Towards better tactical combat

[Derakon]

How about this? Monster is not at full health or player was in LOS within the last 10 turns or monster hunts by scent? Presumably scent will decay after some time.

[Nick]

Currently monsters detect by sight (a monster in LoS of the player sees the player), by sound (represented by a "heatmap"), and by smell (laid around the player and decaying over time). Also, monsters too far away to detect the player in one of these ways become passive - they just sit and do nothing, even if awake, until the player is detected again.

My current list of ways to improve this situation is:

• Have monsters sometimes not notice the player even when in LoS, possibly by monsters having some kind of perception stat (which might also affect noticing of monster traps set by the player)
• Rather than becoming passive, allow monsters to wander (in some manner) when not detecting the player
• Add other methods of detection, like by life force or magic use

[fizzix]

My current list of ways to improve this situation is:

• Have monsters sometimes not notice the player even when in LoS, possibly by monsters having some kind of perception stat (which might also affect noticing of monster traps set by the player)
• Rather than becoming passive, allow monsters to wander (in some manner) when not detecting the player
• Add other methods of detection, like by life force or magic use

Having unaware monsters opens up a lot of options, including rogue stabbing attacks and player invisibility.

The way I was thinking of doing it is having an "awareness" bar, sort of like the current sleep counter. It can be both positive and negative. Negative means the monster is unaware of the player, positive means it is aware of the player.

Every player turn, monster awareness gets updated. Monster perception is rolled against player stealth, with a modifier for stealthy action (no movement/rest), neutral action (player movement without attack) and noisy action (anything else). Of course everything is modified depending on whether the monster can see the player. Player actions which directly affect the monster, also affect monster awareness. These actions should immediately make a monster aware, provided they can see the player. Unaware monsters wander aimlessly, or aimfully but that's more work. Aware monsters track as current. Aware monsters can lose awareness too and go back to unaware if the player is out of sight for long enough and being stealthy.

It may be that we need a third state, "unaware but tracking" in which the monster doesn't know where the player is, but is still actively looking, based off of scent maps.

Intelligent wandering monsters are super tricky, especially when you're dealing with groups.

[Djabanete]

I have found this thread very interesting even though it has been a long time since I've played *band. I agree that much of the fun lies somewhere between strategy and tactics, but closer to tactics.

try to move from having lots of situations that look like, "If I don't do X,Y or Z now I may die next turn" into "If I don't do X, Y or Z now I may die in 3-5 turns" At first glance this looks similar, both require immediate action, but there's actually a huge difference here. It allows for situations where planning ahead is critical.

Yes! I think the reason this is more fun is that if you have exactly one turn to change the situation, only a small number of panic buttons are acceptable options, but if you have 3--5 turns to change the situation, there is more depth to the chain of options you can explore. Derakon correctly pointed out that Angband does already have a lot of situations like this, but I think they are more weighted toward the early game, whereas the immediately lethal situations are weighted toward the late game.

He wants to make it more like Sil, with no one-shotting, but way more dangerous than V nonetheless. I am not convinced it can be done without making V a near clone of Sil.

Right, I agree that the Fizzix's suggestion does not necessarily point toward making the game easier and in theory would make it harder. To oversimplify things, it requires more skill to notice and solve a 6-step life-or-death problem than to solve a 1-step life-or-death problem. I also agree that it's not obvious how to alter the game in this direction and sympathize with Nick's preference not to systematically use this as a guiding principle.

But it might be a good notion to keep in mind, because in my view the second half of the game (very roughly speaking) is less dangerous, less varied, and longer in real time than the first half of the game. In the first half of the game, new tactics are constantly developing as you find items and level up; lethal situations unfold over longer time horizons, giving you time to play with the different tactics; and there's not much need to linger anywhere. Later on, you get into a groove tactically; lethal situations crop up very quickly but require pretty routine responses; and the last 40 or 50 levels are kind of homogeneous in terms of most monsters being fast, summoning a bunch, and doing a bunch of damage.

I don't mean this as a scathing critique. In my view, even the second half of Angband is pretty darn fun. But I like the first half more than the second half, and Fizzix identified a part of the reason that I hadn't really thought of before.

[Pete Mack]

So keep create stairs? Even Sil has this, with Horn of Blasting or whatever it's called.