Voting on implementables! [Updated]

[Wensleydale]

[UPDATE]
Thanks for all the replies and discussion, everyone! It's been a month since this topic was posted, so let's look at the most popularly requested new features thus far:

Tier 1:

Abstract Demigod Worshipers (graciously undertaken by mumra)
Hive Reform
In-Game Autopickup (as an interface implementable, this would be great first-time patch material for aspiring developers!)

Tier 2:

Permabuffs (my personal favorite hurry, galehar!)
Changes to hexes, including making failed hex spells reduce monster MR and new low- and high-level hex spells
General improvements to melee combat, such as type-specific weapon abilities or tactical combat maneuvers (not technically an authorized implementable, but it was so popular that I just had to mention it)

Tier 3:

Changes to rP
Golems
New plant monsters
Improvements to Poison Magic/Alchemy (also not technically an implementable, but very often requested)

Honorable Mentions:

Spider's Nest
Food Reform (Done!)
Player in the arena
Support for translations
God Wrath Overhaul
Make Passage of Golubria more useful
Forest Branch (once again, not technically an implementable, but maybe I'll get started on some portal vaults...)

Oh, and as always, "interface implementables" makes a strong showing but is a bit too vague and sweeping to be included alongside the above. If you really like any of the proposals on the interface implementables page, please reference them by name.
[END UPDATE]

If you've ever visited the mantis or development wiki, you've probably realized that Crawl is a game with a lot of ambition. The process of realizing this ambition, however, is almost entirely undirected; no formal criteria govern the order in which features are implemented. Most of the time, developers simply embark on whatever project most interests them at the moment. However, on the rare occasion where a developer is searching aimlessly for a project, it would be useful if they had a jumping-off point. To this end, dpeg has authorized me to create this thread, where I'll be compiling suggestions from the community regarding which features you'd like to see implemented first. Here's what you need to do:

1) Become familiar with the list of implementables. The canonical list is here on the mantis, although the devwiki planning page for the next release is also fair game. Furthermore, the devwiki may be hiding features that dpeg has indicated are ready to become implementables, but have never been registered as such on the mantis; if you find any such features, they are also fair game.

2) Pick as many implementables as you'd like, and reply to this post with your list of implementables ranked in order of importance. Your list must be ranked in order: the point of this exercise is not to determine what features you'd like to see in the game, but how soon you'd like to see each feature in the game.

Note that Tavern posts are not valid sources for implementables, likewise for devwiki pages that dpeg has not personally endorsed. Once we have enough responses to this post, I'll compile all the lists into a master list of ranked implementables which developers and independent contributors alike can reference when they need inspiration.

But wait, there's more! The second purpose of this post is to assemble prospective features from this forum and the devwiki and determine which have the most community backing. Features with sufficient support may be considered to become implementables. Note that this does not constitute a promise that sufficiently popular features will be implemented. Once again, here is what you must do:

1) Scour the forum, trawl the devwiki, and speak privately with your fellow Taverngoers in order to discover features that you'd like to see in the game.

2) Compile a list of as many features as you'd like, ranked in order of importance. For this list, briefly describe each prospective feature and include one or more links to sources on the Tavern or the devwiki that explain it in detail. No details, no implementable! Include this list in your reply, below your list of ranked implementables.

And, when compiling this second list, please keep in mind the List of Things that Won't Be Done.

[dpeg]

Thanks for the post, I think this is a great idea.

Some words about the development process: it is not fully random. There are no formal criteria, that much is true. However, when a version has been released, and planning for the content of the next version starts, two things happen: someone (I used to do this until 0.8 by way of "0.6 agenda " etc. e-mails) lines up a number of important topics; and coders express topics they're interested in.
From the point of view of global design, a topic can be "important" for a number of reasons: It solves a balance issue; it is a relevant interface improvement; it finished something that ought to be polished; it adds something flashy and cool (each new release should come with something nifty). If a topic meets several criteria, it is even more important. (An example are the Shoals: they've been started version ago; a new branch is really cool; they added some strategic depth and set up the principle of branch rotation.)
These days, the e-mails to the developer mailing list are replaced by a 0.10 development wiki page.

Why does it work like this? Crawl is developed by volunteers. Hunting down bugs and nerfing day in and day out might be the most important things to do, but no coder can be blamed for wanting to have some fun. This is why it's absolutely okay if developers pursue their hobby-horses (which are discussed like any other idea, of course).

Since stability is so important, many pretty cool ideas will never be implemented. (If you have some idle time, browse the development wiki for the list of species or god or branch proposals. Almost all of this will never materialise.) This means that (a) spotting the *really* good stuff in the heap is important; (b) appealing to Crawl players might work. Since roguelikes are a rather peculiar pastime, there are many players who can code. We have started asking for support with vaults, then tiles and later speech. With all of these, support was, and is, enormous. Implementables are just the next step: we're asking code-literate crawlers to help out with certain patches. We have tried to select good ideas, so that patches are almost guaranteed to go into trunk, assuming they run. (Nothing prevents you from sending in a patch for your pet idea, but it might not be compatible with our plans at all. Check the old Sourceforge patch list for a number of examples.)

