SquareLOS experimental branch on CBRO


Although the central place for design discussion is ##crawl-dev on freenode, some may find it helpful to discuss requests and suggestions here first.

Tomb Titivator

Posts: 909

Joined: Thursday, 3rd January 2013, 20:32

Post Tuesday, 28th April 2015, 05:14

SquareLOS experimental branch on CBRO

http://crawl.berotato.org:8080/#play-dc ... relos-0.17

Any devs care to comment on the current attitude toward this implementation and/or the kind of feedback that would be most helpful in evaluating it?

I just played a few minutes and my immediate impression is that targeting short-range spells like Flame Tongue is MUCH more intuitive. Didn't get to wide-open levels where the full square would matter more though (i.e. b/c there are a few more tiles in LOS that could be filled w/ enemies)
Wins (Does not include my GrEE^Veh 15-runer...stupid experimental branch)
User avatar

Shoals Surfer

Posts: 293

Joined: Tuesday, 19th February 2013, 18:55

Post Tuesday, 28th April 2015, 05:42

Re: SquareLOS experimental branch on CBRO

interesting

I kinda like it
I love pitsprint and pitsprint culture.
dpeg wrote:The only good player is a dead player.
User avatar

Tomb Titivator

Posts: 802

Joined: Sunday, 30th March 2014, 21:06

Post Tuesday, 28th April 2015, 06:14

Re: SquareLOS experimental branch on CBRO

It takes some getting used to..
It seems like movement can bring a lot more enemies into LOS than usual.
Comborobin Admin

Tomb Titivator

Posts: 853

Joined: Thursday, 29th August 2013, 18:39

Post Tuesday, 28th April 2015, 14:57

Re: SquareLOS experimental branch on CBRO

yes, YES

chebyshev is my homeboy
User avatar

Barkeep

Posts: 4435

Joined: Tuesday, 11th January 2011, 12:28

Post Tuesday, 28th April 2015, 15:14

Re: SquareLOS experimental branch on CBRO

Huh. I thought it would be really weird and unnatural, but it wasn't. I looked at webtiles, but normally play console; maybe it would look odd in console?

Anyhow, for my two dungeon levels of a HuIE, it seemed nice.
I am not a very good player. My mouth is a foul pit of LIES. KNOW THIS.

Tartarus Sorceror

Posts: 1774

Joined: Tuesday, 23rd December 2014, 23:39

Post Tuesday, 28th April 2015, 17:31

Re: SquareLOS experimental branch on CBRO

xentronium wrote:It takes some getting used to..
It seems like movement can bring a lot more enemies into LOS than usual.

Yes - moving diagonally, it can bring 29 squares into view, while moving orthogonally it brings 15 squares into view. Under circleLOS, moving diagonally brings 23 squares into view, and orthogonally brings 17 squares into view. The advantage of moving orthogonally is therefore substantially greater under squarelos.
streaks: 5 fifteen rune octopodes. 15 diverse chars. 13 random chars. 24 NaWn^gozag.
251 total wins Berder hyperborean + misc
83/108 recent wins (76%)
guides: safe tactics value of ac/ev/sh forum toxicity

Tomb Titivator

Posts: 853

Joined: Thursday, 29th August 2013, 18:39

Post Tuesday, 28th April 2015, 18:04

Re: SquareLOS experimental branch on CBRO

Berder wrote:
xentronium wrote:It takes some getting used to..
It seems like movement can bring a lot more enemies into LOS than usual.

Yes - moving diagonally, it can bring 29 squares into view, while moving orthogonally it brings 15 squares into view. Under circleLOS, moving diagonally brings 23 squares into view, and orthogonally brings 17 squares into view. The advantage of moving orthogonally is therefore substantially greater under squarelos.

Diagonal movement is safer now in that you encounter enemies 8 tiles away, rather than potentially finding them only 5?(not sure on exact number) tiles away. The set of enemies that are dangerous at 8 tiles away is smaller than the set of enemies dangerous at 5 tiles. So yes, you'll reveal more tiles per step, but you're guaranteed to not be surprised by something much closer than your max LOS. (Of course walls change this and block LOS, so there's more to it than that, but we're talking about frictionless open dungeon rooms here.)

The inverse is also true - you now mask 29 tiles by retreating diagonally vs only 15 by retreating orthogonally.

For this message the author johlstei has received thanks:
Arrhythmia

mps

Tomb Titivator

Posts: 886

Joined: Saturday, 3rd January 2015, 22:34

Post Tuesday, 28th April 2015, 18:27

Re: SquareLOS experimental branch on CBRO

njvack wrote:Huh. I thought it would be really weird and unnatural, but it wasn't. I looked at webtiles, but normally play console; maybe it would look odd in console?


Well, remember, squarelos is what almost every other video game you've ever played uses, so...
Dungeon Crawling Advice tl;dr: Protect ya neck.

For this message the author mps has received thanks: 6
Arrhythmia, duvessa, PleasingFungus, rockygargoyle, Sar, tedric
User avatar

Barkeep

Posts: 1788

Joined: Saturday, 29th June 2013, 16:52

Post Tuesday, 28th April 2015, 22:48

Re: SquareLOS experimental branch on CBRO

My squarelos DEFE fell through a shaft for three floors, and landed next to Gastro, a jackal pack, and a hive vault. My un-IDed scroll was ?noise. Then I started an OgBe that found shadow dragon armour on D:1.

I noticed no problems, and haven't really felt the impact of the change at all so far, other than the fact that my instincts re: aiming spells/large rocks is off. I bet it's a bigger difference on bigger floors, as others have said.
User avatar

Tartarus Sorceror

Posts: 1881

Joined: Saturday, 7th September 2013, 21:16

Location: Itajubá, MG, Brazil.

Post Tuesday, 28th April 2015, 22:55

Re: SquareLOS experimental branch on CBRO

I dont really see the reason for the change: what is it supposed to fix?
my posts are to be read in a mildly playful tone, with a deep, sexy voice.

