Alternatives to energy randomization

[and into]

Based on my own experiences and some comments that came up in other threads, I was wondering how devs feel about energy randomization. I think I understand why it was implemented, but it feels like a rather strong-armed fix that feels awkward in the context of so many other, much more nuanced and balanced ways of introducing risks or penalties to (over)using certain tactics in Crawl.

If the behavior to be discouraged is stuff like running around in circles for a while to regain HP ("pillar dancing" and the like), then the solution of energy randomization seems both too broad and yet also inefficient. Too broad, because many normal repositioning tactics can be arbitrarily punished at random; inefficient, because you can still perform indefinitely long stalling movements against dudes. Energy randomization does not so much help "force a confrontation" by letting you escape or letting the enemy advance, but in many cases prolongs one's retreat even in cases where a player, through good play, kept a dangerous melee opponent at one or two tiles' distance—a bit of bad luck and the enemy is next to you, and now while backing up you have to worry about a random dice roll completely independent of how you played may put the enemy next to you.

The problem, I think, is one of "lengthy movement stalemates" between players and enemies at the same speed. I think Crawl devs can come up with better ways to end those stalemates, but here's a quick proposal of my own:

  Code:

One alternative to energy randomization
Enemies who have done nothing but move while following you for a long period of time will tend to disengage, giving up the chase. However, they will make a great deal of noise after doing so. "The yak bellows loudly in frustration, before turning back." "The orc gives up, shouting curses." And so on. These shouts would have higher "noise" ratings than usual shouts, basically simulating an off-screen temper tantrum after the enemy has given up the chase.

In these sort of scenarios, the player has managed not to die to an enemy and retreat successfully for a long time, but occasionally this results in a "movement stalemate"; as opposed to energy randomization, my proposed alternative would resolve such stalemates technically in player's favor, but at a certain cost: The enemy you've escaped from makes a lot of noise and draws attention of other enemies on the level. This can also be expanded to other measures that enemies might take against you, based on intelligence; for instance, an orc pack might regroup and form an ambush party at some strategically important point on the level (like near a stair case), perhaps redistributing their weapons so that the strongest member of the group gets the best weapon, and making preparations for you, etc.

Such "revenge" AI effects would not ever trigger during tactical repositioning, nor during *efficient* retreats when you use the layout to break LOS to lose your pursuer or move quickly back to a staircase and go to another level. The noisiness penalty would only be triggered in cases where the player runs in circles or does similar for a long period of time without definitively eluding the enemy. Note that this would also punish very lengthy "kiting" sessions of enemies that you could run from but choose not to, and could easily be adapted to punish "door dancing" enemies capable of opening doors. The point wouldn't be to punish running away (which is often a good idea in Crawl and is usually interesting), but it would punish *very poorly conducted* retreats as well as abusive "pillar dancing" tactics that constitute the sort of un-fun degenerate behavior that players sometimes indulge in because they don't want to use up one of the 12 charges on that wand of fire they have, or something.

The advantage here, over energy randomization, is that it allows better fine tuning so that you never have situations in which good play and reasonable tactics (including reasonably executed retreats) are randomly punished by having enemies move extra tiles. Since good positioning is *the* major component of what makes good play in Crawl dynamic, interesting, and fun, even a small chance of having that screwed up for reasons independent of how poorly/well someone was playing is very bad for DCSS design.

[crate]

so once a melee enemy gets next to my hunter I want to go find a stair and a pillar so I can dance around the pillar near the stair so the enemy starts retreating (so I can shoot it to death as it does so)? (and then I take the stair to get away from the enemies woken up by the noise, which has helpfully pulled them away from the other stairs on the level)

I still don't really understand why energy randomisation exists in the first place. It doesn't really bother me too much as a player but I don't understand the design reasoning behind it. Why does it need a replacement? If players being able to lead monsters around indefinitely is not desirable, then it seems to me the solution is to just make monsters faster than players (this still encourages backing up until the monster gets a "double move" and moves next to you during it, but I'm not sure that this is undesirable behaviour when it is going to happen within ~10 moves, instead of something that will merely happen within an unbounded "eventually").

[and into]

Sure, you could have players get "winded" and suffer a slight movement delay (for instance) when you have been moving for X turns with a monster in sight, where X is relatively large.

Or you can keep something like energy randomization, and just have it only work in one direction (i.e., it either always causes monster to catch up, or always lets you put distance between yourself and monster). In this case I'd say energy randomization should only have chance of triggering when you have been moving away from a monster in LOS for X turns, where X would not be so large as in the above scenario.

