My final thoughts on WJC, and moving on.


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

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Saturday, 18th March 2017, 10:54

My final thoughts on WJC, and moving on.

I suppose that, given the title of this post, you may have been expecting some Tavern drama. Fear not, I'm not upset or annoyed at the dev team at all, on the contrary, working on the god has been a very fun experience that has taught me a lot. My reasons to withdraw from working on (and discussing) WJC are mostly pragmatic and in favor of focusing more on other projects.

Yesterday, a couple changes were made to the god that I have strongly argued against. Whether this makes the god altogether better or worse, I have to admit I'm not experienced enough to say. What I can do is point out at the end of this post the potential issues I perceive with those changes (since I've tried variations of these in the past) and hope that, in case these predictions come true, they can be solved somewhat scientifically.

My real reasons to withdraw are that I've found working on a feature like this to be extremely time consuming. Those who saw the commit count of my initial PR will tell you that I've easily put between one and two hundred hours in tweaking and redesigning the god since its initial inception. Since the god made it to trunk, that time has transformed in time spent discussing it, which is exhausting in a different way. While I don't mind any of this, my wake up call came yesterday when I realized that changes had been made without me realizing at all (I was checking !messages in ##crawl-dev every day since I live on a different timezone, and nobody sent me a ping before this patch). Again, I'm not trying to express anger at the devs for not "respecting my design", which would be a shit attitude to have, but realizing that no matter how much time I spend, and how strongly I argue, I don't have control over the idea anymore, and on the flipside no responsibility for watching over it with the same degree of involvement. If I had been asked before these changes were made, maybe I would've felt compelled to remain in the conversation, but at the moment I don't feel like I can keep up, especially when I'm asleep through most of the live discussion. I wish I could, because it's a lot of fun to discuss the design with the devs and to gather feedback from Webtiles!

I've always admitted that I'm not a very experienced crawl player. The only reason why I could be of value in these conversations lately is because I have a disproportionate amount of experience with WJC in particular, for obvious reasons. This means that even though my mechanical knowledge is lacking, I have seen dozens of variations of the god mechanics fail in funny (and enlightening ways). However, as the design moves away from the original, my advantage is going to disappear, since I'm going to be as inexperienced with the new mechanics as everyone else, and my input will become less valuable. This is another reason why I'm going to let go of it.

So as a last point, I'm going to at least leave my prediction of what issues the new mechanics are going to cause, and hopefully some pointers that will help fix them in the future.

With the current changes to the way martial attacks work, the dependency between the three has been broken. Before, the efficiency of lunge was a consequence of having used the other two. Now there's no such dependency, so the optimal way to play the god will degenerate to retreating until lunge is available, or failing that, constant swapping with a summon to apply the damage bonus from lunge repeteadly. The god will now be optimal in corridors. I have seen this happen in two or three iterations of the god. The natural consequence will be to nerf the Lunge damage bonus, but this will not help. Without a dependency between the moves, for any given game lunge will either be optimal, and degenerate in the ways I explained, or not be worth it at all, in which case players will completely forgo lunge and instead wait and tab, which is tactically superior. I believe it is essential for lunge to depend on the other two moves and I've expressed it multiple times, so I'm planting this flag early so that, in case these issues manifest (and I hope I'm wrong) this can serve as a starting point.

Without the defensive use of wall jump, it is not justifiable to use any of the two AoE attacks (whirlwind and wall jump) against multiple enemies except by high mitigation characters. Low HP/AC characters will have no incentive to attack in AoE without a sufficiently strong escape tool. This is going to compound the previous issue making worshippers stick to corridors again. This restriction to wall jump is going to make the god significantly less fun and I'm certain we're going to see that reflected in the immediate feedback. Just a sample from a SA post:
Wu Jian is definitely the most fun god in the entire game, holy shit. It was looking for a while like 0.20 might be mostly a sidegrade compared to 0.19 but now I can kite hydras with a polearm by kickflipping off walls

I'm not being controversial if I say these posts will stop coming up, because they are mostly due to defensive wall jump. How much that weights compared to game balance is not my decision to make.

By making wall jump unusable as an escape tool, Serpent's Lash is hurt indirectly, since most creative uses of it go out of the window. A few players were starting to pick up on the advanced ways to use it as an escape, which often involved moving towards enemies first. With these changes, Serpent's Lash will strictly be saved to walk away, and no good player will ever use it together with martial attacks. With wall jump gone, Serpent's Lash is best saved as a way to gain those extra two tiles towards the stairs. Altogether, these new changes will make the player use the kit less as the danger increases.

In case these predictions manifest, here's how I think the god should have evolved instead: The damaging parts of whirlwind and wall jump would be reduced (if necessary, to the extreme of taking whirlwind damage down to 70% and removing the damage part of wall jump altogether). The AUT expenditure on wall jump would be increased, and a conduct added to reduce the efficiency of high mitigation, either through noise associated to encumbrance level, or a drop in AC and SH. This would reduce tedium by making tab viable for easy fights while augmenting the connection between the moves, forcing the player to be creative with the kit to survive.

Now, you can debate any of the points I brought up, but I won't reply. This isn't meant to start a discussion, I already did with the "Balancing the Council God" thread. This is just a summary of my thoughts, on what I personally believe will happen, and only time and player feedback will be able to prove that. I know (and I'm not saying it ironically) that I'm leaving the god's future in good hands, but in case the problems I discussed appear, I hope someone in the dev team will remember this post and consider that I am saying this not out of a desire to protect "my baby", because I've never been like that, but because I have direct experience in seeing variations of this new design fail.

I hope I'm wrong, and in any case enjoy WJC! :)

For this message the author Steel Neuron has received thanks: 6
bananaken, Brannock, Heia, NhorianScum, Wolfechu, Xenobreeder

Dungeon Master

Posts: 388

Joined: Monday, 18th August 2014, 20:04

Post Saturday, 18th March 2017, 13:09

Re: My final thoughts on WJC, and moving on.