For this message the author Hirsch I has received thanks: 2
Sar, XuaXua

Sar

User avatar

Ziggurat Zagger

Posts: 6418

Joined: Friday, 6th July 2012, 12:48

Post Tuesday, 28th April 2015, 23:00

Re: SquareLOS experimental branch on CBRO


For this message the author Sar has received thanks:
Hirsch I
User avatar

Tartarus Sorceror

Posts: 1881

Joined: Saturday, 7th September 2013, 21:16

Location: Itajubá, MG, Brazil.

Post Tuesday, 28th April 2015, 23:06

Re: SquareLOS experimental branch on CBRO

damn, I almost felt loved for a moment here. I can go back to cutting myself.
serioulsy now, thanks for the link. I missed the discussion, tought it happened on the wiki or the IRC channel.
my posts are to be read in a mildly playful tone, with a deep, sexy voice.

Dungeon Master

Posts: 1613

Joined: Thursday, 16th December 2010, 21:54

Post Tuesday, 28th April 2015, 23:30

Re: SquareLOS experimental branch on CBRO

The commit also has a good explanation.

For this message the author Kate has received thanks: 2
Sprucery, tedric

Ziggurat Zagger

Posts: 8786

Joined: Sunday, 5th May 2013, 08:25

Post Tuesday, 28th April 2015, 23:33

Re: SquareLOS experimental branch on CBRO

Also maybe relevant

Experimental square LOS has been on big servers 2 times before that I can remember; it was great! I'm not going to have the time to waste on crawl for at least a few weeks so I hope this time it either lasts a while or finally makes it into master.

Yes, square LOS changes tactics, because the elements that were important with circular LOS and ranges - approaching monsters from diagonals, making melee monsters approach you from not-diagonals, reaching, digging, etc. shenanigans - aren't around anymore. This is kind of the whole point of changing to square LOS in the first place, those tactics arise specifically because of crawl's inconsistent geometry.

The main balance implication is that something with a range of, say, 5 now actually has a range of 5, instead of a range of "5 when you want it to be 5 and 3 when you want it to be 3". This is a big buff to ranged monsters that are being approached in the open since they get up to twice as many opportunities to attack before you can hit them in melee, and of course a nerf to stealth. But that can be fixed, and is a small price to pay for getting rid of the glaring geometry inconsistency that affected every fight in the game.

The number of vaults that are engineered around existing LOS, and actually work, is pretty small, btw. Most of them don't have monsters!

For this message the author duvessa has received thanks:
Arrhythmia
User avatar

Tartarus Sorceror

Posts: 1762

Joined: Monday, 14th October 2013, 01:05

Post Wednesday, 29th April 2015, 00:11

Re: SquareLOS experimental branch on CBRO

Hirsch I wrote:I dont really see the reason for the change: what is it supposed to fix?

Image
With circle LOS, you can see almost twice as far at 90 degree angles as you can diagonally. This means that if you approach monsters from diagonals, they have way less time to react to you(which makes a big difference when approaching things like smiting/tormenting/hellfiring/summoning/sleeping enemies).

Crawl should not have magic directions that it is safer to approach enemies from(or that make it easier for enemies to approach you, since this fucks with player range as well. If a spell has 5 range, it should have 5 range in every direction. Not so with circle LOS).

Mines Malingerer

Posts: 35

Joined: Friday, 16th January 2015, 08:14

Post Wednesday, 29th April 2015, 04:27

Re: SquareLOS experimental branch on CBRO

I much prefer circular LoS and always thought the 1.41 movement cost on diagonals change was better, but it seems like everybody else loves squarelos, so...

Tomb Titivator

Posts: 909

Joined: Thursday, 3rd January 2013, 20:32

Post Wednesday, 29th April 2015, 05:37

Re: SquareLOS experimental branch on CBRO

Penalizing diagonal movement doesn't actually solve the problem. In fact, it probably makes it worse! Because now you still have the fucked-up unintuitive targeting issues, AND you've given the player an incentive to go to great lengths to minimize the chance that they will be forced to move diagonally during combat. Ever.

Also it's worth noting that the commit reduces max LOS from 8 to 7 (and reduces the range of many spells accordingly). For those keeping score: Berder's math above is correct regarding how many tiles this reveals diagonally vs. horizontally; johlstei's comment about diagonal movement being even safer because 8>5 is also still correct, but 1/3 less so.
Wins (Does not include my GrEE^Veh 15-runer...stupid experimental branch)

For this message the author tedric has received thanks: 4
Arrhythmia, duvessa, Rast, Sar

Lair Larrikin

Posts: 28

Joined: Monday, 13th April 2015, 13:19

Post Wednesday, 29th April 2015, 12:57

Re: SquareLOS experimental branch on CBRO

Personally I like the quirks of circle LoS when it comes to approaching enemies diagnally. I always internalized this as flanking rather than approaching head-on.

For this message the author tannisroot has received thanks:
Rast
User avatar

Barkeep

Posts: 4435

Joined: Tuesday, 11th January 2011, 12:28

Post Wednesday, 29th April 2015, 15:41

Re: SquareLOS experimental branch on CBRO

Is... is there actually a chance this would make it to Trunk for widespread experimentation? That would be baller.
I am not a very good player. My mouth is a foul pit of LIES. KNOW THIS.

For this message the author njvack has received thanks:
Sar

Slime Squisher

Posts: 411

Joined: Saturday, 9th March 2013, 14:22

Post Wednesday, 29th April 2015, 17:19

Re: SquareLOS experimental branch on CBRO

wouldn't that be too big a change to just impose it on everyone eventually? i personally wouldn't like it on the sole basis it bothers my sense of aestethics, i'd likely not play crawl with a squared los.