The first proof how well implementables can work is the slime god. This was an old idea, mentioned briefly by someone on Sourceforge (in an unrelated thread even). The idea caught attention and was fleshed out after a while. Then nothing happened: the slime god has a lot of appeal, but it is clearly less urgent than whatever our coders were doing. So (this was still pre-Mantis) we said that a patch is welcome. Within days, 78291 sent in a patch.

On Mantis, we have institutionalised this with the Implementables category. These are of different difficulty, many are for the interface (almost all Crawl developers started by improving the interface in some way; this seems to be a good way to get to know the code in a local and controlled manner). One goal is to raise regular contributors this way. (The patcher is not just an unpaid coding slave. There are often design gaps to fill, and the Implementable is just a suggested specification. A coder can really shape the subject.)

In this thread, I am curious to read what you would think are good candidates for further Implementables. Such features should be quite detailed -- almost all god, species, branch proposals on the wiki are not detailed enough, for example. An Implementable is not the place to brainstorm: a potential patcher should get a very clear idea of what is desired (and ideally also why). Please refrain from adding Implementables of your own. Part of the appeal is that these are "devteam-approved". You can lobby for them elsewhere, for example here.

[mumra]

Favourite implementables

1. Spiders Nest
A lot is being and has been done already but there's plenty left.

2. Abstract Demigod Worshippers
Might inspire me to try out a few Demigod characters for a change!

3. In-game selection of auto-pickup items
I'm slightly less bothered about this since knives were removed but later on in the game, it'd be great in particular to turn off spellbooks and low-level wands

4. Hive as a portal vault
Initially I wasn't too fussed about this; but actually, Hive is usually just an annoying food / royal jelly dive, and anything to spice it up a bit would be welcome.

5. Forest Branch
I seriously doubt it'll happen all that soon, but I'm a big fan of the idea.

Favourite could-be-implementables

(Not particularly in any order; except the first which I think would have the biggest impact)

Alchemy ( https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:magic:alchemy and https://crawl.develz.org/tavern/viewtopic.php?f=8&t=2133 )
Rebranding poison magic and having a reason to train it higher (and in particular a spell converting gold to silver projectiles...)

Snail Trails ( https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:monster:snails )
Implementation could be somewhat similar to spider webs; and could add some environmental character to Lair.

New Spiders ( https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:dungeon:branch:spider )
Perhaps Spiders Nest needs a small number more monsters and uniques. My favourites are barb-spitting and web-spitting (spiders lack ranged threats right now). And of course the Hellspider Also an egg-laden spider releasing swarms could be a good monster or unique.

New Plants ( https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:monster:plants )
Especially if the forest branch were to happen, a new range of flora-based threats would be a must. There are a whole load of ideas I like on that page. Also it'd be nice to have some variation on the standard Plant tile, maybe that could be achieved just with purely cosmetic variation.

Revolving Door Gods ( https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:god:revolving_doors (inspired by https://crawl.develz.org/tavern/viewtopic.php?f=8&t=2431 and
https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:god:doorokhloe) )
Two-for-the-price-of-one-god!

A couple of my ideas from a while back:

Vortices ( https://crawl.develz.org/tavern/viewtopic.php?f=8&t=1912&p=21423 )
High-level translocation spell, needs some sorting out and writing up on the wiki.

Thunder Bugs ( https://crawl.develz.org/tavern/viewtopic.php?f=8&t=1976&p=21905 )
Electricity-themed walking, then stationary, then flying enemies (maybe suitable for an elec-themed branch idea someone mentioned on ##crawl)

[Wensleydale]

Mumra gives an excellent example of what we're looking for. Feel free to use his post as a template.

[MyOtheHedgeFox]

1: Announce -cTele and Bazaar/Lab. A very nice feature. Those who forget about such things, like me, could benefit from this. =)

2: Interface Implementables. I must vote for this.

3: Vintermann's Food Reform. This suggestion is very well organized. =)

4: Make rP only 90% effective. There would be a good reason to carry a potion or two with you, and perhaps even to make small "healing bases" in some well-locked places.

5: Demigods: Abstract Worshippers. This would really spice up the gameplay.

6: Hive Reform. This could be a large Crawldevelzorg Orb Of Nerfation for Spriggans, but also a large and nice rod of happiness for those who finnally find the updated Hive.

I know this is not an implementable, but:
6a: New Plant Monsters. (what?) This idea looks really nice, and having these would really make a player consider choosing Fedhas Madash as a deity for his adventurer - just like choosing Lugonu for easing the Abyss travel.

Of course,
7: Add the possibility to translate the game easily. This task requires a lot of work, but it can make the game even more popular.

[mumra]

7: Add the possibility to translate the game easily. This task requires a lot of work, but it can make the game even more popular.

I also support this. It would be a huge task, but would actually have additional hidden benefits. Currently, a huge amount of Crawl's text is scattered around the source code; in order to make translations work, all text would need moving out into separate files. This would be an Extremely Good Thing in my opinion; it would effectively standardise the way that text is defined - making it much, much easier to tweak existing text as well as providing solid guidelines when adding new text.

[ryzol]

Quick note - Could people please link to the features? This way others can easily view them and decide if they like them or not.

[MyOtheHedgeFox]

The things I voted for can be easily found in the Mantis list of Implementables and in the 0.10 plan. If there is no link, seek it there. =)

7: Add the possibility to translate the game easily. This task requires a lot of work, but it can make the game even more popular.

