FR: Speed 11+ branch

[tedric]

I would really like to try a branch where there was no such thing as a speed 10 monster -- i.e. where the lowest possible innate monster speed was either 11 or [edit: NOPE FINE WHATEVER JUST MAKE IT 11]. This kind of thing gets talked about all the time whenever folks start griping about Crawl's inherent flaws, but I'm not aware that it has ever actually been tested. Unfortunately I can't code such a branch myself, so I hope somebody else will. Thanks in advance.

[Pollen_Golem]

either 11 or PlayerSpeed+1 (preferably the latter)

The latter would benefit nagas and make spriggans impossible, and make goliath beetles the new sigmunds

[Hirsch I]

what's a goliath beetle?

[tasonir]

playerspeed + 1 would make Chei interesting. I assume you mean the player's base speed. It'd probably be better to not adjust for race either, and just set it to 11.

[CanOfWorms]

The new wizlab, Golubria's Singularity Collider!

[reaver]

Isn't Spider basically this?

I guess it has emperor scorpions and entropy weavers etc. but most of the threats in Spider are fast.

[duvessa]

Spider doesn't have anything dangerous in it though, so it's not useful for trying this out. I think that by "branch" the OP meant a git branch rather than an in-game branch. Probably changing the speed of all monsters that routinely generate before Lair would be enough to gather all the useful data/experience.

[tedric]

Yes, git branch.

I appreciate that basing it on player speed would interact weirdly with speed-modified player species, so maybe 11 minimum for monsters is the way to go. But I guess if it appeared to be a desired change after a bunch of playtesting with normal-speed races, the whole concept of faster/slower player species would be thrown into question, so maybe Naga/Spriggan/Felid considerations are unimportant for early testing. Anyway, the point is I would like to actually experience a game where the full variety of monsters I meet are faster than me, but I don't have to play only Naga or Chei characters.

[Siegurt]

Why not simply add 1 to the base movement delay of all races if that's what you're after (Less changes to be made, probably easier to implement)

Unless you're specifically seeking a branch where all creatures *attack* faster than you as well (That is the sort of change you could do yourself pretty quickly/easily)

(In fact you could do this in one very very simple place, find the end of player_movement_speed and change "return mv;" to "return mv+1;")

[dowan]

You could call this branch "Shout it from the stairways" or "Killhole mounted sword turret".

I think this sort of change would only serve to make crawl much more tedious, as it effectively hampers player mobility, in a game where the 'experts' seem to agree that the most interesting aspect is positioning.

[Sar]

PlayerSpeed+1

why would you do that

[Berder]

Spider doesn't have anything dangerous in it though,

That really depends on your character. Spider is unbelievably dangerous for high-ev low-ac characters, to the point that I'd often rather do shoals first on a felid or octopode. Why? Webs destroy EV/SH (while being far less dangerous if you have 30 AC), and paralysis is an absolute death sentence if you don't have enough ac to tank it. Also there are ghost moths who are a huge problem if you are a blaster caster without sInv+invis or strong melee.

Spider is really only easy if you are a tanky fighter.


About the OP, spider qualifies as a fast-monster branch and so does shoals, really. Also lair. No, not everything in lair is faster than you - just the most deadly things (black mambas, spiny frogs, blink frogs). Plus there's swamp, full of hydras, shambling mangroves (once they root you they're faster than you), ravens, spiny frogs, komodo dragons, wasps, and spriggan riders. And there's freakin snake. Nagas are slow, you say? Well anacondas, shock serpents, and the other snakes aren't, and you can't just walk away from a pack of slow nagas if there are fast nasties with them.

Really all it takes is one dangerous fast monster in a pack, and you cannot safely walk away from the pack, so it's as if the whole pack is fast.

[tedric]

You could call this branch "Shout it from the stairways" or "Killhole mounted sword turret".

I think this sort of change would only serve to make crawl much more tedious, as it effectively hampers player mobility, in a game where the 'experts' seem to agree that the most interesting aspect is positioning.

this is exactly the sort of "armchair game design" that my request for a test branch is meant to resolve

[duvessa]

walking away from monsters that are faster than you is actually exactly what you should do almost all of the time

[tedric]

find the end of player_movement_speed and change "return mv;" to "return mv+1;"

If it is that easy to change, surely someone with more experience writing patches (I have literally zero, so the bar is not high) could implement it as an experimental branch and slap it up on a server, yes? I know I phrased my OP as "I would like to play" but really the implication was "I would like a bunch of us to be able to play and collect data about", which means online play.