For this message the author adozu has received thanks:
Berder

Ziggurat Zagger

Posts: 8786

Joined: Sunday, 5th May 2013, 08:25

Post Wednesday, 29th April 2015, 17:26

Re: SquareLOS experimental branch on CBRO

Why is it worse than imposing circular LOS on everyone?

For this message the author duvessa has received thanks: 3
Arrhythmia, PleasingFungus, Sar

Dungeon Master

Posts: 3618

Joined: Thursday, 23rd December 2010, 12:43

Post Wednesday, 29th April 2015, 17:30

Re: SquareLOS experimental branch on CBRO

adozu wrote:wouldn't that be too big a change to just impose it on everyone eventually?
No, of course not. I don't think this change is bigger than what Crawl is going through anyhow.

i personally wouldn't like it on the sole basis it bothers my sense of aestethics, i'd likely not play crawl with a squared los.
That's fine, but for us, gameplay trumps lesser (desirable) features such as realism, flavour, aesthetics. Whether squares look uglier than circles is a subjective matter anyway... I have a feeling that most players will cope fine.

Slime Squisher

Posts: 411

Joined: Saturday, 9th March 2013, 14:22

Post Wednesday, 29th April 2015, 17:30

Re: SquareLOS experimental branch on CBRO

this isn't just shortening lair branches or removing a spell school, it changes how the game is perceived by players on a fundamental level. Same could be said if it was squared for decades and suddendly turned circular. At most you'd ought to give player a choiche there.

it's not like squared los is whitout it's own set of issues (can't really argue on an absolute basis that square > circular)
User avatar

Barkeep

Posts: 4435

Joined: Tuesday, 11th January 2011, 12:28

Post Wednesday, 29th April 2015, 17:32

Re: SquareLOS experimental branch on CBRO

adozu wrote:wouldn't that be too big a change to just impose it on everyone eventually? i personally wouldn't like it on the sole basis it bothers my sense of aestethics, i'd likely not play crawl with a squared los.

Have you actually tried playing it? I thought it would bug my aesthetic sense, but it felt really natural. If it just switched, I'd notice it but it wouldn't be a "WTF????" moment.
I am not a very good player. My mouth is a foul pit of LIES. KNOW THIS.

Tomb Titivator

Posts: 853

Joined: Thursday, 29th August 2013, 18:39

Post Wednesday, 29th April 2015, 17:37

Re: SquareLOS experimental branch on CBRO

adozu wrote:this isn't just shortening lair branches or removing a spell school, it changes how the game is perceived by players on a fundamental level. Same could be said if it was squared for decades and suddendly turned circular. At most you'd ought to give player a choiche there.

it's not like squared los is whitout it's own set of issues (can't really argue on an absolute basis that square > circular)

This is status quo bias - how long something has been some way should have no bearing on whether it changes or not, it should only be based on if it's an improvement. Personally I think harmony of design is more important than the concerns expressed - it is very confusing that LOS uses fake-euclidean distance while movement uses chebyshev distance, and making movement use fake-euclidean distance would be a nightmare, so having both use chebyshev seems like the idea solution to me.

By the way people threaten to quit crawl over every single change, some dude said he quit because rings of regeneration became amulets of regeneration, you're not going to sway anyone with your hunger strike. Sorry if you genuinely stop playing crawl if this change goes in, but most of the time people saying stuff like that are just posturing.
Last edited by johlstei on Wednesday, 29th April 2015, 17:42, edited 3 times in total.

For this message the author johlstei has received thanks: 5
archaeo, Arrhythmia, Curio, elmdor, tedric
User avatar

Barkeep

Posts: 4435

Joined: Tuesday, 11th January 2011, 12:28

Post Wednesday, 29th April 2015, 17:38

Re: SquareLOS experimental branch on CBRO

adozu wrote:it's not like squared los is whitout it's own set of issues (can't really argue on an absolute basis that square > circular)

... such as?

The situations described in the commit log range from encouraging tedious tactics (see dangerous monster, position to be either diagonal or orthogonal depending on whether I am afraid of its ranged attack or it should be afraid of mine) to bizarre (the thing with diagonal movement before casting to get additional shots).

I think that in Crawl, you can absolutely argue that square is better than circular.

I guess you can argue that "space does not work like squarelos" but space does not work like circlelos either; the closest you could probably get is a hex grid. But that's not going to be exact either because they're all grids and space does not work like grids.
I am not a very good player. My mouth is a foul pit of LIES. KNOW THIS.

mps

Tomb Titivator

Posts: 886

Joined: Saturday, 3rd January 2015, 22:34

Post Wednesday, 29th April 2015, 17:41

Re: SquareLOS experimental branch on CBRO

adozu wrote:this isn't just shortening lair branches or removing a spell school, it changes how the game is perceived by players on a fundamental level. Same could be said if it was squared for decades and suddendly turned circular. At most you'd ought to give player a choiche there.

it's not like squared los is whitout it's own set of issues (can't really argue on an absolute basis that square > circular)


There's always older versions. If it's a real problem, I expect there'll be a fork, but I really doubt this will be the issue that leads to a serious competing variant.
Dungeon Crawling Advice tl;dr: Protect ya neck.

For this message the author mps has received thanks:
Arrhythmia

Slime Squisher

Posts: 411

Joined: Saturday, 9th March 2013, 14:22

Post Wednesday, 29th April 2015, 17:41

Re: SquareLOS experimental branch on CBRO

ok, how about we change crawl from squares to hexes? it solves most of the circular los issues. (just a hypothesis)

it would be an equally drastic and effective solution. let's assume it could be made to work (let's ignore for a moment how hard would the implementation be with existing content adaptation etc) would it still be the same game?

would it be ok to just turn crawl into hexes? I think a matter like this can't be evaluated on a purely gameplay basis. Then again mine is only a singular opinion.