I think the purpose of energy randomization is to try to prevent long "movement stalemates" as these are usually pretty boring. Having to run from a dangerous guy, even through explored terrain, can be fun and interesting, but it becomes boring if you are on a loop and doing it either to regenerate HP/MP or to wait for energy randomization to kick in and let you definitely escape up some stairs or whatever. (In this sense energy randomization can actually make these sort of situations worse, not better.)

Those sort of stalemates are much more common in early game (when you may not have consumables to escape them) and also amongst newer players who are still learning when/how to use consumables effectively. I do think it would improve the game if those sort of situations were avoided, because repetitive behaviors aren't fun. Obviously you could also try to solve this in more of a case-by-case basis. One of the common enemies to do this sort of thing with is jellies and other enemies that you just want to avoid getting into melee with at all, even if you can beat them. (Assuming you lack a suitable wand or whatever to take them out with.)

But it may well be the case that I don't fully understand why energy randomization was added.

[crate]

I think the purpose of energy randomization is to try to prevent long "movement stalemates" as these are usually pretty boring.

well it demonstrably does not do this, so this is a very puzzling justification if that is so

[reaver]

Here would be my alternative:

Implement the "variable wait" idea that's been around for a while. This means when you hit ".", instead of always letting 10 auts passing, it waits 1 aut then checks if anything has happened (most critically: enemies moving) and stops if something has happened. This lets players use "." instead of weird ring/weapon swap tricks.

Remove energy randomization from every monster except orc priests, since this is apparently the one monster randomized energy works well on. This would be a transitionary phase to some other solution to the pillar dancing problem, such as adding a monster flag which modifies the AI to handle pillars.

I'm not sure how good removing energy randomization will be, but it seems trivial to revert and a good change to try, even if you assume it isn't likely to stay.

[duvessa]

My approach would just be to get rid of randomized energy and give orc priests speed 11/haste/slow/whatever.

[Leafsnail]

Make standard races 1 slower and give them Swiftness as an activated ability, or something. I don't know, short of really radical changes (like trying to avoid ever having players that are the same speed as monsters) I don't see how you could fix this.

[KoboldLord]

Energy randomization does do an awful lot of strange things, and it doesn't really address the design issue I believe it was intended for, but there is one aspect of energy randomization that I really like. Without energy randomization, you can count steps when the enemy is across the screen and run some basic arithmetic to work out exactly how many actions you can take before you have to worry about any actual risk. You can then take the maximum available actions and still be absolutely safe because that one last step the monster has to take might as well be a thousand miles. Seeing a D2 ogre or an early hydra charging your inexperienced character should be an alarming proposition, not a counting exercise. You should not be able to deal with a D2 ogre by shoot six times, walk upstairs, alternate from a second staircase until you get lucky on damage with absolutely zero risk of taking damage.

[Leafsnail]

Doesn't energy randomization just mean you have to count the squares then subtract 1 to be safe?

[KoboldLord]

It isn't quite that simple, but yes, the problem is just smaller and not eliminated completely. Most problem sets in the game take place at closer ranges than the edge of LOS, so making the problem smaller does help if you are willing to put up with the odd behaviors that come as a side effect.

[Leafsnail]

When does energy randomization lead to a monster making two extra moves?

[Patashu]

When does energy randomization lead to a monster making two extra moves?

It's rare but possible - think 0.1% level odds.

[twelwe]

Energy randomization does do an awful lot of strange things, and it doesn't really address the design issue I believe it was intended for, but there is one aspect of energy randomization that I really like. Without energy randomization, you can count steps when the enemy is across the screen and run some basic arithmetic to work out exactly how many actions you can take before you have to worry about any actual risk. You can then take the maximum available actions and still be absolutely safe because that one last step the monster has to take might as well be a thousand miles. Seeing a D2 ogre or an early hydra charging your inexperienced character should be an alarming proposition, not a counting exercise. You should not be able to deal with a D2 ogre by shoot six times, walk upstairs, alternate from a second staircase until you get lucky on damage with absolutely zero risk of taking damage.

Make just one staircase then

[Patashu]

Energy randomization does do an awful lot of strange things, and it doesn't really address the design issue I believe it was intended for, but there is one aspect of energy randomization that I really like. Without energy randomization, you can count steps when the enemy is across the screen and run some basic arithmetic to work out exactly how many actions you can take before you have to worry about any actual risk. You can then take the maximum available actions and still be absolutely safe because that one last step the monster has to take might as well be a thousand miles. Seeing a D2 ogre or an early hydra charging your inexperienced character should be an alarming proposition, not a counting exercise. You should not be able to deal with a D2 ogre by shoot six times, walk upstairs, alternate from a second staircase until you get lucky on damage with absolutely zero risk of taking damage.