Steel Neuron wrote:Yesterday, a couple changes were made to the god that I have strongly argued against.
...
While I don't mind any of this, my wake up call came yesterday when I realized that changes had been made without me realizing at all (I was checking !messages in ##crawl-dev every day since I live on a different timezone, and nobody sent me a ping before this patch). Again, I'm not trying to express anger at the devs for not "respecting my design", which would be a shit attitude to have, but realizing that no matter how much time I spend, and how strongly I argue, I don't have control over the idea anymore, and on the flipside no responsibility for watching over it with the same degree of involvement. If I had been asked before these changes were made, maybe I would've felt compelled to remain in the conversation, but at the moment I don't feel like I can keep up, especially when I'm asleep through most of the live discussion.
...
In case these predictions manifest, here's how I think the god should have evolved instead...


I don't get it, if this is how you think the god should have evolved, why didn't you submit a PR? It seems really strange to be saying that you have no control over the idea after the initial PR got merged beyond just talking to devs. The one IJC PR that I spot that you submitted after the initial merge (granted, bugfixing) was merged within 12 hrs.

FWIW no one on the dev team needs to ping anyone, dev team or otherwise, when they merge something.

For this message the author advil has received thanks:
duvessa

Vaults Vanquisher

Posts: 443

Joined: Thursday, 16th February 2017, 15:23

Post Saturday, 18th March 2017, 14:18

Re: My final thoughts on WJC, and moving on.

My take: Crawl development is opening up in that there is more interest and support for alternatives to dcss. People like you who have the energy and interest to do new and creative things should think seriously about going outside the dcss process. It's true that the devteam has a sort of groundedness individual developers sometimes lack (see Circus Animals), but you can get that from other sources if you look for it.

The truth is that you can change crawl a lot with fairly small amounts of coding and that the real obstacle to fast resolutions to many standing problems or just substantially new directions in content or balance is this process of discussion with a large number of interested parties. If you have a vision and a few taste testers, you can go pretty far pretty fast.

If you or others decide to go that way, it sounds like gnollcrawl.tk might be a good place to incubate new forks with an existing audience.
*Lana Del Rey voice* , video games...

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Saturday, 18th March 2017, 14:20

Re: My final thoughts on WJC, and moving on.

advil wrote:
I don't get it, if this is how you think the god should have evolved, why didn't you submit a PR? It seems really strange to be saying that you have no control over the idea after the initial PR got merged beyond just talking to devs. The one IJC PR that I spot that you submitted after the initial merge (granted, bugfixing) was merged within 12 hrs.


I thought we were still discussing / gathering data before making balance changes. I had proposed those changes informally, and I was told it was still early, and we needed to have more evidence before balance changes were needed. Obviously we got to that phase already, I just wasn't aware of it.

advil wrote:FWIW no one on the dev team needs to ping anyone, dev team or otherwise, when they merge something.


That's completely understandable. Really, I'm not complaining. I'm just manifesting that I won't be able to keep up (because just paying a bit less attention to the IRC logs for a week made me completely miss that a change was happening) and summarizing my thoughts on the change in one place, since I won't have time to be actively involved in the conversation anymore.

I notice it's going to be hard to convince people that this post is harboring no ill will against anyone :). I dislike these changes, but I understand where they are coming from and why I can't be a direct influence on where the god goes from now on. I'm just trying to be scientific and making concrete predictions about the issues WJC is going to show in the following weeks, with the hope of my position gaining credibility down the line when some dev considers further changes.

Ziggurat Zagger

Posts: 8786

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

Post Saturday, 18th March 2017, 21:15

Re: My final thoughts on WJC, and moving on.

Steel Neuron wrote:This restriction to wall jump is going to make the god significantly less fun and I'm certain we're going to see that reflected in the immediate feedback. Just a sample from a SA post:
Wu Jian is definitely the most fun god in the entire game, holy shit. It was looking for a while like 0.20 might be mostly a sidegrade compared to 0.19 but now I can kite hydras with a polearm by kickflipping off walls

I'm not being controversial if I say these posts will stop coming up, because they are mostly due to defensive wall jump.
My posts may have slightly fewer backhanded insults against the devs than this goon's posts, but I for one find the restriction to wall jump makes the god significantly more fun

Temple Termagant

Posts: 5

Joined: Friday, 5th February 2016, 05:15

Post Sunday, 19th March 2017, 10:41

Re: My final thoughts on WJC, and moving on.

For me personally this change has made the god completely useless.
Before it provided a viable alternative to tabbing, replacing that with something a bit more interesting. Now that you've got to tab with this god too? Naaah, I'll pass.

Slime Squisher

Posts: 352

Joined: Monday, 14th December 2015, 00:43

Post Sunday, 19th March 2017, 11:36

Re: My final thoughts on WJC, and moving on.

Xenobreeder wrote:For me personally this change has made the god completely useless.
Before it provided a viable alternative to tabbing, replacing that with something a bit more interesting. Now that you've got to tab with this god too? Naaah, I'll pass.


I generally agree with the sentiment, except I wouldn't call it useless, instead I'd say it's too much micromanagement for too little reward. Sure the cleaveslow and endless escapes were very strong, but it was a reward for not using the be-all and end-all Tab key. Now I just don't want to bother anymore.

Abyss Ambulator

Posts: 1217

Joined: Sunday, 14th April 2013, 04:01

Post Thursday, 23rd March 2017, 04:42

Re: My final thoughts on WJC, and moving on.

On topic: thank you so much for all your work on this god, and for providing me with literally the most fun thing I've experienced in crawl since Ru's release.

Semi-off topic: Yeah, I'm done beelining this god, the walljump changes absolutely ruined it for me. Make it exhaust, make it use MP, make every other ability nerfed into the floor, I don't care, but all of the fun is gone and replaced with constant 'you can't do that' messages every time I accidentally bump into a wall.
Three wins: Gargoyle Earth Elementalist of Ash, Ogre Fighter of Ru, Deep Dwarf Fighter of Makhleb (0.16 bugbuild :( )

Lair Larrikin

Posts: 18

Joined: Friday, 3rd March 2017, 15:24

Post Thursday, 23rd March 2017, 15:07

Re: My final thoughts on WJC, and moving on.