[tedric]

some monsters are fast

The point of my OP request is not to experience what it's like to have fast monsters exist, it is to experience what it's like to have literally every monster in the game be faster than you. Maybe you fell prey to the "dungeon branch"/"git branch" confusion referenced above? I'm sorry about that, everyone, but English is weird, what can you do.

Anyway, the only monsters that actually prevent you from running from normal- or slow-moving packs are those that prevent you from moving entirely (confusion/paralysis/constriction/petrification) and those with Blink Allies Encircling (or just Blink, if they happen to block off a 1-tile escape corridor). Speed has nothing to do with it -- in fact I will often let a fast enemy nip at my heels while retreating until I can put a hard barrier between us and the slower pack following, or at least until I reach a chokepoint. Then I can take out the fast enemy 1-on-1. This is a very reliable tactic thanks to stairs, doors, stealth, corners, 1-tile hallways, etc. On the other hand, if I can't take out the fast enemy before the pack catches up, AND I can't survive the fast enemy picking at me while I walk away, then I actually can't take the fast enemy at all -- so whatever I do next is exactly what I would do if I met that enemy without a slower pack behind it.

[celem]

Spider doesn't have anything dangerous in it though,

Really all it takes is one dangerous fast monster in a pack, and you cannot safely walk away from the pack, so it's as if the whole pack is fast.

I have the opposite experiences in general with this mechanic. To take one of your examples... I'm in Snake, I see a pack of Nagas I dont much want to fight and they brought anacondas. The fact that they have different movements speeds excludes them from being a pack at all in my eyes, they are intrinsically seperateable (if I outrun the nagas then I can kite the anacondas until the nagas lose the trail, then stop and kill the fast snakes. Yes constriction is kind of tailored to combat exactly this, but its still not going to stop me trying (and often succeeding). Infact constriction forces me to choose this tactic, since like webbing it is very nasty for low-AC characters. Its the same problems you get as a necro with AD, you are not suddenly a player character pack, unless you actually set your speed to match fastest ally.

Where fast monsters are genuinely scary is where the whole pack is fast. Nothing sums this feeling up for me better than killer bees, which are quicker than almost everyone and famed early-D killers. Later on you get the same vibe with the more exotic batty types, shrikes/scarabs and the like. Spriggan squads in Depths... I dont much find 1 fast monster intimidating, no matter what slow beasties he has as buddies, because he wont stick on them and I can (will) split him. Note that im not advocating any alteration to monster AI, if stuff couldnt be lured out like this then you reduce the number of things I can actually do to manage the situations the game is throwing at me, straight up difficulty increase. Encounters do not need to be binary, saying go around is not enough. Conflicts should not be reducable to fight it all right now or go away, I like being able to whittle away at a problem.

[all before]

I would also like to try a base speed 11 branch.

[Berder]

I have the opposite experiences in general with this mechanic. To take one of your examples... I'm in Snake, I see a pack of Nagas I dont much want to fight and they brought anacondas. The fact that they have different movements speeds excludes them from being a pack at all in my eyes, they are intrinsically seperateable (if I outrun the nagas then I can kite the anacondas until the nagas lose the trail, then stop and kill the fast snakes

Just walking away from anacondas at speed 10 - especially if my HP wasn't full at the start - is an extremely dangerous proposition with most of my chars. Anacondas are fast enough to hit you a lot while chasing you, and they can hit very hard and constrict you. If this move is not crazy, then your char is ridiculously overpowered for snake.

[dowan]

Now imagine if for some reason that anaconda got to hit you more times as you walked toward the choke point, making it more likely to constrict you out in the open before you got to your chokepoint. Now what do you do? Pop a teleport scroll every time you see an anaconda with nagas? Just play strong combos that can win this fight in the open? Just shout a lot from safe chokepoints?

The solution only seems to create even more tedium than the problem it claims to solve. If I can't move fights to an area that suits me, I will just make sure the fights come to me, in an area that suits me. It's true that right now that's an optimal tactic, but the ability to lure creatures means I don't have to do that. The current system is balanced between danger and tedium to a degree, but all you're doing is making the dangerous but fun option even more dangerous, and making the tedious but safe option even more necessary.

Luring may be problematic in some ways, but the reason players usually do it is because you have to do it not to die. Luring is a symptom of the real problem, which is that you have to make sure fights happen in advantageous positions to be able to win them. I'm not sure I'd even call that a problem, it's really just basic tactics. So maybe if the problem is that people have to lure monsters across half a map, the real solution is to make sure there are more hallways generated to fight in, so you don't have to drag monsters so far. Or maybe the solution is to make monsters generally easier so you can just fight them all out in the open.