Make just one staircase then

Or introduce Brogue Stairs (tm)

[Baldu3]

Isn't the benefit of energy randomization that it makes it possible to escape from adjacent enemies and get to some stairs without using consumables/spells/evokes, even if it takes a while and some hits need to be taken?

[Galefury]

Isn't the benefit of energy randomization that it makes it possible to escape from adjacent enemies and get to some stairs without using consumables/spells/evokes, even if it takes a while and some hits need to be taken?

No, that's what weapon swapping is for.

[Lasty]

Give all monsters the ability to use a 2-3 duration swiftness triggered by the move action, with the standard player swiftness penalty when it wears off. The chances of monsters using it on each move are 1 in (100 - number of sequential turns moving towards the player). The counter resets whenever the monster loses track of a player or otherwise stops trying to path towards the player.

This would result is a much more dramatic (but fixed-scale) increase in monster speed that -could- happen any time (to prevent math) but generally only happens during prolonged chases. It gives the monster the chance for an extra hit or so, and then the monster will be slowed and fall behind, giving the player the ability to escape.

[Quazifuji]

Give all monsters the ability to use a 2-3 duration swiftness triggered by the move action, with the standard player swiftness penalty when it wears off. The chances of monsters using it on each move are 1 in (100 - number of sequential turns moving towards the player). The counter resets whenever the monster loses track of a player or otherwise stops trying to path towards the player.

This would result is a much more dramatic (but fixed-scale) increase in monster speed that -could- happen any time (to prevent math) but generally only happens during prolonged chases. It gives the monster the chance for an extra hit or so, and then the monster will be slowed and fall behind, giving the player the ability to escape.

You might be able to create more interesting AI for this than just it being used randomly. But giving every monster a "sprint" ability does seem like it could be a more intuitive, more flavorful, and less luck-based way to end prolonged retreats or pillar dances. If the only benefits of energy randomization are that it makes retreating/pillar dancing speed 10 monsters less reliable and makes the number of ranged attacks you can use before a monster closes a gap less predictable, the this would solve both of these problems too. And if the problems with energy randomization are that it's incredibly unintuitive and opaque and sometimes punishes legitimate, non-abusable tactics with random chance, then this would eliminate both of those things.

I think the lack of opaqueness to this idea is actually my favorite part. I spent a long, long time assuming that any monster that could hit me while I was running away was faster than me, and any time I gained distance from a monster while running away it must be slower than me. The situations where a gap was created but immediately closed while I retreated were just plain confusing. On the other hand, seeing the message "the ogre starts sprinting towards you as fast as it can!" followed by the ogre catching up and getting a couple attacks in, follow by "the ogre has exerted its energy and slowed down" and me getting away would make perfect sense and I'd understand exactly what just happened.

Not only does this create a scenario where running away from an enemy will reliably let them get a few hits in, but guarantee escape if you survive (which seems like a much better alternative to "you'll randomly take hits or gain distance when running away"), but it also communicates that perfectly. I don't think we'd ever have long, rant-filled, confused threads of people wondering why running away seems so unreliable in Dungeon Crawl advice if it worked like this. It might take some tuning to get the right balance between making escape mechanics more consistent and interesting and avoiding cases where an ogre entering LOS two squares away from you from around a corner and sprinting isn't instant death if you don't have ?blink, but overall it seems like a much, much more elegant solution than energy randomization, both in terms of gameplay and flavour, to all the issues that people have discussed in this thread.

[damiac]

Do not give orc priests speed 11, haste, or slow, for the love of god. They're already murder machines in the early D, but at least you can run from them.

I don't understand the point of energy randomization, and it seems like all it really accomplishes is occasionally killing players through no tactical mistake, or occasionally letting players escape who did screw up. Most often all it accomplishes is make me run around the pillar a couple more times.

[TeshiAlair]

I'm actually quite a fan, particularly in early game, probably because I'm a badplayer. I like that if I'm running from an OOD ice beast, that rather than burn some consumables, I can lure it to stairs, then pillar dance until we have a space and go back down them. It is far less annoying than burning things on an early floor only to run into something nasty 5 minutes later, or pillar dancing till at full HP, trying to fight, failing, and repeating like in the Olden Days.