PS: let's not go on ad hominems, i'm not calling a hunger stirke, in my case i'd just play an older version. I don't think there's any need to escalate the topic i just meant to express a pow. I did not expect to stir such a reaction that i can't even type a reply whitout getting a more replies notification.

Tomb Titivator

Posts: 853

Joined: Thursday, 29th August 2013, 18:39

Post Wednesday, 29th April 2015, 17:47

Re: SquareLOS experimental branch on CBRO

adozu wrote:ok, how about we change crawl from squares to hexes? it solves most of the circular los issues. (just a hypothesis)

it would be an equally drastic and effective solution. let's assume it could be made to work (let's ignore for a moment how hard would the implementation be with existing content adaptation etc) would it still be the same game?

would it be ok to just turn crawl into hexes? I think a matter like this can't be evaluated on a purely gameplay basis. Then again mine is only a singular opinion.

PS: let's not go on ad hominems, i'm not calling a hunger stirke, in my case i'd just play an older version. I don't think there's any need to escalate the topic i just meant to express a pow. I did not expect to stir such a reaction that i can't even type a reply whitout getting a more replies notification.

I'm not trying to ad hominem you - there are changes that would make me stop playing new versions of crawl as well, but my thoughts on how crawl should work seem to be pretty well-aligned with that of the devs, so I'm not too worried. It's an okay thing to be, but I'm mostly just calling your bluff. Sorry if you perceive that as an attack - sometimes I say things I don't mean and I appreciate it when people call me on it, but it's understandable if you don't feel that way.

Yes, it would absolutely be okay to turn crawl into hexes and still call it crawl. It would be a huge undertaking and I don't know if it would be an improvement or not, but it'd be cool and I'd play it if that's what happened in 0.19 or something. It would be a much more drastic change than squareLOS, but it might be worth it.

PS http://mbays.freeshell.org/hexcrawl/ you can try out some ancient version there
User avatar

Pandemonium Purger

Posts: 1283

Joined: Thursday, 16th April 2015, 22:39

Post Wednesday, 29th April 2015, 17:57

Re: SquareLOS experimental branch on CBRO

That hex crawl is totally ASCII. I'm guessing that overhauling the graphical presentation would take the most work, while having blank spaces between tiles would be immensely ugly and distracting.

Unless... you do this:
  Code:

     ----
    /****\
   / **** \
   \ **** /
    \****/
     ----

Where the 16 asterisks represent the original square tile graphic, and the rest is filled in with background.

mps

Tomb Titivator

Posts: 886

Joined: Saturday, 3rd January 2015, 22:34

Post Wednesday, 29th April 2015, 18:10

Re: SquareLOS experimental branch on CBRO

I think the objection to actual LoS being square on an aesthetic level is totally crazy, but the objection to square shaped spell effects is more reasonable -- for example, square tornadoes with monsters pushed around inside along square shaped paths is a genuinely weird idea.

I would just say that spell effects do not have to be constrained by any rules of geometry. If it looks better and there's some tactical interest to it, by all means bring back circlish AOE effects.
Dungeon Crawling Advice tl;dr: Protect ya neck.

For this message the author mps has received thanks:
chequers
User avatar

Barkeep

Posts: 4435

Joined: Tuesday, 11th January 2011, 12:28

Post Wednesday, 29th April 2015, 18:18

Re: SquareLOS experimental branch on CBRO

mps wrote:I would just say that spell effects do not have to be constrained by any rules of geometry. If it looks better and there's some tactical interest to it, by all means bring back circlish AOE effects.

As long as it's OK for the spell to have the same kinds of odd range behavior that circlelos does, then yeah. But moving enemies in a "circle" pattern would increase and decrease their range to the player, which could have some really strange effects. In a game with Chebyshev movement, "two moves away" definitionally describes a square.
I am not a very good player. My mouth is a foul pit of LIES. KNOW THIS.

Tomb Titivator

Posts: 853

Joined: Thursday, 29th August 2013, 18:39

Post Wednesday, 29th April 2015, 18:31

Re: SquareLOS experimental branch on CBRO

mps wrote:I think the objection to actual LoS being square on an aesthetic level is totally crazy, but the objection to square shaped spell effects is more reasonable -- for example, square tornadoes with monsters pushed around inside along square shaped paths is a genuinely weird idea.

I would just say that spell effects do not have to be constrained by any rules of geometry. If it looks better and there's some tactical interest to it, by all means bring back circlish AOE effects.

Hey man, you gotta dance with the beautiful topology of crawl, this is in fact a circle of radius 2 in crawltopia:
  Code:
#####
#####
#####
#####
#####

For this message the author johlstei has received thanks:
Arrhythmia

Ziggurat Zagger

Posts: 8786

Joined: Sunday, 5th May 2013, 08:25

Post Wednesday, 29th April 2015, 18:39

Re: SquareLOS experimental branch on CBRO

adozu wrote:Same could be said if it was squared for decades and suddendly turned circular.
This is exactly what happened in 0.7 when ranges were made circular instead of square. I wonder why you didn't complain about that?

mps wrote:I think the objection to actual LoS being square on an aesthetic level is totally crazy, but the objection to square shaped spell effects is more reasonable -- for example, square tornadoes with monsters pushed around inside along square shaped paths is a genuinely weird idea.
Why is it genuinely weird in Chebyshev geometry? The Chebyshev equivalent of a Euclidean circle looks like a Euclidean square. That's really all there is to it. The Euclidean equivalent to Crawl's "circles" looks like a star (but with aliasing because of the grid). If your line of sight (or tornado) was a wobbly star in a game with Euclidean geometry, everyone would call it crazy. So why do Crawl, NetHack, and ToME get a free pass for doing the same thing?

For this message the author duvessa has received thanks:
Arrhythmia

Tartarus Sorceror

Posts: 1774

Joined: Tuesday, 23rd December 2014, 23:39

Post Wednesday, 29th April 2015, 18:57

Re: SquareLOS experimental branch on CBRO

I think circleLOS is fine. It has the advantage that the number of squares revealed when moving diagonally and orthogonally are more similar.