I also support this. It would be a huge task, but would actually have additional hidden benefits. Currently, a huge amount of Crawl's text is scattered around the source code; in order to make translations work, all text would need moving out into separate files. This would be an Extremely Good Thing in my opinion; it would effectively standardise the way that text is defined - making it much, much easier to tweak existing text as well as providing solid guidelines when adding new text.

I would also say that allowing translations would also require non-Latin keyboard layouts support.
In example, I'm sure that would be necessary for a Russian translation.

Speaking about control - all the non-English letters might work as their Latin counterparts, i.e. Cyrillic п works like g when pressed during the game, though it still prints "п" when you type something in a textfield.

[ryzol]

1 - Automatic annotations for uniques and ghosts. This is a convenience and would strongly help new players. https://crawl.develz.org/mantis/view.php?id=4488

2 - Make hexes more usable by temporarily reducing MR. If done this would work will with reviving the Arcane Marksman.

3 - Make rP weaker, but make monster poison weaker. There are too many areas that you must avoid without rP. Weaker rP also makes Poison Magic stronger which it probably needs. https://crawl.develz.org/wiki/doku.php? ... ect:poison

4 - Allow player in the arena. Give it a save file and the player is controlled with autofight. Will help with future fighting balance. The sooner this is done the easier it is to test esoteric features. I would also like to see manual player control of player files in the arena as AI fights sub-optimally. This would improve the long term development of crawl by making balance testing easier.

5 - Improve Demigods. Demigods are stupid boring. They need help. https://crawl.develz.org/mantis/view.php?id=3046

6 - Fighting rewrite. Melee characters are less interesting to play than casters because they have fewer choices. However, this would be a huge change. If I had to chose between everything else I listed and a Fighting rewrite I would rather have everything else. https://crawl.develz.org/wiki/doku.php? ... ht_rewrite

7 - More difficult mid game and mid-late game. Especially the midgame. Zot is difficult enough. The last level of a branch seems difficult enough. But Lair is very easy. So are Swamp/Snake 1-4, and Shoals 1-2. I think midgame difficulty somewhere around Orc:3 would be a good target. Bottom line is that the early game should not be so much harder than midgame and lategame.

8 - One new deity. It's always cool getting something flashy in a patch and not just tweaks.

9 - A change to Passage of Golubria. As is it is somewhat useful, but not really worth 4 spell levels. 4 SL is too high for someone with bad casting aptitudes to use, and spell takes 8 mana to use. Furthermore it is situationally less useful than Blink and the Passage placement is inaccurate. I think a bandaid fix by lowering PoG to 2 SL is good for now. Later on I would like to see the spell changed to be useful after Controlled Blink is castable. https://crawl.develz.org/wiki/doku.php? ... or_replace

10 - Have Vehumet gift Ring of Flames and support its use.

11 - Change the name of Enchanters. Their spell book is all hexes. The class isn't based around hexes either, it's based around stabbing.

[Guppyfry]

1 - Revive Arcane Marksmen/Make Hexes more useful - I don't think I've ever trained hexes, I like the idea of it getting boosted with some sort of magic resistance attack and a new background

2 - Fight rewrite - Fighting could be a lot more interesting even if it kept only to passive abilities, so the player experience was still fast paced and simple compared to spellcasters (https://crawl.develz.org/wiki/doku.php? ... ht_rewrite)

3 - Interface implementables - A few of these caught my eye: More messages explaining travel-stops and auto-explore, condensed (flavored) combat messages, and improved auto-targeting (https://crawl.develz.org/wiki/doku.php?)id=dcss:brainstorm:interface:interface_implementables

4 - More plants/Plant monsters/Forest branch - Breaks up the dungeon feel and helps make Fedhas more attractive

5 - Golem/Slow race - There's a bunch of golem ideas around it seems like, I think distilling them and getting a new slow race would be great and gives Chei a little more appeal

6 - Hive as a serial vault - I like squashmonsters idea of the hive being a series of vaults that are little food banks of increasing difficulty. Although as a hoarder I'd like to have a deep branch to stash in if hive turns into a vault. (https://crawl.develz.org/wiki/doku.php? ... rtal_vault)

7 - I like the demi-god worshiper idea and the revolving doors god a lot, as well as all the xom ideas - I think these are very cool, but are a tier below the above ideas for me

[mumra]

2 - Fight rewrite - Fighting could be a lot more interesting even if it kept only to passive abilities, so the player experience was still fast paced and simple compared to spellcasters (https://crawl.develz.org/wiki/doku.php? ... ht_rewrite)

Oh yes, I meant to include this in my list; personally I like the idea of special weapon abilities (see: https://crawl.develz.org/wiki/doku.php?id=dcss:brainstorm:combat:weapon_reform) - I basically never play melee characters because I find repeatedly hitting things a bit boring, compared to the cool and awesome spells that wizards can have

[Galefury]

AFAIK the fighting rewrite mentioned in the 0.10 plan is a behind the scenes thing, not a new features thing. It would make adding new combat mechanics a lot easier, but they are not part of the fighting rewrite. Of course that doesn't mean it is any less important or desirable, and it really should get done. It is far beyond the scope of an implementable however.