[twelwe]

Energy randomization does do an awful lot of strange things, and it doesn't really address the design issue I believe it was intended for, but there is one aspect of energy randomization that I really like. Without energy randomization, you can count steps when the enemy is across the screen and run some basic arithmetic to work out exactly how many actions you can take before you have to worry about any actual risk. You can then take the maximum available actions and still be absolutely safe because that one last step the monster has to take might as well be a thousand miles. Seeing a D2 ogre or an early hydra charging your inexperienced character should be an alarming proposition, not a counting exercise. You should not be able to deal with a D2 ogre by shoot six times, walk upstairs, alternate from a second staircase until you get lucky on damage with absolutely zero risk of taking damage.

Make just one staircase then

Or introduce Brogue Stairs (tm)

crawl lite did this and i wrote about it in that thread. brogue stair trivialize fighting to some extent, especially in open area levels with a staircase leading (or inside of) a corridor.

[Brannock]

What's different specifically about Brogue's stairs? I haven't gotten around to playing Brogue and I don't see anything enlightening turning up on my searches for "brogue stairs".

[Sar]

Monsters can and will use them.

[Patashu]

What's different specifically about Brogue's stairs? I haven't gotten around to playing Brogue and I don't see anything enlightening turning up on my searches for "brogue stairs".

Play Brogue sometime, it's a good roguelike. (Or Crawl Light, since Crawl Light has Brogue stairs too.)
But here's how Brogue staircases work:
1) If a monster is pursuing you when you take the stairs, the game notes how many moves away the monster is from the stairs.
2) On the level you moved to, after you take that many moves, the monster comes up the stairs and continues to pursue you.

[damiac]

So you can essentially find a good place to kill monsters, then lure all monsters there from anywhere else in the game? I can see the issue.

[Arrhythmia]

The thing is, in Brogue, pretty much every monster is capable of beating you in a fair fight and there's not enough food to constantly be backtracking.

[damiac]

Ah. But I see why it wouldn't work well for DCSS

[dpeg]

Alright, for the record: energy randomisation was implemented in order to make pillar dancing less reliable.

If you guys would be responsible for the game and suddenly see everyone and their granny do the pillar dance, you'd try to come up with something as well, I bet.

Sure, the mechanic is far from perfect but whatever happens, simply going back to reliable pillar dancing is not an option.

From reading the thread, I really like the idea of monsters getting increased movement speed after a (randomised while) in sight of you. I didn't understand why they should slow down after another while, though That would solve pillar dancing and some forms of kiting.

[Quazifuji]

I didn't understand why they should slow down after another while, though

Well, from a player standpoint, randomized energy can help or hurt. It doesn't make pillar dancing bad, it makes it a bigger gamble - sometimes it'll let you escape from an enemy you couldn't beat at full health, sometimes it'll get you killed by an enemy when you try to run away. You can see people in this thread who appreciate the first case as well as people who are frustrated by the second case. So having the monster slow down after its burst of speed would sort of combine both cases in a reliable pattern - the monster gains a few tiles, then loses a few, so you're guaranteed to take hits, but get a chance to escape if you survive. But this isn't necessarily the developers' perspective on the matter, since it sounds like the main goal of energy randomization was the cases where you get caught, not the cases where you escape.

Another issue with the random burst of speed is the case where, say, an ogre comes into vision from around a corner two squares away from you, and you try to retreat to a staircase. Ignoring the possibility of encountering other monsters on the way to the staircase: with no energy randomization, escape is guaranteed. With energy randomization, the odds are heavily in your favor, especially if the staircase isn't super far (but even if it's on the opposite end of the floor they're still pretty good). With random speed bursts with slowing afterwards, you're okay as long as you can take a couple hits. With random speed bursts with no slowing, running away is an extremely risky proposition and there's a good chance you should be popping consumables.

So the question here isn't just about pillar dancing, but about retreating in general. How effective should retreating from a speed 10 melee monster be from a given distance? Should the odds be in your favor for getting away cleanly, should the odds be in your favor as long as you can take a few hits, or should the odds be against you unless the monster's at least 4-5 squares away?

[duvessa]

Alright, for the record: energy randomisation was implemented in order to make pillar dancing less reliable.

If you guys would be responsible for the game and suddenly see everyone and their granny do the pillar dance, you'd try to come up with something as well, I bet.

The whole point of the posts in this thread is that randomized energy doesn't actually make pillar dancing less reliable (except against orc priests)

[Leafsnail]

If the fundamental problem is that players benefit from prolonging a movement stalemate the simplest solution would be to halt regeneration while moving/acting. Doing that would probably require a substantial re-balance of the early game though.