TeshiAlair wrote:On topic: thank you so much for all your work on this god, and for providing me with literally the most fun thing I've experienced in crawl since Ru's release.

Semi-off topic: Yeah, I'm done beelining this god, the walljump changes absolutely ruined it for me. Make it exhaust, make it use MP, make every other ability nerfed into the floor, I don't care, but all of the fun is gone and replaced with constant 'you can't do that' messages every time I accidentally bump into a wall.


Second everything you just said. Lost all interest in the god with the wall jump change.

Mines Malingerer

Posts: 36

Joined: Thursday, 6th October 2016, 15:14

Post Friday, 24th March 2017, 15:42

Re: My final thoughts on WJC, and moving on.

Thirded, highly disappointed in the changes, whoever made them clearly did not grok what WJC was about. There is a difference between balancing and perverting the purpose of a proposed god.
User avatar

Shoals Surfer

Posts: 318

Joined: Friday, 6th July 2012, 10:16

Post Tuesday, 28th March 2017, 06:31

Re: My final thoughts on WJC, and moving on.

Yeah, I was kind of shocked to read that duvessa post about the interactions between martial moves being bad. It seems crawl really has become the button-mashing arpg it's detractors have accused it of being.
"No one should have two lives / Now you know my middle names are wrong and right / But baby there's no guidance / When random rules"
User avatar

Tartarus Sorceror

Posts: 1762

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

Post Tuesday, 28th March 2017, 07:01

Re: My final thoughts on WJC, and moving on.

"become"? wouldn't that mean it always has been, since WJC is really recent anyways? also who calls a turn based game a button masher
User avatar

Tartarus Sorceror

Posts: 1762

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

Post Tuesday, 28th March 2017, 07:05

Re: My final thoughts on WJC, and moving on.

I mean, speaking as the dude who beat this game in 37 minutes by picking FoFi, going oka, finding a vampiric axe and slamming my face into the keyboard, I don't really see how the old version of WJC would make it less "button mash"-y when pre-nerf WJC was a god you mashed left and right to talk back and forth in front of pan lords until they died. it seems to me that nerfing whirlwind is a change in, if anything, the complete opposite direction of "button mashing" because it means you don't want to go adadadadadada to use whirlwind all the damn time, instead making it into a situational tactical tool

For this message the author Shard1697 has received thanks: 2
Lasty, nago

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Tuesday, 28th March 2017, 07:11

Re: My final thoughts on WJC, and moving on.

Shard1697 wrote:I mean, speaking as the dude who beat this game in 37 minutes by picking FoFi, going oka, finding a vampiric axe and slamming my face into the keyboard, I don't really see how the old version of WJC would make it less "button mash"-y when pre-nerf WJC was a god you mashed left and right to talk back and forth in front of pan lords until they died. it seems to me that nerfing whirlwind is a change in, if anything, the complete opposite direction of "button mashing" because it means you don't want to go adadadadadada to use whirlwind all the damn time, instead making it into a situational tactical tool


It's a bit annoying when people keep comparing the current version of WJC with the pre-nerf version. The old version was broken too (although I'd argue more fun than the current). I don't know about the others but I'm personally not interested in restoring the old version of WJC as is, but rather taking it in the direction that I am convinced will be better than the two we've experienced.

I have written again and again that my goal was to combat "adadadad" spam, and explained in the CYC thread that the current version not only doesn't fix that, but further incentivizes spamming a single martial attack by shifting the focus towards damage. I don't know how much WJC you have played since the nerf, but I doubt you're finding incentives to mix it up while fighting.

The way to achieve that is to reduce damage on martial attacks across the board and focus on their support effects.

Dungeon Master

Posts: 3160

Joined: Sunday, 5th August 2012, 14:52

Post Tuesday, 28th March 2017, 07:31

Re: My final thoughts on WJC, and moving on.

Crawl is a game where combat depth is designed to come from questions about when and where to fight and which consumables (where consumables means any somewhat-limited-use effects) to use when. Making the process of attacking monster slower and/or more complex once a player has decided that now is the time and place to attack is just a tax on the player's patience; if their patience runs out they can always default back to just attacking normally, because that works. Trying to fight this by adding bonuses to complex-attack modes is just bribing the player to play your minigame. Imagine crawl's maps were on an explicitly-numbered grid and you got combat bonuses for attacking monsters on a sequence of tiles that had some mathematical significance; it would require more thought, and it might not be trivial to achieve in practice, but it wouldn't actually create good gameplay.

If crawl is a "button masher", it's because it's too easy. The fact that tab is effective isn't inherently problematic, because it's a command that means "attack". It's okay for attacking monsters to be a single, simple button press in a game about attacking monsters. Making combat more interesting in this view would mean making the choices about when and where to fight more difficult. You can see attempts at this in the dynamic monsters branch, where we attempted to make choosing your time to fight a more interesting decision. Adding an option that causes attacking monsters to require more button presses and/or more thinking between button presses doesn't add depth or interesting choices to the game, it just makes combat slower.

The thing that works really well about WJC is that it transforms your decisions about when and where to attack, and it gives you new ways to attack while moving, a more complex form of attack. This more complex form of attack is acceptable in my mind when it is something you only want/need to use when it will actually help your positioning. I want to preserve and amplify that aspect by removing the bribes that incentivize players to exclusively use these more complex forms of attack over the normal forms of attack.

I don't want players to feel that they must play a slow minigame during each fight in order to get access to more powerful attacks. To me, that approach is like asking the player to type in a captcha validation during each fight to unlock special bonuses: it adds no complexity or depth, but does tax the player's patience.

For this message the author Lasty has received thanks: 5
Brannock, duvessa, luckless, Shard1697, stickyfingers

Dungeon Master

Posts: 3160

Joined: Sunday, 5th August 2012, 14:52

Post Tuesday, 28th March 2017, 07:35

Re: My final thoughts on WJC, and moving on.

Steel Neuron wrote:I don't know about the others but I'm personally not interested in restoring the old version of WJC as is, but rather taking it in the direction that I am convinced will be better than the two we've experienced.