Another thing that seems too large scale for an implementable that I would really like to see is a god wrath overhaul. God wrath sucks in many ways: it is mostly non-threatening (resource drain, inconvenience) but has high potential to kill you if it happens at a bad time. It lasts way too long, it is usually avoidable by waiting it out somewhere safe, and excruciatingly boring when waited out. Making god wrath shorter, more focused, and less harsh at low levels, but only occurring while actually in interesting situations would be great (no strategic effects in combat, no combat effects when not in combat). Preventing people from waiting it out by switching to an xp/exploration based formula is also good, but doesn't seem like it would work well with the current wrath effects. I haven't seen a precise proposal for this, but it seems like something the devs have been wanting to do for quite a while (listed as deferred from 0.9).

I'm not seeing galehar's skill level fuzzing idea on the 0.10 plan page. Oversight or did it get scrapped/deferred to a later version? I think having this in 0.10 would go a long way towards making automatic training viable.


Onwards to things that might actually work as implementables. Unfortunately most of these are a little low on details:

1. Hive reform (already an implementable, but one I would really love to see realized). The current hive is booooring. Worst branch by far, only interesting strategically (piety now/later, food now/later, a bit of loot now/later), not tactically. It is also very similar to what's planned for the spider nest branch, a place where fast poisonous stuff swarms you. We don't need two branches with the same gameplay, so turning it into a portal vault and adding a little race for food minigame to it seems like a good idea. It honestly cant get much worse than it is.

2. Hell effect overhaul (tension based). No proposal for this on the wiki or mantis, but there's a tavern thread about it that has some detailed ideas. Strategically relevant effects and pure rest stoppers should happen when not in combat, while effects only relevant in combat should only happen in combat.

3. Mephitic Cloud and Evaporate nerf. Making resistability, duration or size of noxious clouds depend on spellpower, then using that to nerf MC and Evap and set them apart a bit more would be great. MC/Evap trivializes many encounters, which is boring. This would also make it easier to nerf transmuter ghosts and swamp drakes. No detailed proposal on this, but it's been talked about a few times, and changing clouds to depend on spellpower is in the 0.10 plan page (in the deferred from 0.9 section).

4. rP overhaul. The all or nothing approach to rP doesn't work very well. Turning rP into a two or three pip resistance and possibly nerfing enemy poison a little to compensate (or providing more ways to get rP) would be great. With rP soon to become even more important due to the addition of the spider branch this seems like something that should be done soon. Also on the 0.10 plan in the deferred from 0.9 section.

5. Interface improvements such as in-game autopickup toggle and automatic ghost/unique annotations are of course always welcome. Most things on the interface improvements wiki page seem nice to have, but nothing really stands out to me. One thing I would really like to see improved is the tiles monster list. I already have an easy at a glance list of enemies, on the regular dungeon screen. Showing me all the monster tiles again provides no additional information, and the at-a-glance accessibility of the information is barely improved over the main screen. Something more in the style of the console monster list would be more useful IMO.

[dpeg]

Wrath overhaul is really important, yes. We even had someone start to implement something, but it didn't get finished, unfortunately. Redoing wrath has a number of positive consequences, my favourite one being that we can redo Lugonu to allow corruption of altars.

[galehar]

I'm not seeing galehar's skill level fuzzing idea on the 0.10 plan page. Oversight or did it get scrapped/deferred to a later version? I think having this in 0.10 would go a long way towards making automatic training viable.

I haven't edited the 0.10 planning yet, but skill level fuzzing will be there. It's in my personal todo list which I should wikify and open up for priority votes

[Saegor]

1) Slow races (golem/treefolk... or nerf ghoul speed => 0.8 or 0.9 because zombies are SLOW)

2) Demigod unleashed (possibility to choose a divine power : confuse/smite/torment/...)

3) Add the possibility to translate the game easily. Please. Pleeease.

4) Timed portals overhaul. Maybe set a trigger when a portal in seen, not just present on the the level (but maybe the spider entry bug is OK now)

[oiseaux]

1) Demigod overhaul / changes: Do something with this race or just remove them.

2) Changing autopickup options in-game: This gets particularly annoying in the late game when a lot of items are no longer necessary/worthwhile.

3) Hive as a portal vault: Hive is sheer tedium and yet another rPois branch. On top of this, rarely do I need extra food at this point in the game, so I just find myself coming back to it if I need a "top up" later on. There's so much food in there I have to make several trips, too. Having said that, we will need somewhere to stash our items if it gets removed.


Things I'm really against:

* Spider branch - No, no, no, no no no noooooo. After the long trip through the lair, and swamp, I can't think of anything worse than yet another uninspired branch that requires rP. Add to this teleporting creatures like jumping spiders and you've got an exercise in tedium. Although there are a lot of spiders in the game I find the whole theme of this branch to be really unimaginative and too similar to Lair. There are so many quality branch ideas that will provide a truly unique experience and theme, but spider is not one of them. God no.

* Adding another race - Felids and Octopodes were both really gimmicky and I'd prefer focus on overhauling what we have to synergise better with gods and backgrounds. Quality, not quantity.

[Galefury]

With the webs and the hellspider (and possibly temporary rP consumables) the spider branch can be called anything but unimaginative. From what I know about it it seems like it will offer a very distinct experience from the other branches.

[mumra]

Having said that, we will need somewhere to stash our items if it gets removed.