In squareLOS, speedrunners or people looking for timed portals have a substantial incentive to explore large areas by moving diagonally (it's almost twice as good as moving orthogonally). And people trying to avoid trouble have the opposite incentive.

Essentially you're trading one kind of directionality for another. It's not a straight upgrade. You're reducing directionality in tactical combat situations, and increasing directionality in exploration situations.

And what exactly is wrong with having tactical quirks in combat? The game is full of tactical tricks. I think it's not an issue that you can sometimes take advantage of this one. In practice, for most of the game I don't find myself worrying at all about circleLOS, and don't find it annoying at all in the cases where it does matter.

Plus I do think circleLOS looks good.
streaks: 5 fifteen rune octopodes. 15 diverse chars. 13 random chars. 24 NaWn^gozag.
251 total wins Berder hyperborean + misc
83/108 recent wins (76%)
guides: safe tactics value of ac/ev/sh forum toxicity

For this message the author Berder has received thanks: 2
Rast, The Ferret

Ziggurat Zagger

Posts: 6454

Joined: Tuesday, 30th October 2012, 19:06

Post Wednesday, 29th April 2015, 19:35

Re: SquareLOS experimental branch on CBRO

Given the choices availaible in a vacumn, I would prefer:

1. HexLOS
2. CircleLOS with psudo-euclidean movement and ranges
3. SquareLOS
4. CircleLOS with psudo-euclidian ranges, and Chebyshev movement (what we have now)

1. Is hard to implement in ascii, (You need to use two offset characters to represent each hex) and would require more drastic changes in crawl (pretty much all the vaults would need to be tossed, and a lot of the graphics would need to be redone, map generation would need to be rewritten, combat AI would change drastically etc. etc.)

2. is hard to explain (It takes more time to move diagonally than horizontally, because you're moving further, this can get you killed if you aren't paying attention) and is harder to do (because how do you define 'adjacent' for melee ranges? It gets ugly and weird) and probably doesn't fit well in crawl (You'd pretty much end up with all new mechanics, since you'd have to impose some sort of range-based mechanic on diagonal melee, which would be weird and hard to balance)

So I guess I consider 3. to be an upgrade, but I consider it a *very small* upgrade to the "is more consistent" department. But I actually have fun, personally, manipulating the different ways you can handle diagonals, I realize it's inconsistent and weird, and there's various circumstances where being aware of the topology can result in optimal movement being strange and nonintuitive, but in some ways I like it that way.

If we do move to SquareLOS, I'll be a little sad, not because it makes the game objectively worse in any way (I think it makes the game objectively better), but because we'd be losing one of the weird quirky elements of the game, and I like weird quirky things.

/Offtopic/
Spoiler: show
So a long while ago I played this PS2 game called 'SSX' (it's a snowboarding racing game) and the first one was clunky, and a little awkward, but still a lot of fun.

The second one (SSX2 "Tricky") they cleaned up the mechanics a LOT it's much cleaner the interface works better, but there's still some clunk in there, different characters don't always perform quite as you'd expect in comparison, and different board types don't perform as well for different things, which means that you have to play *differently* depending on what you pick (And in some cases are limited in what you can do, one character might not rotate as much as another for example), there's also weird spots where you can kind of break the game here and there (you can break out of the map where they didn't intend you to).

With SSX3, they cleaned up the interface to the point where everything is crisp, all the characters behave exactly the same given the same stats, and all the boards perform tricks in the same way, if you go off a jump at a certain speed, you can be sure of doing just so much rotation with a given amount of stats and can land the same trick with any character, and there is absolutely no "breaking" it.

Naturally the 'smooth, clean, quirk-free' version is actually the least fun for me, I can intellectually recognize it as superior, but I just lose interest much more quickly.

This reminds me a little bit of that.
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!

For this message the author Siegurt has received thanks: 2
adozu, njvack

mps

Tomb Titivator

Posts: 886

Joined: Saturday, 3rd January 2015, 22:34

Post Wednesday, 29th April 2015, 20:31

Re: SquareLOS experimental branch on CBRO

Actually, the topology induced by either metric is the same, both on a grid (where the topology is discrete) and in the plane, where it's the standard product topology, but anyway...

duvessa wrote:
mps wrote:I think the objection to actual LoS being square on an aesthetic level is totally crazy, but the objection to square shaped spell effects is more reasonable -- for example, square tornadoes with monsters pushed around inside along square shaped paths is a genuinely weird idea.
Why is it genuinely weird in Chebyshev geometry? The Chebyshev equivalent of a Euclidean circle looks like a Euclidean square. That's really all there is to it. The Euclidean equivalent to Crawl's "circles" looks like a star (but with aliasing because of the grid). If your line of sight (or tornado) was a wobbly star in a game with Euclidean geometry, everyone would call it crazy. So why do Crawl, NetHack, and ToME get a free pass for doing the same thing?


I think multi-tile objects on the screen engage the visual imagination in a certain way that makes circles, even bad approximations to them, look right in certain contexts, among them cyclones, one of those classically circular phenomena. Magical effects should be allowed a certain artistic license. As a tactical matter, it's not obvious that square effects or round effects are better or worse. For example with round AOE effects you can do stuff like this:

  Code:
@**
****
****
 **


Maybe that's good, maybe it's not, but I don't think the right way to decide is geometric consistency. That said, I'm completely on-board with the arguments about range and LoS rules matching with distance in the sense of movement, of course.

(Also, I would point out to nvjack that current tornado already moves monsters in a way that pulls them closer then pushes them away! Maybe it's good, maybe it's wack, but that's the way it is.)
Dungeon Crawling Advice tl;dr: Protect ya neck.