I have written again and again that my goal was to combat "adadadad" spam, and explained in the CYC thread that the current version not only doesn't fix that, but further incentivizes spamming a single martial attack by shifting the focus towards damage. I don't know how much WJC you have played since the nerf, but I doubt you're finding incentives to mix it up while fighting.

The way to achieve that is to reduce damage on martial attacks across the board and focus on their support effects.

If you want to pursue this model and don't feel you can convince the dev team to try it in trunk, I would suggest that you reach out to one of the active forks as a way to try it out. I mean this as a serious and not dismissive idea -- one of the exciting things about having so many active trunks is that we can try more ideas in parallel, even ideas that one group sees as incompatible with their goals for their project. I feel strongly that you should pursue your direction and determine whether it ends up creating better gameplay. As I've made clear before, I don't believe that it will, but if you give it a try one of us will learn something, and the worst-case scenario from your perspective is that it will be you. Learning something new is always a good outcome, and if it turns out that I'm way off base here and that your design is the better one, I will be happy that I learned something new.

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Tuesday, 28th March 2017, 07:54

Re: My final thoughts on WJC, and moving on.

Lasty wrote:If you want to pursue this model and don't feel you can convince the dev team to try it in trunk, I would suggest that you reach out to one of the active forks as a way to try it out. I mean this as a serious and not dismissive idea -- one of the exciting things about having so many active trunks is that we can try more ideas in parallel, even ideas that one group sees as incompatible with their goals for their project. I feel strongly that you should pursue your direction and determine whether it ends up creating better gameplay. As I've made clear before, I don't believe that it will, but if you give it a try one of us will learn something, and the worst-case scenario from your perspective is that it will be you. Learning something new is always a good outcome, and if it turns out that I'm way off base here and that your design is the better one, I will be happy that I learned something new.


Oh, I play the branch I'm describing routinely and it works much better than the current trunk and the former version. I can clean it up and PR it, or even nag johnstein to open an experimental, but what purpose would it serve? Again my problem is the lack of metric. At what point would it be considered a success? Would you or any other devs play a full game with it?

Even if you did, I don't care about "teaching anyone something new" honestly, I'm not in this to be told I was right. I want to see a design that works well in the game, so if opening a branch could be conducive for that goal I will do it, but if it's just about winning an argument I won't.
Last edited by Steel Neuron on Tuesday, 28th March 2017, 08:08, edited 1 time in total.

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Tuesday, 28th March 2017, 07:58

Re: My final thoughts on WJC, and moving on.

Lasty wrote:I want to preserve and amplify that aspect by removing the bribes that incentivize players to exclusively use these more complex forms of attack over the normal forms of attack.

I don't want players to feel that they must play a slow minigame during each fight in order to get access to more powerful attacks. To me, that approach is like asking the player to type in a captcha validation during each fight to unlock special bonuses: it adds no complexity or depth, but does tax the player's patience.


For the record Lasty we want exactly the same thing. I would ask you to play a game under the current Trunk version and tell me honestly whether the god achieves this purpose as is, without comparing it to the former revision.

Dungeon Master

Posts: 3160

Joined: Sunday, 5th August 2012, 14:52

Post Tuesday, 28th March 2017, 09:27

Re: My final thoughts on WJC, and moving on.

Steel Neuron wrote:Oh, I play the branch I'm describing routinely and it works much better than the current trunk and the former version. I can clean it up and PR it, or even nag johnstein to open an experimental, but what purpose would it serve? Again my problem is the lack of metric. At what point would it be considered a success? Would you or any other devs play a full game with it?

Even if you did, I don't care about "teaching anyone something new" honestly, I'm not in this to be told I was right. I want to see a design that works well in the game, so if opening a branch could be conducive for that goal I will do it, but if it's just about winning an argument I won't.

At what point would _you_ consider it a success? One measure of success is whether you're pleased with what you've created, and you clearly are. Another is whether your creation has pleased other people, and even the versions you found imperfect did that. Another is whether your work can reach a large audience, and it can right now; my suggestion above was a way to give your preferred version of your work a larger audience. However, it sounds like to you the project is only a success if it gets into DCSS trunk in the version that you prefer, and that seems tough, since some of your goals for the project don't seem to line up well with some DCSS devs' goals.

As for what counts as success within DCSS trunk, you're right that there's no one fixed metric. I have every intention of playing full games with the god -- I already played a full game with a version I didn't care for -- but I have been busy enough that it hasn't happened yet. I played a bunch of games with Pakellas, even though I ultimately felt that the god was fundamentally a problematic design. Each developer no doubt has a stack of explicit and implicit goals that they're applying, and all of them have the freedom to apply those goals as they see fit. You may come up with the perfect design for one developer and still run into problems with another. On top of that, as I've said before, nothing ever gets to stand as permanently "a success" -- everything gets evaluated repeatedly over time, both by existing developers and by new developers who haven't yet joined the project. Was Ely a success? Was Ely's rework a success? Can Nemelex ever be a success? Were death knights a success, and if so, why did I remove them? This is all fundamentally in the realm of the subjective.

You say you're not interested in teaching anything and you're not interested in winning an argument. You don't mention whether you're willing to learn anything, which is what I actually suggested. If you're neither interested in teaching nor learning then you may be a bad fit for contributing to DCSS.

Edit:
Steel Neuron wrote:
Lasty wrote:I want to preserve and amplify that aspect by removing the bribes that incentivize players to exclusively use these more complex forms of attack over the normal forms of attack.

I don't want players to feel that they must play a slow minigame during each fight in order to get access to more powerful attacks. To me, that approach is like asking the player to type in a captcha validation during each fight to unlock special bonuses: it adds no complexity or depth, but does tax the player's patience.


For the record Lasty we want exactly the same thing. I would ask you to play a game under the current Trunk version and tell me honestly whether the god achieves this purpose as is, without comparing it to the former revision.


I explicitly _do not want_ players to have an incentive to use martial attacks in every fight, or even in the large majority of fights, which I think differs from your ideal situation. I also do not want players to feel that once they've used one martial attack they need to continue using martial attacks. I would prefer that Tab stay the default method of launching a melee attack even when playing with a god that offers martial attacks.

