What Eddie Plays

[Derakon]

Random silly idea: reduce all monster damage values by a factor of 5, but make all player spells that aren't immediate offensive actions take 5 turns to go off. Want to phase door? Cast it 5 turns in advance. One turn from death? You should've drunk a healing potion five turns ago.

[TJS]

Everything ties together. If we're really going to fix this, Angband is going to be a lot more broken, balance-wise, before it gets better.

Yes totally agree with this. And if it isn't time to fix it now when everyone is already complaining the balance is broken then when would be a good time to do it?

[fizzix]

I sort of would also like to see a full overhaul. Linearizing the tables has been on my list for a long time. (I actually did it once for HP a long time ago, but that just meant the midgame was too easy.)

The way I see the progression is this, but this is really Nick's domain:

1) change combat to be more transparent. We could use O style, or the finesse/power from v4, or Sil's system. I have no idea which is objectively best, but I'd like it to be reasonably simple (simpler than DCSS which is a reasonable benchmark.) Sil level should be a good target.

2) Rebalance monsters/player stats around the new combat. This involves figuring out how long we want the game to be. The length unit is the number of times monsters go from challenging to windowpane. My crude guess is we should be aiming for about 5 transitions for a 50 level game, and ~8 for an 100 level game.

3) Rebalance gear to fit in with player stats and monster abilities.

4) Fix game mechanics that are overpowered or broken. A split between movement speed and action speed is high on the list here. Also on the list is archery, teleportation, destruction, and many others. In tandem, identify unfair advantages, both for the player and the monster. Descending stairs and immediately getting breathed on by 10 hounds and dying is an unfair advantage for the monsters. Being able to immediately teleport level out of danger is probably an unfair advantage for the player. Once these are identified, eliminate them. A delay on teleportation is a useful tactic (DCSS uses it).

5) Balance drop rates with statistics and player feedback. Figure out a way to include strategic gameplay (inventory management perhaps) in addition to the more tactical nature.

So basically, what I'm saying is that we should start by picking a combat system and then start mucking with the player stats.

This would look like a very different game, probably more different than any change to date. It will likely make a lot of people unhappy.

[Estie]

I am very sceptical regarding the straightening out of all formulas. There are certain changes of value in the game, like the fact that you want daggers early and heavy maces late, or the fact that you dont care for constitution early on, but want it lategame, that I find important to make the game interesting.

They pester the value function for shop prices and artifact evaluation, but imagine the game where you always want the same - the best weapon for lvl 50 is also the best for lvl 1 etc.

That point where you find yourself wielding a dagger which does good damage and then you find an axe of Westernesse, with half the damge of the dagger but FA and SI. I would miss that.

I would much rather have more such weird bumps created than the existing ones removed.

[Derakon]

Of course there should still be uses for both daggers and axes, and the player should feel that tension between damage and useful abilities. But there has to be a better way to achieve that tension than the current system.

[luneya]

Actually, I think the Vanilla melee system only needs to be tweaked, not replaced. It makes good sense that a weak, low-level character can swing a dagger or rapier more effectively than a lance or maul. The problem is that it takes too much stat gain for the heavy stuff to become effective. Once a warrior gets str and dex around 18/100, they should be able to wield either heavy or light weapons without much difference (maybe with something like +1-2 blows per turn on the light weapon, but this being offset by the bigger damage die on the heavy one). And with str and dex of 18/200, the warrior should have max blows on anything lighter than Grond.

As for the other classes, I think priest, paladin, and ranger should follow the same pattern of starting out with light weapons and being able to transition effectively to heavier stuff by the midgame, though not with as many blows as warriors would get. For mage and rogue, it makes more thematic sense for them to have to stick to light weapons through most of the game if they want to melee effectively. Perhaps allow rogue to have as many (or more) max blows as warrior, but never to be able to achieve this peak with any weapon heavier than a rapier. After all, that sort of stabbing with light weapons is what rogues traditionally do in RPGs.

[TJS]

I sort of would also like to see a full overhaul. Linearizing the tables has been on my list for a long time. (I actually did it once for HP a long time ago, but that just meant the midgame was too easy.)

The way I see the progression is this, but this is really Nick's domain:

1) change combat to be more transparent. We could use O style, or the finesse/power from v4, or Sil's system. I have no idea which is objectively best, but I'd like it to be reasonably simple (simpler than DCSS which is a reasonable benchmark.) Sil level should be a good target.