For this message the author mps has received thanks:
Berder

Slime Squisher

Posts: 411

Joined: Saturday, 9th March 2013, 14:22

Post Wednesday, 29th April 2015, 20:36

Re: SquareLOS experimental branch on CBRO

why, tornado is a mess, isn't that what tornadoes are anyway?

Ziggurat Zagger

Posts: 8786

Joined: Sunday, 5th May 2013, 08:25

Post Wednesday, 29th April 2015, 20:41

Re: SquareLOS experimental branch on CBRO

mps wrote:(Also, I would point out to nvjack that current tornado already moves monsters in a way that pulls them closer then pushes them away! Maybe it's good, maybe it's wack, but that's the way it is.)
Yes, because it's subject to the exact inconsistency that squarelos is supposed to get rid of, like every other "circle" in the game.

Siegurt wrote:2. is hard to explain (It takes more time to move diagonally than horizontally, because you're moving further, this can get you killed if you aren't paying attention) and is harder to do (because how do you define 'adjacent' for melee ranges? It gets ugly and weird) and probably doesn't fit well in crawl (You'd pretty much end up with all new mechanics, since you'd have to impose some sort of range-based mechanic on diagonal melee, which would be weird and hard to balance)
2. either requires removing the grid from the game, or makes crawl attempt to use 3 different geometries at once instead of 1 or 2. I actually explained this more thoroughly in the thread I linked:
duvessa wrote:This doesn't actually fix any of the problems with Crawl geometry. Because Crawl takes place on a grid, you cannot have Euclidean geometry in Crawl. Period. Consistent geometries possible on a 2D grid include Chebyshev and Taxicab geometry. Crawl currently attempts Chebyshev in some places (movement, etc), and has a weird, terrible Fake Euclidean geometry in other places (LOS, ranges, explosions, etc)

Changing the diagonal movement cost to 1.4 solves absolutely nothing, and in fact actually makes the problem worse - instead of the mix of Chebyshev and Fake Euclidean that the game currently has, you'd have a mix of Chebyshev, Fake Euclidean, and Geometry Where Pi Is Equal To 3.1016. In addition, and more importantly for gameplay, Crawl actions take place on a "grid" just like Crawl geometry: performing five 14 aut actions is not equivalent to performing seven 10 aut actions, because monsters, the environment, etc. are able to act during those actions, while the player character is not. The result is the kind of disgustingly bizarre behaviour outlined in this post. I guarantee if this were implemented in Crawl I'd play either old versions or not at all until it got un-implemented, because tactical positioning - the entire reason I like Crawl - would become utterly nonsensical.

If you want to make Crawl geometry consistent, you have to either get rid of the grid or use a geometry that actually works on the grid. My preference would be to switch to pure Chebyshev; square LOS, square ranges, explosions, tornadoes, etc. Unfortunately, problems still arise because Crawl also tries to have continuous angles which are not really meaningful on a grid and result in aliasing; I gave a very short explanation of the problem with that here. You cannot eliminate angle aliasing on a grid as far as I know (many games such as NetHack attempt to hide this by only allowing you to shoot in 8 directions, but that just moves the problem from targeting to positioning). Here the only option to eliminate it is to get rid of the grid entirely, which is an obviously impractical change for Crawl. You could greatly reduce aliasing by switching from a square grid to a hexagon grid, but that's still a huge change to the game. So I support just switching to Chebyshev geometry, which fixes the inconsistencies even if it does leave the aliasing.

Tartarus Sorceror

Posts: 1774

Joined: Tuesday, 23rd December 2014, 23:39

Post Wednesday, 29th April 2015, 20:49

Re: SquareLOS experimental branch on CBRO

Changing the shape of your LOS does not actually do anything to the "geometry." circleLOS is still using Chebyshev geometry if you're measuring distance by movement. It's just a shape overlaid on the grid. The LOS could be a triangle and it would still be Chebyshev geometry. There's no law saying every shape in a given geometry must be a ball in that geometry.
streaks: 5 fifteen rune octopodes. 15 diverse chars. 13 random chars. 24 NaWn^gozag.
251 total wins Berder hyperborean + misc
83/108 recent wins (76%)
guides: safe tactics value of ac/ev/sh forum toxicity

Tartarus Sorceror

Posts: 1739

Joined: Tuesday, 13th March 2012, 02:48

Post Wednesday, 29th April 2015, 21:20

Re: SquareLOS experimental branch on CBRO

Image

Tartarus Sorceror

Posts: 1888

Joined: Saturday, 9th July 2011, 20:57

Post Wednesday, 29th April 2015, 23:44

Re: SquareLOS experimental branch on CBRO

I love being able to Magic Dart things from seven squares away on the diagonal. It feels almost decadent.

For this message the author nicolae has received thanks:
Arrhythmia

Ziggurat Zagger

Posts: 8786

Joined: Sunday, 5th May 2013, 08:25

Post Thursday, 30th April 2015, 00:14

Re: SquareLOS experimental branch on CBRO

fun fact: in old versions you could crossbow things from eight squares away on the diagonal. didn't matter that you couldn't see them, you could crossbow them

Berder wrote:Changing the shape of your LOS does not actually do anything to the "geometry." circleLOS is still using Chebyshev geometry if you're measuring distance by movement. It's just a shape overlaid on the grid. The LOS could be a triangle and it would still be Chebyshev geometry. There's no law saying every shape in a given geometry must be a ball in that geometry.
I mean sure, if we're going to be pedantic, but generally players would not expect how far you can see to depend on which direction you're looking, nor expect an explosion to be star shaped, nor expect a crossbow bolt to travel different distances depending on the direction you shoot it, etc. So most people intuitively conclude that the circular LOS/ranges/etc is an attempt at Euclidean simulation that fails because there is a grid and because movement/etc is Chebyshev, rather than concluding that circular LOS/ranges/etc. are in the game just for the hell of it. Even I'm not THAT uncharitable with Crawl's developers.