This got me thinking so I wrote a new thread.

* Spider branch - No, no, no, no no no noooooo. After the long trip through the lair, and swamp, I can't think of anything worse than yet another uninspired branch that requires rP. Add to this teleporting creatures like jumping spiders and you've got an exercise in tedium. Although there are a lot of spiders in the game I find the whole theme of this branch to be really unimaginative and too similar to Lair. There are so many quality branch ideas that will provide a truly unique experience and theme, but spider is not one of them. God no.

Well, the plans haven't exactly been secret, and there's been plenty of time to raise specific objections on the wiki. The branch is half-implemented already and has plenty of developer momentum behind it.

I certainly agree that there are some excellent branch ideas on the wiki. Part of the focus of this thread is about raising ideas you'd like to see considered for future implementation; so perhaps you could let people know exactly which branch ideas you think particularly deserve looking at?

[oiseaux]

Well, the plans haven't exactly been secret, and there's been plenty of time to raise specific objections on the wiki. The branch is half-implemented already and has plenty of developer momentum behind it.

I certainly agree that there are some excellent branch ideas on the wiki. Part of the focus of this thread is about raising ideas you'd like to see considered for future implementation; so perhaps you could let people know exactly which branch ideas you think particularly deserve looking at?

Sorry, I overreacted a bit there. I'm just very surprised that the idea got so much support considering it is just aggravating one of the worst design flaws in the game - that in the early-mid game, if you don't have rPois, your choices are severely limited. In addition to the fact that it's a natural creature branch with a very limited diversity of monster types, I just think it's a shame that more outlandish and creative ideas are being overlooked.

Two ideas that I can see on the wiki that I would prefer:

https://crawl.develz.org/wiki/doku.php? ... ting_rocks
https://crawl.develz.org/wiki/doku.php? ... len_heroes

And here's something that I've been thinking of recently, it's essentially a ghost and sInv branch.

https://crawl.develz.org/wiki/doku.php? ... mnal_ruins

[mumra]

Sorry, I overreacted a bit there. I'm just very surprised that the idea got so much support considering it is just aggravating one of the worst design flaws in the game - that in the early-mid game, if you don't have rPois, your choices are severely limited. In addition to the fact that it's a natural creature branch with a very limited diversity of monster types, I just think it's a shame that more outlandish and creative ideas are being overlooked.

There's an idea floating around to ease up the rPois problem in spider with a rotting consumable (e.g. spider eggs) than grants temporary rPois. Spider is intended to have T&D be more relevant. Also, there's a view that limited diversity actually helps define a branch (e.g. slime - how many monster types does that have?)

The problem with very outlandish and creative ideas is that they are far more difficult to design, code and balance (although the Hellspider is certainly somewhat outlandish!) so I can see why they might get left to one side in favour of more easily-obtainable

https://crawl.develz.org/wiki/doku.php? ... ting_rocks

Floating rocks is a very cool concept - perhaps the stated goal of replacing Vaults makes it a bit hard to contemplate; I'd love to see it at least as a portal.

https://crawl.develz.org/wiki/doku.php? ... len_heroes

Dwarf branch idea has certainly been around for a while, and (as evident from that wiki page) an amount of work has been done already on it. I'm not sure why it's not actively being worked on, but my own feelings are:
a) Without something to distinctly differentiate it, it's very close to NetHack
b) The difficulty might actually be quite a bit harder than Elf, possibly it should be later in the game
c) It's perhaps a bit 'obvious' having an Elven Halls / Dwarven Halls rotation

Dwarves are definitely a significant race in Crawl and it's a shame no work has happened for a while in giving them some airtime. I have my own somewhat different proposal for a dwarven factory / forge branch which I've been meaning to write up for a while (and am about to get around to...)

https://crawl.develz.org/wiki/doku.php? ... mnal_ruins

I really like the story behind the idea, it conjures a nice picture; where I think it might meet resistance is being "yet another undead branch". However - I don't know if you read my idea for Stash Branches; I could certainly see it working as a very nice theme for a stash map. Also, it might make nice use of the new web trap which has been added for spider (especially since incorporeal monsters can pass freely through webs).

[dpeg]

oiseaux: I have no idea why you don't (try to) trust in our will to make the new branch diverse. We postponed Shoals until we felt they're good enough, the same will happen with Spider.

Some facts you might not know (but you should have expected something along these lines):
1) Branch roulette will probably be Spider/Snake, Shoals/Swamp. This takes some pressure on the rP front.
2) There are plans for Spider to provide a limited rP item.
3) Spider will feature webs, making T&D and Str more relevant than elsewhere in the Lair system.
4) The Hellspider will be a unique boss fight.

I don't think it will come as a surprise that Spider, as a Lair subbranch, features natural monster. Note that the Spider branch jibes well with spider form. In the future, there could be a Spider God.

I find it strange that you think Crawl developers overlook outlandish ideas. We have a Slow God, a Slime God, a non-healing species, etc.