[skjarl]

What about giving all players the ability to sprint a short distance? The distance could be determined by your strength, agility, weight, or some combination thereof. If monsters can sprint, it will remove our ability to get away in a lot of cases. There must be a counter that doesn't rely on consumables that may or may not be available. Being able to sprint away would also give folks an alternative to pillar dancing.

[Lasty]

skjarl, your proposal would allow you to always escape from speed 10 monsters without taking damage, which would be a straight up buff to players. It would also obsoletize the Swiftness spell.

[johlstei]

To be more clear, that already exists and is called "swiftness".

[skjarl]

Yes, I'm proposing Swiftness be made an ability instead of a spell. Is there something holy and special about it that takes it off the table for any changes? Folks are talking about buffing all monsters with sprint so that players can't effectively run away anymore. This seems like a reasonable balance to that and it would lead to interesting tactical scenarios. Also, they could not always escape speed 10 monsters as many of them have missile attacks and/or friends.

[Lasty]

skjarl: the proposal you're responding to does not buff all monsters; it makes arbitrary speed adjustments to monsters into non-arbitrary ones, and it would guarantee that you can outrun any monster as long as you start running early enough and can take a hit. It would also mean that starting to run after you're already beaten by that monster is no longer arbitrarily either safe or fatal, but is instead predictably dangerous.

Please consider how your proposal would change the current state of the game. The only difference after your proposal is that players can always escape speed 10 monsters if they want to. This is not "balanced"; it's straight up making the game much easier for all non-chei characters.

[johlstei]

Yes, I'm proposing Swiftness be made an ability instead of a spell. Is there something holy and special about it that takes it off the table for any changes? Folks are talking about buffing all monsters with sprint so that players can't effectively run away anymore. This seems like a reasonable balance to that and it would lead to interesting tactical scenarios. Also, they could not always escape speed 10 monsters as many of them have missile attacks and/or friends.

But what does that actually change? Most characters have access to swiftness - it may come later if you are unlucky, but unless you are worshipping trog or chei, the fact that you have to type z to use it instead of a isn't all that interesting. Considering all the monsters that were consciously designed around some characters not outrunning them, it seems foolish to just hand that ability to all characters from XL:1 and send them off. Would you then buff a large variety of monsters to match your swift speed instead of your regular speed? Sounds finicky. If you want to guarantee swiftness you can always pick AE.

[skjarl]

"As long as you can take a hit" is the operative phrase here. Many times you can't take a hit, not even one. GSC vs most characters = 1 shot in the early game. If monsters and players can both sprint, I'm unclear on how that guarantees you will always escape them, especially if you get tired before they do. By "predictably dangerous" you mean "100% fatal" because the monster will be stomping your guts out every step of the way as you try to run. At least with the current randomization mechanics you had a chance of escaping. I think I'd rather have column dancing back than the proposed system.

[Lasty]

As far as I can tell, no one is suggesting that the speed boost be enough for (e.g.) that ogre to catch up with you if you're at least one tile away and running away. If the ogre is right next to you, then the current system already does exactly what you're suggesting is unforgivable: give the ogre one more hit on you.

It wasn't at all clear that you were suggesting that both players are monsters can sprint, but if they could that would be pointless:
The ogre puts on a burst of speed!
You put on a burst of speed!
The ogre begins to slow down.
You begin to slow down.
Etc.
The whole point of all of these systems is to make sure monsters and players don't just keep chasing each other at the same speed forever.

[skjarl]

If the speed boost isn't enough to catch up, what's the point? You're back to column dancing. I think the idea is to force players to fight and most likely die if they failed to escape earlier in the fight. Also, I did suggest that your speed/duration/re-use timers would depend on your stats and/or encumbrance. Same with the monsters, I suppose. I very strongly dislike the idea of an ogre rounding a corner and catching up to my caster before I can get back to the stairs, which is what this will lead to unless I'm misunderstand something.

[Lasty]

The point is that if you're at least one tile away the fight already has an ending: you escape up the stairs. If you're less than one tile away, this prevents pillar dancing from being optimal by guaranteeing that the enemy will get a chance to hit you and then fall behind, giving you space to get up stairs safely, and thus resolving the encounter by ensuring that you are either safe or dead instead of leaving you tediously running in circles.