For this message the author duvessa has received thanks: 3
Arrhythmia, Lasty, Sar

Tartarus Sorceror

Posts: 1774

Joined: Tuesday, 23rd December 2014, 23:39

Post Thursday, 30th April 2015, 01:11

Re: SquareLOS experimental branch on CBRO

duvessa wrote:I mean sure, if we're going to be pedantic, but generally players would not expect how far you can see to depend on which direction you're looking, nor expect an explosion to be star shaped, nor expect a crossbow bolt to travel different distances depending on the direction you shoot it, etc.


If it were really confusing and against expectations, there would be tavern posts from bona fide new players asking about it. Do you see any? I would characterize squareLOS as coming from very experienced players and devs who have a background in math and want crawl to conform to some math ideas that aren't necessarily gameplay improvements.

SquareLOS has its own quirks. Generally players would not expect to reveal twice as much territory moving on a diagonal compared to orthogonally. Nor would they expect an explosion or a tornado to be square shaped.
streaks: 5 fifteen rune octopodes. 15 diverse chars. 13 random chars. 24 NaWn^gozag.
251 total wins Berder hyperborean + misc
83/108 recent wins (76%)
guides: safe tactics value of ac/ev/sh forum toxicity

Ziggurat Zagger

Posts: 8786

Joined: Sunday, 5th May 2013, 08:25

Post Thursday, 30th April 2015, 01:33

Re: SquareLOS experimental branch on CBRO

Berder wrote:If it were really confusing and against expectations, there would be tavern posts from bona fide new players asking about it. Do you see any?
https://crawl.develz.org/tavern/viewtopic.php?f=8&t=4979

For this message the author duvessa has received thanks:
Arrhythmia

Ziggurat Zagger

Posts: 6454

Joined: Tuesday, 30th October 2012, 19:06

Post Thursday, 30th April 2015, 01:43

Re: SquareLOS experimental branch on CBRO

Hm, maybe I should start compiling my crawl with EXTERNAL_FLAGS=-DEUCLIDEAN (Per a comment a few down from the linked post)
Spoiler: show
This high quality signature has been hidden for your protection. To unlock it's secret, send 3 easy payments of $9.99 to me, by way of your nearest theta band or ley line. Complete your transmission by midnight tonight for a special free gift!

Tartarus Sorceror

Posts: 1774

Joined: Tuesday, 23rd December 2014, 23:39

Post Thursday, 30th April 2015, 01:45

Re: SquareLOS experimental branch on CBRO

duvessa wrote:
Berder wrote:If it were really confusing and against expectations, there would be tavern posts from bona fide new players asking about it. Do you see any?
https://crawl.develz.org/tavern/viewtopic.php?f=8&t=4979

Fair enough. Still, very few new players ask about it.
streaks: 5 fifteen rune octopodes. 15 diverse chars. 13 random chars. 24 NaWn^gozag.
251 total wins Berder hyperborean + misc
83/108 recent wins (76%)
guides: safe tactics value of ac/ev/sh forum toxicity

Tartarus Sorceror

Posts: 1888

Joined: Saturday, 9th July 2011, 20:57

Post Thursday, 30th April 2015, 03:01

Re: SquareLOS experimental branch on CBRO

Berder wrote:some math ideas that aren't necessarily gameplay improvements.


I'm not sure I agree with you 100% on your game design work there, Lou.

nicolae wrote:I love being able to Magic Dart things from seven squares away on the diagonal. It feels almost decadent.


SquareLOS has its own quirks. Generally players would not expect to reveal twice as much territory moving on a diagonal compared to orthogonally. Nor would they expect an explosion or a tornado to be square shaped.


While I haven't taken a poll, I suspect that "why the hell can't my ranged attacks go as far diagonally as orthogonally" was a bigger issue for new players than "how come tornados are squares" is going to be.

For this message the author nicolae has received thanks: 2
Arrhythmia, duvessa

dck

Vestibule Violator

Posts: 1653

Joined: Tuesday, 30th July 2013, 11:29

Post Thursday, 30th April 2015, 05:55

Re: SquareLOS experimental branch on CBRO

I believe the status quo lends itself better to creating interesting situations given how crawl works, and is the more desirable choice over squarelos.
To go into more detail than "no it's great really" I would like to point out the commit's mention of repositioning prior to engaging in combat with a sleeping monster. Whenever this example is brought up (and it is a fair example as it addresses the weakest point of circlelos) it is always enunciated as a very transcendent part of the game that is being made worse by circlelos, I would like to differ on that regard.

First of all since the game does not take place in a vacuum most if not all situations in which a sleeping monster is spotted will not be immediately followed by engaging said monster, certainly not if the circumstances actually make the encounter in any way relevant. At the very least there will be repositioning to see what else is in the neighborhood of the monster in question and to gain information on the terrain surrounding it; this is assuming that the player came to notice the monster from the perfect angle to engage it from, relative to his escape or fight routes that are explored.
Now let us really consider all of that because it is a lot, and take into account that it all has to happen for sqlos to be an actual improvement over what we currently have in this alledgedly very important regard. To shorten that somewhat, no proper gameplay situations that aren't of planetary alignment rarity will see themselves affected at all by this. This is because the intended impact of the change is that the player no longer has to reposition when spotting a sleeping monster in situations where attacking is the chosen route; but there are tens of features that go vehemently against that very notion, such as the ability to wake monsters up while they are out of your LoS, wandering monsters that may be on the level or why not levelgen itself not being particularly (thankfully) fond of huge empty spaces everywhere.
And not only this, but this is actually the most naïve approach at an actual gameplay example of this change's impact possible. This assumes that the player will at least be in neutral positioning with regards of his explored terrain when spotting the monster, and certainly not in an openly negative one such as having say 3/4s of terrain outside of LoS actually unexplored. The problem with this is that unlike the vanishingly rare situation above mentioned, this circumstance arises all the time in games that take place in reality due to autoexplore existing and not being perfect.

