, you get some stats points every level, which can be invested in stats with diminishing returns and stat potions work like restore + temporary boost to a given stat.

Another good idea IMHO is to get rid of initial stats points, just make a given class/race have a given stats. It is allmost like this anyway. Everybody uses some proven best stat distribution.

That's not quite correct. In Unangband, every few levels, you get multiple stat points to spend on abilities - however, you can only improve each ability once when you do so. Then, every 10 levels, you get 5 stat points, which mean you can boost almost all your stats.

This means you can't take advantage of the exponential curves quite so much as a single stat gain per level - and forces you to diversify a little.

Additionally every stat (including charisma) benefits (almost) every class. And you actually spend a fair amount of time missing monsters in melee - so to hit bonuses are much more important than vanilla. And so on.

It does require quite a big redesign.

[Antoine]

I know you personally are against min-maxing, and that's totally fine, but I can't see any feasible way to allow customizing characters from the start without also allowing min-maxing. The best we can do is make certain that the "max" isn't too absurd.

The best cure to min-maxing is to make mins hurt.

A character with 6 Wis should be a sucker against enemy spells, a character with 4 Cha should be just about unable to use the shops and so forth.

A.

[buzzkill]

~no rage eeee.

[LostTemplar]

Random is bad, definitely. It simply does not fit, only provokes restarting until (allmost)best possible stats appear. You cannon make anyone to play bad char so why not have only one possibility, the best one.

[Derakon]

So I can play a character for 10 hours before discovering that his ultimate potential is below-average? No thanks.

Similarly, you can't force the player to accept random stats at the beginning of the game; they'll just keep creating new characters until they get ones they like.

I know you personally are against min-maxing, and that's totally fine, but I can't see any feasible way to allow customizing characters from the start without also allowing min-maxing. The best we can do is make certain that the "max" isn't too absurd.

[buzzkill]

Use the standard roller for starting stats. Then use a variation of it again, again modified by race and class, to cap stats at a random amount (non 18/220) unknown to the player until they reach it.

[Derakon]

I think Magnate is referring to how WOW race / class combos start with fixed stats (as opposed to distributing stat points) rather than choose which stats to boost at levelup.

Angband may not be about grinding for levels, but it is about grinding for stats (stat potions, etc) or at least it has been.

I second (third?) the fixed starting stats. It allows racial / class differentiation at the beginning of the game for free without any additional changes due to the different stats each starting combo will have.

And I think it is more newbie-friendly than expecting them to not only know which stats to boost, but what number to boost them to (e.g. Int 18/150 for mages, blows thresholds, etc).

Since the stat rolls have an average of 13, just replace the roll in chargen with +13 and you're done.

So, what, a Hobbit Warrior will start with a STR of only 16? 13 + 5 from class - 2 from race. That kind of thing is going to make the early game very hard unless we also completely rescale all of the stats (and/or the difficulty of all monsters up to stat gain) too. You could make the class bonuses and penalties more extreme (e.g. warriors get +8 STR instead of +5), but then we'd be basically just force-allocating the point-buy stuff for the player. Right now we suggest how the player would want to allocate their points instead. And usually it's a good suggestion! Newbie-wise the current system is plenty friendly.

This feels to me like a solution in search of a problem. Can someone better-motivate to me why this needs to be changed?

[EpicMan]

I'm actually not a fan of this approach. I also disagree that it's newbie friendly, I think it's the opposite. It requires a player to know what stat would be most useful at clevel 50 when he gets to clevel 2.

Also, I thought Angband was designed to be around getting better equipment rather than grinding for levels.

I think Magnate is referring to how WOW race / class combos start with fixed stats (as opposed to distributing stat points) rather than choose which stats to boost at levelup.

Angband may not be about grinding for levels, but it is about grinding for stats (stat potions, etc) or at least it has been.

I second (third?) the fixed starting stats. It allows racial / class differentiation at the beginning of the game for free without any additional changes due to the different stats each starting combo will have.

And I think it is more newbie-friendly than expecting them to not only know which stats to boost, but what number to boost them to (e.g. Int 18/150 for mages, blows thresholds, etc).

Since the stat rolls have an average of 13, just replace the roll in chargen with +13 and you're done.

[fizzix]

What do other people think about this idea? It has a lot of newbie-friendly appeal. It's how World of Warcraft does it, in fact, and IIUC there's no shortage of character customisation in that game ...

I'm actually not a fan of this approach. I also disagree that it's newbie friendly, I think it's the opposite. It requires a player to know what stat would be most useful at clevel 50 when he gets to clevel 2.

Also, I thought Angband was designed to be around getting better equipment rather than grinding for levels.

[Magnate]

Another good idea IMHO is to get rid of initial stats points, just make a given class/race have a given stats. It is allmost like this anyway. Everybody uses some proven best stat distribution.

What do other people think about this idea? It has a lot of newbie-friendly appeal. It's how World of Warcraft does it, in fact, and IIUC there's no shortage of character customisation in that game ...