[skjarl]

Here's my current understanding:

1. Player gets in a fight with monster.
2. Monster kicks the crap out of player.
3. Player tries to column dance/run away, but monster sprints and keeps beating the shit out of the player.
4. Player dies in a pool of his own blood.

or

1. Player walks around a corner and sees a deadly monster 5 tiles away.
2. Players runs away from the monster and toward the stairs.
3. Monster sprints and catches up to the player.
4. Player dies in a pool of his own blood.

What am I misunderstanding?

[Lasty]

Thanks for giving concrete examples; I think that might help us to understand each other. Here's what I'm picturing in your two examples:

Version A:
1. Character gets in a fight with monster.
2. Monster kicks the crap out of character.
3. Character tries to column dance/run away, but monster sprints and gets {x} additional hits in, where {x} can be calibrated but should probably be around 1-2. (In practical terms, monster movement decreased by 1-2 AUT over 2-5 turns, as appropriate -- numbers not yet set in stone)
4. Monster then slows down to below normal speed. If the character hasn't already died in a pool of their own blood, they continue walking away and gain at least 1 tile of distance and can escape upstairs.

Version B:
1. Character walks around a corner and sees a deadly monster 5 tiles away.
2. Character runs away from the monster and toward the stairs.
3. Monster sprints but does not catch up to the character, gaining up to but no more than 1 tile of distance over a short period and then falling behind even further.
4. Character escapes easily, as per now.

Given the way the examples are phrased, I think you missed the aspect of the proposal where the monster speed gain is 1) very small and 2) very short-lived and 3) followed by below-normal speed.

[skjarl]

I see. Let me think about it for a while before forming an opinion. Thanks for the explanation.

[skjarl]

Ok, I can live with this. It's better than the pure randomness of the current system. Hell, this might let nagas actually get away from a fight occasionally.

[Igxfl]

So, if pillar dancing isn't okay, does that mean engaging each monster by throwing a stone and running back fifteen tiles before fighting isn't okay either?
I know it's currently optimal and it probably won't change just out of the difficulty of messing with it, but is that good game design? Really?

[and into]

So, if pillar dancing isn't okay, does that mean engaging each monster by throwing a stone and running back fifteen tiles before fighting isn't okay either?
I know it's currently optimal and it probably won't change just out of the difficulty of messing with it, but is that good game design? Really?

I don't see how selectively getting an enemy's attention and repositioning in anticipation of a fight, being careful about how much gets in your LOS at once, etc., all of which is what makes melee interesting (and thus isn't a design problem), is comparable to pillar dancing, which usually isn't fun or interesting. What am I missing here?

[dpeg]

Alright, for the record: energy randomisation was implemented in order to make pillar dancing less reliable.

If you guys would be responsible for the game and suddenly see everyone and their granny do the pillar dance, you'd try to come up with something as well, I bet.

The whole point of the posts in this thread is that randomized energy doesn't actually make pillar dancing less reliable (except against orc priests)

Thank you for explaining the thread to me, I almost didn't understand it's purpose.

People have been asking why energy randomisation was implemented; I told them why. I don't fully agree with "doesn't make pillar dancing less reliable" -- it does change things slightly. I agree it doesn't change things enough, and that's the other reason why I replied. My proposal that monsters gain quickness was serious. Some details and ramifications:

A monsters knows when it is chasing the player (i.e. this can be reasonably implemented). When it is chasing and doesn't make progress, there is a chance it becomes quick. This chance could trigger after a fixed number of turns chasing, and be constant for each turn chasing afterwards. Once a monster is quick, it never becomes unquick again. Display a quick gnoll as "quick gnoll".

Some half-assed analysis: this method means that approaching monsters for a fight is a more serious decision. I like that is also helps against kiting, which is a really annoying tool. It might be overdoing things (just like energy randomisation underperformed), and I won't push the issue in the near future. (I think there's a reason that no other dev felt like replying in this tread.) Anyway, that's what I'd do about pillar dancing and kiting these days.

[some12fat2move]

Perhaps we could make the player's normal movement speed a non-integer number (for example, 9.5 or 9.8), that way you are never the same speed as an enemy?

[damiac]

Dpeg, that quickness thing would be a huge increase in difficulty. Suddenly the strategies of 'run away when things get dangerous' or 'try to attract enemy attention, then lead somewhere safe' doesn't work. Meaning, essentially, the RNG becomes more powerful, and player choices become less powerful.

That's not to say it couldn't be balanced, but right now it sounds like 'Play a powerful character or die to the first enemy you can't beat, because there's no escape'. Wake up grinder on D2? Better hope you're next to a staircase.