2) Rebalance monsters/player stats around the new combat. This involves figuring out how long we want the game to be. The length unit is the number of times monsters go from challenging to windowpane. My crude guess is we should be aiming for about 5 transitions for a 50 level game, and ~8 for an 100 level game.

3) Rebalance gear to fit in with player stats and monster abilities.

4) Fix game mechanics that are overpowered or broken. A split between movement speed and action speed is high on the list here. Also on the list is archery, teleportation, destruction, and many others. In tandem, identify unfair advantages, both for the player and the monster. Descending stairs and immediately getting breathed on by 10 hounds and dying is an unfair advantage for the monsters. Being able to immediately teleport level out of danger is probably an unfair advantage for the player. Once these are identified, eliminate them. A delay on teleportation is a useful tactic (DCSS uses it).

5) Balance drop rates with statistics and player feedback. Figure out a way to include strategic gameplay (inventory management perhaps) in addition to the more tactical nature.

So basically, what I'm saying is that we should start by picking a combat system and then start mucking with the player stats.

This would look like a very different game, probably more different than any change to date. It will likely make a lot of people unhappy.

Sil gets so much stuff right, combat, stealth, monster AI, perception, evasion/armour it is worth borrowing heavily from that.

I'd also get rid of multiple blows entirely personally. They make weapons completely unintuitive and impossible to balance.

One of the main problems with the game is the whole detection mini-game. You've got instakill monsters around so you need to be able to see where they are before you get near them. So you detect all the time and avoid them or if you stumble on them you teleport them or yourself away. There's not much fun game decision making involved. Then when you get reliable detection and escapes the game becomes a trivial grind to get the gear to fight the end bosses.

It is possible to have instakill monsters without detection, but you would need to design game mechanics around it. For example the player could have a perception stat that gives clues to where a monster is on the map or smaller monsters could flee away from where these large breathers are coming from.

I can imagine the paranoia when you know one is close by but you're not quite sure where would be great fun. Do I drink this last potion of perception to figure out where he's coming from or hide in this corner and hope he doesn't see me? Fun game decisions rather than just the usual detect every couple of steps and then avoid anything deadly.

Trouble is the game is old and the people playing it aren't always the type of people who like change. You'd have to break a lot of stuff before putting it back together again.

[Nick]

The way I see the progression is this, but this is really Nick's domain:

You gave me such a nice agenda though! Still, I'm going to go through and critique it.

1) change combat to be more transparent. We could use O style, or the finesse/power from v4, or Sil's system. I have no idea which is objectively best, but I'd like it to be reasonably simple (simpler than DCSS which is a reasonable benchmark.) Sil level should be a good target.

I guess I would like to simplify to the point where if someone said "How does Angband combat work?" or even "What does +2 to hit mean in practice?" I could finish my answer before they fell asleep.

O combat - I like the way damage dice are important, and the way criticals are extra dice, but to-skill and (especially) to-deadliness are less clear. I get the general idea of v4 but never really played it enough to get the feel, and Sil combat I think is a bit too board-gamey for Angband. So I'd probably favour some variant of O combat.

2) Rebalance monsters/player stats around the new combat. This involves figuring out how long we want the game to be. The length unit is the number of times monsters go from challenging to windowpane. My crude guess is we should be aiming for about 5 transitions for a 50 level game, and ~8 for an 100 level game.

I have plans to redo races and classes in 4.2, and combat for 4.3, which reverses your order. That may change, but I'm looking at leaving both of those essentially unchanged in 4.1 (I think there are will be plenty of changes in that already). For monsters, this may be a decent guide, but it sounds a bit ... inorganic to me. I'm inclined to be a bit more capricious

3) Rebalance gear to fit in with player stats and monster abilities.

Well, yes; I guess there will probably be many small changes affecting gear.

4) Fix game mechanics that are overpowered or broken. A split between movement speed and action speed is high on the list here. Also on the list is archery, teleportation, destruction, and many others. In tandem, identify unfair advantages, both for the player and the monster. Descending stairs and immediately getting breathed on by 10 hounds and dying is an unfair advantage for the monsters. Being able to immediately teleport level out of danger is probably an unfair advantage for the player. Once these are identified, eliminate them. A delay on teleportation is a useful tactic (DCSS uses it).

I don't know if I consider any of these mechanics overpowered or broken, and some of them might actually be fundamental to Angband. Not to say there should be no changes, but I'm concerned about hollowing out the game and leaving a sterile shell.