That said, I will happily play WJC under the current trunk version as soon as I get an opportunity.

For this message the author Lasty has received thanks:
duvessa

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Tuesday, 28th March 2017, 10:00

Re: My final thoughts on WJC, and moving on.

Lasty wrote:You say you're not interested in teaching anything and you're not interested in winning an argument. You don't mention whether you're willing to learn anything, which is what I actually suggested. If you're neither interested in teaching nor learning then you may be a bad fit for contributing to DCSS.

Lasty, I have (and I still am) learned a lot developing this god. I don't think I have to bring up again how radically the god has changed. Even though it may look like it in hindsight, I didn't retrofit the ideas that I was offered and credit myself with them. I'm fully aware many of my design decisions were wrong, and I am not so deluded to think the god is in a perfect state right now, or that it would be with the changes that I'm proposing.

That said, when you sink dozens or even hundreds of hours (and believe me, I have) into a project you start to develop strong opinions about certain things. After trying out multiple variations of these ideas, I'd like to think I'm in a position to know that at least a handful of ideas lead into a dead end, at least in the particular ecosystem that is WJC. There are many, many things I'm willing to try and be proven wrong about, but there are a few that I know for a fact aren't functional, because I've tried them already. This isn't even about the fact that things were changed without my "permission", which is a ridiculous idea. If the only change made were the one to wall jump, I would've grumbled a bit but not oppose it completely. That specific whirlwind change however is bad for the purpose that the people who implemented it want to achieve.

Since the last patch I have played 15 different games with the current Trunk version, and about 30 with my custom version. I'm comparing them day to day with an open mind, I promise you, and the current trunk version is best played by repeating the same martial attack for every fight (including trivial fights), which is against what you and I want.

Lasty wrote:I explicitly _do not want_ players to have an incentive to use martial attacks in every fight, or even in the large majority of fights, which I think differs from your ideal situation. I also do not want players to feel that once they've used one martial attack they need to continue using martial attacks. I would prefer that Tab stay the default method of launching a melee attack even when playing with a god that offers martial attacks.

Sometimes I think you guys just don't read me.

I'll quote myself from the past few weeks on this very forum:
SteelNeuron wrote:Honestly, this is the thought process I'd like a player to follow:
  • Is the encounter difficult?
    • No -> Tab.
    • Yes -> Martial attacks.

SteelNeuron wrote:Whirlwind is essentially super-tab at the moment, and this is not my design intention. I explicitly want whirlwind to be good against a) groups, and b) single monsters that you anticipate will fight for a bit.

SteelNeuron wrote: Some players are more interested in the strategic aspect, while combat tactics are something they might want to focus on only for very specific difficult encounters. For this kind of player, knowing that martial attacks are unconditionally going to be better than tabbing, the god will result in frustration, as they will be forced to constantly and attentively engage with a part of the game that isn't their main focus.

SteelNeuron wrote:The caveat to people enjoying the constant adaptation, is that they stop enjoying it when it's forced on every fight. This is why I was campaigning to make tab relevant again by giving it the role of simple, efficient damage dealer, reserving the martial attacks for difficult fights where the player wants to be 100% engaged and use the kit.

SteelNeuron wrote:Otherwise [martial attacks] become a worse version of tab with more keypresses and less built in quality of life. This is the point that I'm trying to make: No martial move should be mindlessly spammable, and for this to work, tab needs to be useful as a direct, hands-free damage dealer while martial moves need to be just positioning utility

Dungeon Master

Posts: 3160

Joined: Sunday, 5th August 2012, 14:52

Post Tuesday, 28th March 2017, 10:24

Re: My final thoughts on WJC, and moving on.

Steel Neuron wrote:That said, when you sink dozens or even hundreds of hours (and believe me, I have) into a project you start to develop strong opinions about certain things.

There's nothing wrong with strong opinions, and I'm certainly not trying to discourage you from having them.
Steel Neuron wrote:After trying out multiple variations of these ideas, I'd like to think I'm in a position to know that at least a handful of ideas lead into a dead end, at least in the particular ecosystem that is WJC. There are many, many things I'm willing to try and be proven wrong about, but there are a few that I know for a fact aren't functional, because I've tried them already.

Since the last patch I have played 15 different games with the current Trunk version, and about 30 with my custom version. I'm comparing them day to day with an open mind, I promise you, and the current trunk version is best played by repeating the same martial attack for every fight (including trivial fights), which is against what you and I want.

But here's where the problem comes in: you seem to be stating that your strong opinions are fundamentally better than other people's because you've played the god a lot.

I think you're correct that the current trunk makes repeating martial attacks ideal, but we want to fix it in different ways. You want to fix it in a way that provokes the player to cycle through attacks. I want to fix it by not prompting players to use the attacks at all unless it helps with positioning. I think both solutions are likely better than status quo, but I also think MPA's change is the first in a likely series of changes that move things in the direction I think is best.
SteelNeuron wrote:
Lasty wrote:I explicitly _do not want_ players to have an incentive to use martial attacks in every fight, or even in the large majority of fights, which I think differs from your ideal situation.

Sometimes I think you guys just don't read me.

I'll quote myself from the past few weeks on this very forum:
SteelNeuron wrote:Honestly, this is the thought process I'd like a player to follow:
  • Is the encounter difficult?
    • No -> Tab.
    • Yes -> Martial attacks.

SteelNeuron wrote:Whirlwind is essentially super-tab at the moment, and this is not my design intention. I explicitly want whirlwind to be good against a) groups, and b) single monsters that you anticipate will fight for a bit.

SteelNeuron wrote: Some players are more interested in the strategic aspect, while combat tactics are something they might want to focus on only for very specific difficult encounters. For this kind of player, knowing that martial attacks are unconditionally going to be better than tabbing, the god will result in frustration, as they will be forced to constantly and attentively engage with a part of the game that isn't their main focus.

SteelNeuron wrote:The caveat to people enjoying the constant adaptation, is that they stop enjoying it when it's forced on every fight. This is why I was campaigning to make tab relevant again by giving it the role of simple, efficient damage dealer, reserving the martial attacks for difficult fights where the player wants to be 100% engaged and use the kit.