Note that new branches are a LOT of work. You want a new layout algorithm, an at least partially new monster set and a clear idea of how it is different in threat. There are interesting ideas on the wiki, but these make up maybe 5% of the work. And no, the usual "some details are left to be settled" will not do it. Shoals were done because some coders like the idea enough; Spider is done because one developer was motivated enough to start the Spider portal vaults, and another one was motivated enough to do a lot of work on layout, webs etc.
I keep saying it, but the repetition seems to be worth it: a new branch is considerably more work than a new god or a new species. If you want to push a particular branch idea, try to make a portal vault for it. You can start with one or more fixed maps. If it needs special rules, try to use lua; ask for help on ##crawl-dev. This can really go a long way: if players and developers can see the value of your branch idea in actual games, this may be exactly what it takes to motivate one to go the next step.

[oiseaux]

oiseaux: I have no idea why you don't (try to) trust in our will to make the new branch diverse. We postponed Shoals until we felt they're good enough, the same will happen with Spider.

Some facts you might not know (but you should have expected something along these lines):
1) Branch roulette will probably be Spider/Snake, Shoals/Swamp. This takes some pressure on the rP front.
2) There are plans for Spider to provide a limited rP item.
3) Spider will feature webs, making T&D and Str more relevant than elsewhere in the Lair system.
4) The Hellspider will be a unique boss fight.

I don't think it will come as a surprise that Spider, as a Lair subbranch, features natural monster. Note that the Spider branch jibes well with spider form. In the future, there could be a Spider God.

I find it strange that you think Crawl developers overlook outlandish ideas. We have a Slow God, a Slime God, a non-healing species, etc.

Note that new branches are a LOT of work. You want a new layout algorithm, an at least partially new monster set and a clear idea of how it is different in threat. There are interesting ideas on the wiki, but these make up maybe 5% of the work. And no, the usual "some details are left to be settled" will not do it. Shoals were done because some coders like the idea enough; Spider is done because one developer was motivated enough to start the Spider portal vaults, and another one was motivated enough to do a lot of work on layout, webs etc.
I keep saying it, but the repetition seems to be worth it: a new branch is considerably more work than a new god or a new species. If you want to push a particular branch idea, try to make a portal vault for it. You can start with one or more fixed maps. If it needs special rules, try to use lua; ask for help on ##crawl-dev. This can really go a long way: if players and developers can see the value of your branch idea in actual games, this may be exactly what it takes to motivate one to go the next step.

Again, I must apologise for disparaging the Spider branch having not played it or allowed it time to be tested and balanced. I do not mean to belittle anyone's hard work on this fantastic game. On top of this, if Spider is part of the Lair roulette then I can only support it -- one of the main things about Crawl is that each game should be different and the player learns to adapt with the cards he or she has been dealt. Adding another roulette branch means more diversity but also means the branch isn't a requirement for every game. I hope to see this happen more in the future - like, for instance, the Dwarf Halls being a replacement for Elf, etc.

I appreciate that just throwing down an idea and saying "why don't you do this!" is possibly the most unconstructive thing, so I think offering up a portal vaul as per your reccomendation is probably the best option. Having said that, since you've seen the wiki page, do you think most of the ideas can be achieved via lua? I don't know at this point if it's sufficient to alter monster behaviour or produce some of the effects/features proposed, or if they'd have to be hardcoded. But if a really solid, well designed portal vault within the limitations of the system will go a long way, then great. Particuarly, I would like to know what would be required for a floor creation algorithm-- do you want an informal psudeo-code implementation, or could you base it off a well-designed bunch of portal vault floors?

In fact, I should stop de-railing this thread and raise one of my own. Again, apologies for the negative attitude, it was out of line.

[ldierk]

My Top 3 consists of interface implementables

Items under plants Most annoying because it always takes quite some time to get what's happening.
Traps in the way Not that annoying because a quick look on the overview map often reveals the trap.
More verbose travel stop messages Somewhat a generalization of the first point.


Keep up the good work. 0.9 is fantastic release!

[mumra]

Dwarves are definitely a significant race in Crawl and it's a shame no work has happened for a while in giving them some airtime. I have my own somewhat different proposal for a dwarven factory / forge branch which I've been meaning to write up for a while (and am about to get around to...)

Presenting: Dwarven Factory branch

[Roderic]

Here's my contribution:

1. New plants.
They have a lot of potential (even treefolk! "7" ) and can add new unique effects. Also eventually could be used to improve Fedhas late-game. Besides, forest vaults, swamp and lair can benefit in diversity and risk from them and make Fedhas more attractive to be worshipped.
2. Improve demigod: abstract worshippers.
Demigods deserve it and if you don't like the idea you still can go with any other atheist character.
3. Revive Arcane Marksmen.
They would be a good background for some magic affine races but they need a good book and also see 6.
4. A new god.
Only one it's enough for now, probably the door themed god(s).
5. Spider branch and demote hive to portal.
Better than the hive for a bug themed branch. Hive is plain and dull. Hive is good as a portal to gain a very nutritious food as a reward.
6. Improve Hex school and/or Improve poison school
Add high level new ones. For poison mages: acid/corrosion/sickness.
7. Learning spells from Vehumet .
Instead of receiving books like Sif Munnites.
8. Improve the melee combat.
a) New fighting techniques when having high UC and b) new text to gain richness in descriptions. The a) is ambitious but really interesting for many backgrounds yet b) it's easier to implement.
9. Q for kenkus! Diversify them to be found in D and used more in vaults. Damn, I'm gonna post something sooner or later about it.
10. Add a new slow race.
Golem / treefolk are good candidates but I don't complain if they are not present in next implementations. We have a lot of races now.
11. Improving Lugonu Make she more attractive by the ability of corrupt altars. It would be interesting if having the chance to be tempted to the dark side if some altar appear in the dungeon suddenly

[Danakh]

This is not on the list but i would like to see some kind of perma buff to ease playing hybrid.
Forum topic : viewtopic.php?f=8&t=1873
Wiki : https://crawl.develz.org/wiki/doku.php? ... r_duration

[Marten]

Apart from everything which was mentioned earlier, I strongly believe, that a huge emphasis in the future releases should be based on improving the variability of monsters/branches.