And this brings me to my next point which is autoexplore. This exists and impacts the way the player encounters monsters greatly; as I just said it is not perfect, nor do I in any way expect it to. First because it would be literally impossible to match manual movement because the player can know things about how even non-fixed levels generate. Second because it would have to account for monsters that are known to be wandering, as well as stairs. Of course nobody seeks to develop an autoexplore that trumps or goes anywhere near the possibilities of manual movement because it is easy to see that the returns for the effort would be meager at best.
And yet we accept autoexplore to be part of the game and to be one of the most pressed buttons in every game played, even though it ruins the player's approach to potential monsters almost every time and is a cause for constant repositioning when the monster does not become aware of the player instantly.
Could this be because we realize crawl is not so brutal that one improper move will shatter all chances of winning? Should we not also realize that crawl has so much is going on that having sqlos allow to engage without repositioning in the rare chance autoexploring into a monster goes perfectly and having the monster stay asleep changes next to nothing of the actual game?

All that said, I have yet to address the elephant in the room that is ranged combat. I frankly do believe squarelos is a bad thing for ranged combat, because it makes it intrinsically stronger and crawl's quality as a game is inversely proportional to how relevant the ability to damage enemies at range is.

All core components of crawl rely on positioning and LoS manipulation as a cornerstone, ranged combat is in itself heavily detrimental to both of those things as it removes weight from errors the player makes in how to deal with threats and is just so much more forgiving in terms of misjudging one's ability to shut down a given source of problems when compared to melee. To the point ranged combat actually allows for the player to bruteforce a fair amount of problems by resetting fights and trying again until the damage rolls are in his favor with such ease it is a perfectly valid option to use in certain situations, provided the generated architecture does not strictly go against that approach.

Now, damaging enemies at a distance is of course acceptable as a feature due to flavor and variety and general fun, but I strongly believe it should be given much less importance than the pivotal part that melee combat is -I would even go further and say that if one could encourage ranged users to move around more and reveal more tiles with potential enemies even through artificial means such as circlelos it would be beneficial to the game, as there is a chance the new monster in conjunction with noise mechanics for example can create a challenging situation whereas none would have existed before. I understand circlelos can be confusing at some very basic levels regarding ranged combat, but I've always acknowledged the peculiarities as simply rules of the game, not unlike dying when under 1 hp any other basic mechanic.

Regarding taking ranged damage from enemies, well that's a different beast entirely. They are by default more interesting than melee enemies as the only conditions they must fulfill to be relevant is to be present in the screen, sometimes even ignoring line of fire blocking altogether. And yet with enemies that are so unbelievably easy to make prevalent and to have overshadowing other monsters that lack ranged abilities, shouldn't there be some mechanic that actually allows one to tempt the player to actually engage the monster in a situation where less relevant friends are actually capable of joining in and perhaps actually making a difference?
Because that is down to the letter what being able to approach monsters at a diagonal in circlelos achieves. It has been downright stated in this thread that "Crawl should not have magic directions that it is safer to approach enemies from". Well I don't think that is necessarily true; what I believe is that the point of monsters is to present a puzzle to the player, and monsters can generally only do so when several of different types are present and environment is welcoming to it. Crawl does generate quite a few interesting situations on its own but they are few and far inbetween compared to what the player himself can get into. In my perspective circlelos' diagonal walking is a great double edged tool that actually offers an interesting benefit for the dangers it presents.

To exemplify, I think walking at a diagonal into LoS of a sleeping unique as a stabby player and having to compare the possibilities of taking advantage of that first step already taken towards the stab with the dangers that diagonal walking presents (revealing a ton of tiles and the unique waking up while being in a bad position) is very interesting. Of course walking into monsters' LoS will still happen under sqlos, it will just not present those sort of situations without actual environmental help and that type of engagement method with all of its consequences will be for the most part gone.
Not only this, in fact, but stabbing as a whole will be a lot less interesting without circlelos allowing for diagonal shenanigans, because sqlos' diagonal moving makes the dangers of approaching at an angle several orders of magnitude greater and one cannot give any sort of stealth benefit to make up for the sheer amount of terrain being revealed or he'd replicating the perceived problems circlelos with stabbing at a different scale!

To close, I do think that at a fundamental level the reasoning for the change is not acceptable because it assumes having a feature as important as LoS with inconsistencies such as the ones circlelos presents has actually led crawl to have worse gameplay than it would have under squarelos. For the reasons above stated I don't find that to be true.
I also believe Siegurt's parallel to the evolution of the SSX series does illustrate the reasoning behind flawed basic premises being able to create more organic and distinct gameplay than the more clinical approach, superior as it should be on paper or more aseptically elegant as it may be in theory.

While I won't chime in about the aesthetic advantages or disadvantages of the current system, common sense assures there will be quite a few more people wondering why their tornado is a square than the linked '12 post regarding circlelos' quirks.

For this message the author dck has received thanks: 2
and into, Sar
User avatar

Pandemonium Purger

Posts: 1283

Joined: Thursday, 16th April 2015, 22:39

Post Thursday, 30th April 2015, 06:04

Re: SquareLOS experimental branch on CBRO

I suggest a compromise: LOS is a square with radius 7 that has 3 tiles clipped off each corner. Targeting and AOE spells may or may not be Chebyshev. :idea:
  Code:
XX+++++++++++XX
X+++++++++++++X
+++++++++++++++
+++++++++++++++
+++++++++++++++
+++++++++++++++
+++++++++++++++
+++++++@+++++++
+++++++++++++++
+++++++++++++++
+++++++++++++++
+++++++++++++++
+++++++++++++++
X+++++++++++++X
XX+++++++++++XX
Next

Return to Game Design Discussion

Who is online

Users browsing this forum: No registered users and 26 guests

cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by ST Software for PTF.