I genuinely think you're pushing the god in a direction that improves it from where it was and which meets the goals you have for it. I think that your ultimate goal is very different from mine, and in that respect I don't think you have understood my goal at all: I really would prefer if players get no bonuses from martial attacks aside from positioning, because I think they are most interesting and least tedious as a tool that's only purpose is to change position as part of your attack. I think that incentivizing players to use the martial attacks purely to get non-positional combat bonuses is inherently problematic, whether it's used in every fight or only tough fights. Further, I believe that whatever is ideal in a tough fight is also ideal in a trivial fight unless there's an explicit cost tied to it that makes it a limited use ability.

SteelNeuron wrote: Some players are more interested in the strategic aspect, while combat tactics are something they might want to focus on only for very specific difficult encounters. For this kind of player, knowing that martial attacks are unconditionally going to be better than tabbing, the god will result in frustration, as they will be forced to constantly and attentively engage with a part of the game that isn't their main focus.

I think this quote is the closest to acknowledging what I see as being the heart of the matter, but I haven't seen any suggestions from you about resolving it. I have consistently had the impression that your goal is to incentivize the player to use these tools as a combat boost rather than make them available for the player to use only when they need positioning. That impression is reinforced for me because I haven't seen you discuss a plan that would move in that direction; I have only seen plans that incentivize the player to rotate through the martial attacks.

The only way that whirlwind would be ideal against tough targets and not weak targets is if the player could correctly forecast that whirlwind's damage + lunging/vaulting to finish the fight would result in more AUT spent in combat + healing than just tabbing against weak targets. I'm not sure how that's possible without making the same true against tough targets, but even if it were possible, it would be fairly difficult to communicate that accurately enough to the player to allow them to correctly forecast it.

For this message the author Lasty has received thanks: 2
duvessa, mattlistener

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Tuesday, 28th March 2017, 10:41

Re: My final thoughts on WJC, and moving on.

I am happy because it looks like we're reaching a point close to understanding. If we drill down a bit more we might find (as I've suspected for a while) that we want the same thing but we are shit at communicating.

Lasty wrote:But here's where the problem comes in: you seem to be stating that your strong opinions are fundamentally better than other people's because you've played the god a lot.

I wouldn't say that, but I'd say that opinions of people that have played the god a lot are better than those from people that haven't, and the great majority of testers that have put more than 20 games on the god seem to agree with me. Keep in mind that not all of them have been unconditional fans of my ideas, I'm counting people here that have been very critical of my decisions before.

Lasty wrote:I think you're correct that the current trunk makes repeating martial attacks ideal, but we want to fix it in different ways. You want to fix it in a way that provokes the player to cycle through attacks.


No, no, no! This is at the core of the misunderstanding we keep having. I think the opinions of some people that like the god for different reasons that I do keep being conflated with mine.

I don't intend for WJC to be a combo god. I doubt I've described it as such before, and if I did I apologize, because I wasn't fully aware of the implications. The idea was never to create a flowchart of moves leading into each other. My focus all this time is to not make any move overcentralizing. The bane of the god, and what causes all the tedium, is for a martial attack to be repeatable without a drawback. The solution that I want is not to force the player to cycle through them at all. The solution I want is for all three moves to be desirable if you need the positioning, while keeping tab as the only indefinitely repeatable one, i.e. the only open loop. If the player wants to keep chaining martial attacks because the positioning benefits them (i.e. staying in the fringes of a pack) that's fine. If they choose to start tabbing at any point, that's also fine and should have a clear, enforced role.

Lasty wrote: I think both solutions are likely better than status quo, but I also think MPA's change is the first in a likely series of changes that move things in the direction I think is best.


And I strongly contest that. MPA's change is completely backwards. If you read my point at the other post about damage frontloading, you'll see that current whirlwind is strictly better than tab for single targets, which means there is no reason why you'd want to tab in the current implementation. Reducing the damage in absence of a secondary effect also doesn't work because of the relationship between move delay and attack delay. Whirlwind needs a non-stacking effect or some form of diminishing return to be functional, or a new idea that nobody has come up with yet that would definitely not be in line with MPA's change.

Lasty wrote:I genuinely think you're pushing the god in a direction that improves it from where it was and which meets the goals you have for it. I think that your ultimate goal is very different from mine, and in that respect I don't think you have understood my goal at all: I really would prefer if players get no bonuses from martial attacks aside from positioning, because I think they are most interesting and least tedious as a tool that's only purpose is to change position as part of your attack. I think that incentivizing players to use the martial attacks purely to get non-positional combat bonuses is inherently problematic, whether it's used in every fight or only tough fights.


An interesting insight about this is that I never used slow as a debuff because of the combat bonuses it provides, but merely for the positional element. Positioning is not only about where you are, but where your enemies will be. That's why I chose distract and slow. However, a possibility for the god, and a way in which I think your goals and mine would align, would be to have a new status effect (let's call it dizzy for a second) that only impairs mobility but not other actions. Whirlwind could cause this dizzy state (and deal less damage as I have been proposing), so a player would only be interested in using whirlwind if they plan to make use of the resulting change of mobility on the enemy for crowd control or lunges, not to reduce incoming dps.

Lasty wrote:The only way that whirlwind would be ideal against tough targets and not weak targets is if the player could correctly forecast that whirlwind's damage + lunging/vaulting to finish the fight would result in more AUT spent in combat + healing than just tabbing against weak targets. I'm not sure how that's possible without making the same true against tough targets, but even if it were possible, it would be fairly difficult to communicate that accurately enough to the player to allow them to correctly forecast it.


But at that point you're delving too deep in optimal play. The forecast is entirely personal. Some players will use the martial attacks more, some less, but by giving tab a facet at which it comparatively excels, nobody is strictly punished for making either of the choices.
Last edited by Steel Neuron on Tuesday, 28th March 2017, 11:24, edited 4 times in total.

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Tuesday, 28th March 2017, 10:57

Re: My final thoughts on WJC, and moving on.