Or maybe there's actually not a problem, so no solution is needed.

[Siegurt]

find the end of player_movement_speed and change "return mv;" to "return mv+1;"

If it is that easy to change, surely someone with more experience writing patches (I have literally zero, so the bar is not high) could implement it as an experimental branch and slap it up on a server, yes? I know I phrased my OP as "I would like to play" but really the implication was "I would like a bunch of us to be able to play and collect data about", which means online play.

Yes, it's easy to implement, the problem is getting someone with an active server to implement said experimental change and put it in an experimental branch, it's not a technical problem, it's a political one (i.e. you have to convince someone to do it)

You'd probably be better off chatting on ##crawl-dev and see if you can get someone to align their interests with yours.

[bel]

Just walking away from anacondas at speed 10 - especially if my HP wasn't full at the start - is an extremely dangerous proposition with most of my chars. Anacondas are fast enough to hit you a lot while chasing you, and they can hit very hard and constrict you. If this move is not crazy, then your char is ridiculously overpowered for snake.

You are talking about the worst case. The typical case is something like this. I see a Naga band, perhaps a Naga ritualist and lots of snake at the edge of my LoS. I retreat a fair bit, letting the snakes catch up to me, while Nagas are left behind. I might take a couple of hits, but not more. Then I deal with the snakes. Then I rest in a safe place, and go back to the Nagas.

Really, speed 11 is basically to just make sure that infinite luring and pillar dancing don't work so easily. Luring still works. Example: a bunch of yaks in the open. You walk to a chokepoint or water while merfolk etc., while taking a few hits. Then deal with yaks in the chokepoint.

Also, no harm in trying this out to see how hard it would be. My guess is not that hard. But then I don't play much junk species. I have played a fair number of Op though. (Not saying Op is junk, it is a fairly good species)

[Sandman25]

Now imagine if for some reason that anaconda got to hit you more times as you walked toward the choke point, making it more likely to constrict you out in the open before you got to your chokepoint. Now what do you do? Pop a teleport scroll every time you see an anaconda with nagas? Just play strong combos that can win this fight in the open? Just shout a lot from safe chokepoints?

Solution to this is "don't explore open terrain when you cannot deal with potential consequences". Leaving some areas unexplored is not functionally different from putting exclusions on sleeping monsters except you don't know what monsters are sleeping there. Similarly don't explore dead end corridors when there is some dangerous monster wandering on the level. Also "dive to next level when there is wandering foo on the level" is in the same category.

[ontoclasm]

If you're wondering about how difficult this is to implement, it is a one-line change to apply this to almost every monster in the game (you just set the cost of walking one square to 9 instead of 10). I did it once just to see how hard it would be code-wise and how much of an impact it would actually have (results: it's nigh-unto undetectable unless you're looking for it; you get hit like once or twice while running away from stuff but that happens sometimes anyway).

I guess it's like a 4-line change if you remove energy randomization at the same time.

[dowan]

Now imagine if for some reason that anaconda got to hit you more times as you walked toward the choke point, making it more likely to constrict you out in the open before you got to your chokepoint. Now what do you do? Pop a teleport scroll every time you see an anaconda with nagas? Just play strong combos that can win this fight in the open? Just shout a lot from safe chokepoints?

Solution to this is "don't explore open terrain when you cannot deal with potential consequences". Leaving some areas unexplored is not functionally different from putting exclusions on sleeping monsters except you don't know what monsters are sleeping there. Similarly don't explore dead end corridors when there is some dangerous monster wandering on the level. Also "dive to next level when there is wandering foo on the level" is in the same category.

Well, not really, the solution is 'yell a lot when in advantageous territory'.
The idea of just having to hug the walls all day does not sound like a more fun crawl to me, and I don't see how it reduces tedium. All you've done is make the 'o' key a really bad idea to press, and the 't' key really important to press a lot.

[Sandman25]

Well, not really, the solution is 'yell a lot when in advantageous territory'.
The idea of just having to hug the walls all day does not sound like a more fun crawl to me, and I don't see how it reduces tedium. All you've done is make the 'o' key a really bad idea to press, and the 't' key really important to press a lot.

Yes, 'tt' is useful here indeed. Yet it has 2 limitations:
1) Shouting is not that loud. For instance, there are many Elf 3 vaults where shouting does not allow player to get all monsters to kill hole
2) It can be dangerous to attract too many monsters.