5) Balance drop rates with statistics and player feedback. Figure out a way to include strategic gameplay (inventory management perhaps) in addition to the more tactical nature.

Balancing drop rates - certainly. My first instinct is to run a mile from strategic gameplay. Angband is a tactical game - it's all about reacting to what the dungeon throws at you, and philosophically that's what it should be.

So, in summary, I think I probably want to do things in a different order to your suggestions, and slower, and not do some of them. But I'm sure we'll talk about this again

[Pete Mack]

The trouble with O is it's already been done. Not everyone likes it. Without skills and abilities it is impossibly tedious, and skills and abilities are not newbie friendly: You have to deal with the tedium before getting the abilities. Damage-based melee has the benefit that even a weak character can get respectable damage with the right equipment (Careth Asdraig and a ring of dex). Short cuts like this in O are much rarer.

[spara]

Another silly random idea. How about capping the total number of points in attributes making it impossible to max everything? Make it so that the player at best can max out something like three stats and the rest would be at 10 or something when evenly distributed.

Would probably make the end game characters noticably different and would give stat swap potions a whole new meaning.

[Derakon]

I like the ideas behind v4 combat. Naturally, I made them. It never really got a proper balance pass though, so we don't know that it'll work in practice. I suspect there are some big issues that'd need to be addressed. The most notable that I recall from v4 were that finesse characters (multiple weak blows instead of few strong blows) had trouble with armored monsters, and that substantially more weapons are "junk" because most classes are biased towards either finesse or power. I never did figure out an elegant way to handle critical hits, either.

I'm sure these are all solvable though. I'd say the biggest takeaway from v4 though is that the game works just fine if you determine hit rate as (75 + player's accuracy mod - enemy's evasion mod)%. Make accuracy mods and evasive enemies be uncommon, so that they're special. You can give class-specific modifiers (warriors are more accurate, mages less) if you want, but the basic formula worked fantastically.

[Nick]

The trouble with O is it's already been done. Not everyone likes it. Without skills and abilities it is impossibly tedious, and skills and abilities are not newbie friendly: You have to deal with the tedium before getting the abilities. Damage-based melee has the benefit that even a weak character can get respectable damage with the right equipment (Careth Asdraig and a ring of dex). Short cuts like this in O are much rarer.

I was purely talking about the melee combat mechanics - effects of damage dice, plusses, player and monster armour values - and principally the fact that in O combat the damage is mainly decided by the dice of the weapon rather than to-dam values on the weapon and other equipment. I'm not suggesting replacing V with O in its entirety. Or I wasn't, but now you mention it...

[Pete Mack]

Nick--
While O combat has a loyal following, it is less popular than damage based combat. A look at the ladder shows Pos running first, V second, and Sil a ways back in third for recent posts. Pos, of course, uses V-style damage, but has variety in other ways.

[AnonymousHero]

I was purely talking about the melee combat mechanics - effects of damage dice, plusses, player and monster armour values - and principally the fact that in O combat the damage is mainly decided by the dice of the weapon rather than to-dam values on the weapon and other equipment. I'm not suggesting replacing V with O in its entirety. Or I wasn't, but now you mention it...

FWIW, I personally found O combat to be incredibly bland... i.e. little variety and only quite "graduated" upgrades when switching weapons (unless you get incredibly lucky). Part of the kick you get when playing is getting a massive upgrade once in a while... rather than many small frequent upgrades. FTR, I've only won O (1.1.0) once, so I may have biases based on that. EDIT: I think the ideas by Derakon in v4 are interesting, but for my part that's just theorycrafting. These ideas do have a little precedence from AD&D where certain enemies are immune to e.g. slashing damage. Not that I recommend the latter blanket immunity... it's just more reason to get only specific weapons and ignore all the others even though they may be better generally.

I think the main problem here may actually be that are way to many weapon types. This leads to TMJ -- along with the whole Priest/Paladin blunt/non-blunt thing and the general light/heavy thing. Ultimately, whether @ is wielding a Scythe, Scythe of Slicing, MoD, SoC or whatever, is immaterial -- it's just damage dice that matters. (Extrapolate to shields.)

I'm not sure what it would actually do the game, but I'd really like to see an experiment where the number of weapon/armor/shield types were drastically reduced.

[Antoine]

I'd be a bit careful of following the v4 line, lest you end up in the same place.

This whole discussion has a bit of a v4ish feel about it.

No offense meant.

A.