Just to expand on this misunderstanding problem about whether this is a "combo god", let me say that it's completely fine by me if a player starts tabbing after successfully applying a martial attack. In fact, there are several valid reasons why you'd want to do this in the revision I'm proposing, for example:

  • After a successful whirlwind (slow), it's valid to start tabbing instead of setting up the other moves, because this will maximize the time you've spent adjacent to the enemy dealing damage while it's still suffering the slow effect. The alternatives, while safer, would consume a lot of this time. With the dizzy alternative this would be slightly different, in that you wouldn't have used whirlwind to begin with if you didn't want to exploit the positioning. Both options are fine.
  • After a successful lunge, for instance against a sleeping target, it's also perfectly OK to start tabbing because relying on energy randomisation to line up lunge is not worth the price of doing business.
  • After a wall jump (and this is the hardest one to explain) it's valid to tab, because due to the geometrical limitations on the kind of jumps you can do, it's very difficult to lunge directly against an enemy you just wall jumped away from (knight's jump problem). Therefore, to line up a lunge, you run the risk of the enemy waking up from the distraction before you approach, where tab would have been more efficient. This is something that you figure out after playing a few games, at first you want to lunge after each successful wall jump, but at some point you realize this is equivalent to throwing a coin, and you might choose not to unless you think the payout is worth it.

Note how in the current Trunk version, there is no reason why you'd want to tab after a whirlwind, or after a wall jump (which I'm also proposing gets less damage). You should keep spamming and in fact, there is no reason why you'd even want to check the result of your move, with the exception of watching over your HP in a tedious, human substitute of autofight_stop.

bel

Cocytus Succeeder

Posts: 2184

Joined: Tuesday, 3rd February 2015, 22:05

Post Tuesday, 28th March 2017, 11:53

Re: My final thoughts on WJC, and moving on.

Steel Neuron wrote:Note how in the current Trunk version, there is no reason why you'd want to tab after a whirlwind, or after a wall jump (which I'm also proposing gets less damage). You should keep spamming and in fact, there is no reason why you'd even want to check the result of your move, with the exception of watching over your HP in a tedious, human substitute of autofight_stop.

This is wrong. Here's a simple scenario where it is best to tab after a whirlwind. @ is you, M are speed 10 monsters, x and y are positions inside a corridor. It is optimal to first move to x (whirlwind), then take a step back to y, then Tab. In this scenario, you only fight one monster at a time.

  Code:
#y#
#x##
.M@##
.MM.#
...#
..##
..##


I am not claiming that this is a good outcome. As I have said elsewhere, the current version of martial attacks incentivize (and indeed give tools to accomplish it) people to move as quickly as possible to a corridor to Tab.

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Tuesday, 28th March 2017, 12:03

Re: My final thoughts on WJC, and moving on.

Sure, on that sentence you quoted I was referring to single targets. When you get multiple enemies involved, there are many particular scenarios where a combination of moves may be better than repeating one, but they're generally the exception.

bel

Cocytus Succeeder

Posts: 2184

Joined: Tuesday, 3rd February 2015, 22:05

Post Tuesday, 28th March 2017, 12:25

Re: My final thoughts on WJC, and moving on.

Ok, but for a single monster it makes no difference whether you whirlwind or Tab.

There's the issue with the martial attacks triggering at the start of the round, but that should be fixed independently of this point. Also, it cuts both ways: whirlwind increases variance (because of the rounding), so it's not be a good idea to whirlwind in general against single targets (except maybe the last round or so).

But I doubt anyone cares about squeezing every single bit of juice out of the mechanism. For practical purposes, whirlwind is the same as Tab against single monsters.

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Tuesday, 28th March 2017, 13:01

Re: My final thoughts on WJC, and moving on.

Lasty and I have been talking a lot over ##crawl-dev, and have come to a possibly satisfying solution. We don't know it will be perfect, and Lasty has reservations still, so I'm going to throw together a prototype and test it with his help in the future. Let's see where this takes us!

EDIT: And here we go https://github.com/crawl/crawl/pull/502

For this message the author Steel Neuron has received thanks: 3
Brannock, mattlistener, nago

Spider Stomper

Posts: 236

Joined: Saturday, 2nd July 2016, 13:16

Post Tuesday, 28th March 2017, 15:59

Re: My final thoughts on WJC, and moving on.

bel wrote:As I have said elsewhere, the current version of martial attacks incentivize (and indeed give tools to accomplish it) people to move as quickly as possible to a corridor to Tab.

Even more true of the pull version, I'd think!

I really don't see how a god whose attacks all do less single target damage than tabbing, but have special effects that all make it easier to escape to a corridor, can end up being anything other than the "escape to a corridor and tab god."

For this message the author luckless has received thanks:
duvessa

Ziggurat Zagger

Posts: 8786

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

Post Tuesday, 28th March 2017, 17:43

Re: My final thoughts on WJC, and moving on.

I have come to the conclusion that wall jump just doesn't belong in this game at all.

Yes, it gives the player "more options"; there are now more possible actions for players to take on most turns, and yes, that makes the decisions more complex. But an increase in complexity doesn't mean an increase in depth. It just means an increase in complexity. To give an extreme example, Crawl is a lot more complicated than Go or chess, but it's a lot harder to become a good Go or chess player than it is to become a good Crawl player.
The defining feature of wall jump is getting to move multiple squares in one action if you're next to a wall. You can make it good at damage dealing or disabling monsters but neither of those make for interesting core features of the ability. The only features of wall jump that make it unique are using it to move faster - which is almost exclusively useful for running away - and using it to jump over a monster - which is almost exclusively useful for running away. Crawl already has a ton of options for running away, and adding another method of running away doesn't really add any depth. Ultimately, wall jump is style over substance, and while flashiness is not a problem in itself it's also not enough to justify hurting gameplay.