There have already been suggestions for Spider's Nest / Forest / Dwarven Factory, but I wouldn't really mind if these concepts did not make it into a whole branch, but just another random features of the usual Dungeon set such as:

Dwarves - with the highest point of interest being one or two Dwarven Artificers per pack, who would use a Rod as their main arsenal
Spriggans - so far I've never seen a single Spriggan Rider other than when being Misled or a usual Spriggan when it was a Shifter
Thieves - aka. characters likely to use needles other than Posion/Curare, since no others do it that way yet; they would be preferably human and act in a way similar to Maurice unique

As I have already proposed in another thread, I do believe all these concepts could be fused into a single character-specific branch which would be selected according to the highest attribute upon starting the game (Str=Factory; Int=Grove; Dex=Thieves' Den), where the players would be more-or-less guaranteed to find items to complement their playstyle. Perhaps as a replacement for Hall Of Blades (since it kinda sux as a branch concept imo)?

Also, Hive has a lot of unexplored potential, which was already mentioned a lot of times, but I do believe, that the same goes for Pandemonium, since it gets incredibly boring after tens of levels - new vaults with their own specific denizens perhaps?

Anyway, these are some of my most favourite other suggestions, which have been explained by others:
-a bit of an overhaul for current gods rather than adding new ones (especially Okawaru, maybe Fedhas,...)
- -MR from unsuccesful Hexes/Wands with increasing magnitude
-poison magic overhaul to include acid or any other highlevel spells
-new highlevel, lategame spells for schools which lack them
-hell effect overhaul
-permabuffs
-turn Passage of Golubria into a scroll, make portals permanent an unenterable by monsters, so it can function as a stash (town) portal

[mumra]

There have already been suggestions for Spider's Nest / Forest / Dwarven Factory, but I wouldn't really mind if these concepts did not make it into a whole branch, but just another random features of the usual Dungeon set such as:

There has been the concept of "special levels" of the dungeon with themed environments and monster sets (this is a possibility I proposed for Factory).

As I have already proposed in another thread, I do believe all these concepts could be fused into a single character-specific branch which would be selected according to the highest attribute upon starting the game (Str=Factory; Int=Grove; Dex=Thieves' Den), where the players would be more-or-less guaranteed to find items to complement their playstyle. Perhaps as a replacement for Hall Of Blades (since it kinda sux as a branch concept imo)?

It might be more interesting to intentionally select a branch or theme not complementing the player's style; otherwise you're just making it easy for them!

make portals permanent an unenterable by monsters, so it can function as a stash (town) portal [/b]

I mentioned elsewhere - portals are entirely located in distinctly inconvenient places to make handy stashes Besides which, the timed portals serve a purpose in forcing the player to act quickly. However, see my suggestion for Stash Branches which are designed for exactly the purpose you're after...

[galehar]

This is not on the list but i would like to see some kind of perma buff to ease playing hybrid.
Forum topic : viewtopic.php?f=8&t=1873
Wiki : https://crawl.develz.org/wiki/doku.php? ... r_duration

I've updated the 0.10 plan with lot of stuff from my todo, permabuff is one of them. I moved the wiki page and updated the OP link. And I also updated the plan for permabuff.

[Marten]

It might be more interesting to intentionally select a branch or theme not complementing the player's style; otherwise you're just making it easy for them!

My point was that I though Blades was intended for players to find the weapons that need, so I was just elaborating on it further. Plus I don't think additional needles for sneak chars and rods for heavy armour chars is that much of a help if they already prove worthy in such a combat, which shouldn't really be by any means easy. I dunno how true it is, but I think I've seen some posts showing effort to make needles more resist-able and there already is a slight nerf to rods for chars with low Evocations.

[galehar]

there already is a slight nerf to rods for chars with low Evocations.

Not really. The Evocations 0.9 unskilled change is a nerf to wands with low evocations. Rods haven't been touched, they already had low power. And power of both is significantly increased with high skill.

[mumra]

My point was that I though Blades was intended for players to find the weapons that need, so I was just elaborating on it further. Plus I don't think additional needles for sneak chars and rods for heavy armour chars is that much of a help if they already prove worthy in such a combat, which shouldn't really be by any means easy. I dunno how true it is, but I think I've seen some posts showing effort to make needles more resist-able and there already is a slight nerf to rods for chars with low Evocations.

I didn't mean in terms of items rewarded; rather, that a Str-based branch is a more even match for a Str-based character. Whereas a magic-based branch is a challenge for a Str character to deal with. Rock, paper, scissors ...

[Strongpoint]

1. Random gods.
It will lead to unique characters and it's cool

2. Weapon perks and special moves
IMHO, casters get more love from the developers

3. New mixed school spells
It would be cool if any 2 schools had at least one common spell

4. Make spell skills influence mixed spells differently.
For example mephitic cloud with 6 conj, 6 poison, 6 air and mephitic cloud with poison 12 , air 3 and conj 3 should be pretty different

[galehar]

None of this is planned for the near future. This thread is to vote for features which have already been approved and are likely to go in the next release.

[Strongpoint]

Ohh, my mistake then But I think same long term topic would be interesting if not useful
So it's goal to make something appear few days\weaks earlier? Then directly from 0.10 plan:

1. Casting a buff spell with Z makes it "permanent"
2. Skill fuzzing
3. New spells for Arcane Marksmen
4. Allow player in the arena. Give it a save file and the player is controlled with autofight. Will help with future fighting balance

[galehar]

So it's goal to make something appear few days\weaks earlier?

The goal is to help us decide which feature to give priority to. Among all the things planned, not all of them will make it in 0.10.

[XuaXua]

I will list a few that I want that I know might not be on the list, or are too specific on the list, or are not even considered:

1. FUZZING

2. Consistency across interrupted activities

3. Some sort of revision/generalization of Enchant Scrolls/Vorpalize Weapon.

4. Implicit weapon abilities / multi-brand (reaching / returning being the primary offenders)

Also, this has been done; does it get wiped from the wiki or does it go to an implemented page, or what?

[XuaXua]

4) Timed portals overhaul. Maybe set a trigger when a portal in seen, not just present on the the level (but maybe the spider entry bug is OK now)

I'm a big proponent of timers being based on some combination of experience gained + items dropped/picked up since view rather than time detection. Of course, still doesn't help much with shafts.

[galehar]

Also, this has been done; does it get wiped from the wiki or does it go to an implemented page, or what?

It's a wiki, anyone can edit it. I don't think there's anything to be gained from maintaining an implemented page. We can really use some help with the dev wiki, so when you spot some obsolete stuff, don't be shy and delete it.

[XuaXua]

Also, this has been done; does it get wiped from the wiki or does it go to an implemented page, or what?

It's a wiki, anyone can edit it. I don't think there's anything to be gained from maintaining an implemented page. We can really use some help with the dev wiki, so when you spot some obsolete stuff, don't be shy and delete it.

Done.

VOTE:
I would also like to see

1 - Dump Spell Hunger requirements added to the summaries.

2 - Color Classification added to help documentation for items and spells.

[dpeg]

Many thanks for the feedback, it is very interesting to read. It may be useful to highlight the interface bits that seem particularly important, perhaps we should hand out dedicated Implementables for them.

Regarding timers: the system has been changed to be much safer, including a two step timer.

[galehar]

Regarding timers: the system has been changed to be much safer, including a two step timer.

The current behaviour creates an optimal exploration pattern (fully explore a level before moving on). I think it's not good (well, it's not terrible either).
(discussed in mantis #4012).
(sorry for off-topic).

[thenewflesh]

My vote goes for the fighting cleanup. It's not very exciting task and won't immediately be noticed by us players, but it will make adding new exciting things in the future much easier. I think this is the most useful thing out there.

As for the actual new features, I like improving Demigods (abstract invocations, etc.:that will make them less dull for the lack of better word) and adding in-game selection of auto-pickup items (no need for justifications, really, it's just very nice convenience).

[Saegor]

ingame learndb access 1,71M of usefull information must be accessible ingame (shortcut: ??? (triple-?))

[dpeg]

Thanks for input, some comments/questions:

Marten: There are ideas about Okawaru, but what would you like to see changed about Fedhas? (Note that current gods, just like current monsters, spells, species, backgrounds etc. get adapted all the time.)

Marten: If you like the hell effect overhaul, try (a) to keep it alive in the forum by adding new ideas or making a summary and (b) making a wiki page for it. It is worth it!

Strongpoint: daftfad and I once started to discuss how random gods would work. I think it's a concept that could actually work, but it needs a lot of effort spent on design (more than on code, probably): you want piety rules, passive effects, active abilities and potentially a conduct. This should be set up randomly, while still having a theme and not being grossly overpowered. If you really like this idea, start thinking about it and discuss ways how to get here.

Strongpoint: If everything goes well, then 0.10 will feature a prototype of a combat move, for constriction (naga, octopode only). We will see how that works out, if it's interesting and good, there should be more to come.

minmay: Thanks for the list. I think you're often very harsh in your assessment, but it is almost always instructive.

Several: Demigods have got some actual momentum, with mumra uploading patches! I think part of the motivation may have come from this very thread, which is great to see. Please note that the idea is not "abstract invocations" or "divine powers", so please don't expect something that won't come.

Everyone: If you think a particular interface idea (currently buried in the list or not even listed) needs more promotion or should Really Get In, tell us. (For example on ##crawl-dev.) We can set up a dedicated Implementable for it like I did for the uniques/ghost annotations which has been coded already by Zebu -- many thanks again.

[betamin]

Count my votes for:

The hive elimination/revamp.

Poison magic changes

Combat moves.

Some new god and/or species.

An Okawaru modification.

[Zzz]

Do away with hive. There's already bee vaults which fits the theme nicely. And food shops.
Forest branch? I'd like to see a new branch between vaults and d27.


...whisper(mouse support for webtiles)whisper