Lunge and statusless whirlwind don't suffer from this problem because they don't help you escape; you and the monsters are in the same locations whether those abilities exist or not. But they still add something to the game by taking two options that are normally bad (moving towards a monster, moving from one square adjacent to a monster to another square adjacent to the monster) and making them better. They need to be careful not to be too good or they overshadow the other options and end up reducing depth. I think the 502 version of lunge is perfect. As luckless alludes to, I'm worried that Dizzy will make it too easy to run away from monsters (although I'm still disgusted by how people pretend corridors are the best terrain).

For this message the author duvessa has received thanks: 3
Brannock, luckless, Steel Neuron

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Tuesday, 28th March 2017, 20:18

Re: My final thoughts on WJC, and moving on.

You make some good points.

I think that dynamic may be different enough by adding a cost, though. It can be more interesting as an ability by being more limited.

I have to admit I'm getting to the end of my ideas and my drive to keep trying to make it work better though.

Ziggurat Zagger

Posts: 6454

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

Post Tuesday, 28th March 2017, 20:47

Re: My final thoughts on WJC, and moving on.

There are several god abilities with different limitations that allow you to escape or move several squares in one step. Dith, ru, lucy, usk, to name a few all have the ability to move as an escape action. I think some sort of escape ability with some kind of cost is completely appropriate and even desirable.

The issue with wall jump as it is is that "being next to a wall" is not a sufficient cost for the benefit provided, not that the benefit is an unreasonable one altogether.
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!

bel

Cocytus Succeeder

Posts: 2184

Joined: Tuesday, 3rd February 2015, 22:05

Post Wednesday, 29th March 2017, 03:39

Re: My final thoughts on WJC, and moving on.

I see that wall jump in the pull request has a piety cost.

I am hopeful that, eventually, people will do the right thing after exhausting every other option. Namely: martial attacks should be an active ability costing piety, rather than being "always on".

The reasoning is simple, and one I have given before:

(a) To take advantage of the mobility, you want people to fight in the open.
(b) Fighting in the open is only worthwhile if the martial attacks are powerful.
(c) Powerful martial attacks shouldn't be free.

The "not free" part could conceivably be achieved in other ways (like Uskayaw's piety mechanic); I gave some ideas in the other thread.

For this message the author bel has received thanks:
luckless

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Wednesday, 29th March 2017, 06:56

Re: My final thoughts on WJC, and moving on.

bel, what you are proposing is already achieved by Serpent's Lash.

Serpent's Lash is the active trigger that turns a set of abilities that are good for mobility and positioning effects into offensive tools. Gating all martial attacks behind an activation cost would make them much less intuitive and desirable.

If Serpent's Lash is too costly, to the point that conversion can't be done as often as the player would want, the cost can be adjusted.

For this message the author Steel Neuron has received thanks:
Lasty

bel

Cocytus Succeeder

Posts: 2184

Joined: Tuesday, 3rd February 2015, 22:05

Post Wednesday, 29th March 2017, 07:40

Re: My final thoughts on WJC, and moving on.

No, Serpent's Lash is not the same thing at all. It would be somewhat similar if you couldn't use martial attacks at all without Serpent's Lash. But I doubt anyone is in favour of the latter.

The "dizzy" status is basically the same as a "-Swift" status. I proposed a variant of this in the other thread (using stick instead of carrot). As I said there, the solution is bad if the martial attacks are free.

For this message the author bel has received thanks: 2
duvessa, luckless

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Wednesday, 29th March 2017, 12:52

Re: My final thoughts on WJC, and moving on.

Let's focus on wall jump for a second.

I propose these wall jump changes:

  • Wall jump is free and usable everywhere.
  • For a 10 aut move delay character, wall jump takes 20 auts to execute (same as moving there normally).
  • Wall jump distracts all enemies adjacent to the origin and landing spots for 10 auts, and has a HD based chance to distract them for longer. It no longer distracts distant onlookers.

The end result is that, if you're jumping into, or away from an enemy, it would behave as old wall jump as the enemy will only get AUTs for one action. However, more distant enemies can catch up with you at normal speed.

I think this achieves many of the things we want. First, it's not usable as level traversal (since you gain no extra distance). Second, it allows you to gain one tile of distance, but not more, preventing abusable kiting. Third, it has no combat advantage over tab unless you are interested in the AoE.
Last edited by Steel Neuron on Wednesday, 29th March 2017, 14:37, edited 1 time in total.

For this message the author Steel Neuron has received thanks: 3
Floodkiller, Lasty, ohmi

Abyss Ambulator

Posts: 1217

Joined: Sunday, 14th April 2013, 04:01

Post Wednesday, 29th March 2017, 13:56

Re: My final thoughts on WJC, and moving on.

Steel Neuron wrote:Let's focus on wall jump for a second.

I propose these wall jump changes...


That sounds absolutely ideal to me personally, the 20 aut thing I think makes it.
Three wins: Gargoyle Earth Elementalist of Ash, Ogre Fighter of Ru, Deep Dwarf Fighter of Makhleb (0.16 bugbuild :( )

Slime Squisher

Posts: 365

Joined: Monday, 7th January 2013, 08:22

Post Wednesday, 29th March 2017, 14:30

Re: My final thoughts on WJC, and moving on.

It's now part of the PR.

Distract only affects enemies adjacent to the launch point and landing, but even a failed distract will still set the monster back by an amount of time equal to your (normal) move speed, as explained before.

Spider Stomper

Posts: 236

Joined: Saturday, 2nd July 2016, 13:16

Post Wednesday, 29th March 2017, 16:56

Re: My final thoughts on WJC, and moving on.

Steel Neuron wrote:bel, what you are proposing is already achieved by Serpent's Lash.

Serpent's Lash is the active trigger that turns a set of abilities that are good for mobility and positioning effects into offensive tools. Gating all martial attacks behind an activation cost would make them much less intuitive and desirable.

If Serpent's Lash is too costly, to the point that conversion can't be done as often as the player would want, the cost can be adjusted.

Serpent's Lash doesn't encourage you to fight in the open, though it would if it lasted longer or didn't exhaust. At most, it encourages​ you to wait a turn before your Dizzy-facilitated retreat to closed terrain, so you can get a couple big hits.

Return to Game Design Discussion

Who is online

Users browsing this forum: No registered users and 29 guests

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