Skip to content

0.16 Development and Release Plan

Edgar A. Bering IV edited this page Dec 12, 2021 · 1 revision

====== 0.16 Development and Release Plan ====== The 0.16 and the follow-up bugfix 0.16.1 have been released.

=====Release Planning=====

==== Binaries and Packages ====
  • ** <del>OS X</del> ** --- geekosaur is working on setting up a 10.8 VS so we can compile with the 10.7 SDK and support 10.7+. __Done! geekosaur compiled OS X binaries on a 10.8 VS and kindly registered a dev account to sign the release.__
  • ** <del>Windows</del> ** --- It seems we can't build the Windows version with LTO with GCC 4.8. Napkin can cross-compile for Windows (and has automatic Windows builds set up already), so he will likely compile the Windows zips/installer. __We make 32-bit and 64-bit zips for windows without LTO. In the future we may use the i686-w64-mingw32 (32-bit) target only and release zips and the installer.__
  • ** <del>Debian</del> ** --- (gammafunk) I'm able to compile the deb packages with LTO on ubuntu, and current plan is to use these for release to support ubuntu, Debian testing|sid, and other recent Debian distos. Looks like we can't support Debian wheezy because of libsdl2-image. __For 0.16.0 we built against ubuntu utopic, but this wasn't very compatible for other debian OS, so for 0.16.1 I disabled LTO and built against ubuntu trusty.__
==== Tournament ====
  • ** <del>Confirm final Date</del> ** --- Tournament 20:00 UTC Mar 13 to 20:00 UTC Mar 29
  • ** <del>Any final rules tweaks</del> ** --- Draft Rules: http://dobrazupa.org/tournament/0.16/
  • ** <del>Install 0.16 on all participating servers</del> ** --- __All servers were updated to 0.16.0 and eventually 0.16.1.__
  • ** <del>Updated tourney page style</del> ** --- (chequers) __We didn't have time to use this for 0.16 (, sadly), but can do so for 0.17.__

==== New Start Page ==== New landing page is completed. Release instructions (for Napkin):

  • git clone clone git@github.com:alexjurkiewicz/dcss-website.git /path/to/webroot
  • run from cron every minute: ./dgl-status-collect.py /var/www/servers.json /var/www/dgl-status.json /tmp/dgl-status-collect.lock
  • Update download.htm's links when 0.16 launches for real
==== Final Release Steps ====
  • ** <del>Wordpress announcement</del> ** __Done!__
  • ** <del>Final 0.16.0 changelog</del> ** __Done!__
  • ** <del>Push release tag for 0.16.0 </del> ** __Done!__
  • ** <del>Upload binaries and source packages to SourceForge</del> ** __For this release we hosted binaries and the full source package on cdo, with the nodep source packages linked to automatically built ones from github.__
  • ** <del>Update CDO download links</del> (Napkin)** __(gammafunk) Binaries uploaded with the new download page from chequers.__

===== 0.16 Development Planning =====

This page is for managing short and middle term development planning. Not everything that's on here is going to get into 0.16, or perhaps any release at all - these are just a list of changes that are worth //considering// for the near future.

Right now I'm tossing the kitchen sink in - we can move stuff that doesn't seem immediately practical into long-term planning, as it seems appropriate.

See the long_term_plans and possibly 0.15_plan for ideas, moving said entries here as necessary.

==== Character ====
  • Attributes --- consider halving their absolute values to make them more meaningful (cf. IRC discussions). __Not going in 0.16; Plans not specific at this point__
  • <del>Eating</del> --- no more chunk eating for most species (can compensate by spawning more permafood and/or give permafood more nutrition), species with strange diets exempt (Troll, Ghoul, Vampire). //chunkless// Alternatively, making chunk interface better (enable some options by default? not splitting chunks by type, age?) have been suggested to keep more similarity with current gameplay. __Chunks we're slurried in 0.16, but any further food rework in 0.17+ will require some careful thought. Minor tweaks could include merely reducing the amount of permafood.__
==== Combat ====
  • SquareLOS --- there don't seem to be any developers currently on the team who oppose this. perhaps it's time to start investigating it again? __this is not going in 0.16, sadly.__
  • GDR reform --- an invisible number. if it's significant, it should be exposed to the player in some way. ("plate armour provides somewhat more protection against melee attacks than a choko"...?) It might be nice to simply associate it directly with AC, or remove it altogether, but there are significant balance considerations involved. __this is not going in 0.16, sadly.__
  • <del>Unarmed combat heavy armour penalties --- should these exist? If they should, they probably should be de-randomized.</del> __Done!__
  • Acid tweaks --- make the slay & ac penalties for corrosion stacks the same number? (current situation causes confusion.) Decrease chance for further corrosion stacks based on the number the player already has? Make rCorr halve the effects of corrosion, rather than halving the chance? (Perhaps something similar for rPois...?) Make acid stacks wear off one-by-one, rather than all at once? (doy) __No further acid tweaks (heh) going in 0.16, but former ideas possible for 0.17__
    • Decreasing chance for corrosion stacks based on current count is in, though probably the exact numbers need tweaking.
  • <del>Weapon brand distribution cleanup --- //reaverb has ongoing work on this.//</del> __And it's done!__
  • Shield skill requirements --- I think I'd like to see these become stepdowned or something of the sort, rather than hard breakpoints on the penalties, the way armour strength requirements were changed. --wheals --- seconded -PF __this is not going in 0.16, sadly.__
  • Banishment --- should take severity/power of the banishment source into account: early banishment (distortion weapon on D:4) should send characters to Abyss:1, Elf:3 banishment to Abyss:3. Can either use absdepth of the banishment source, or player XL, or banisher HD, or perhaps another scale. [Rationale: Abyss:1 is an interesting challenge for early game, but falls flat later on; the Abyss levels work well to distinguish between escape and rune hunt; the only missing piece is sending different characters to different levels). __this is not going in 0.16, sadly.__
==== Interface ====
==== Items ====
  • Unlimited ranged weapon ammo --- Should basic ammo for launchers be made unlimited? Primarily affects the early game - big buff for arkane markspeople. __Not going in 0.16__
  • Automulching for all ammo --- Make all ammo mulch on hit, multiply generated quantities? Makes playing hunter involve less tedious running around after ammo, but lots of side effects, not all desirable... __Not going in 0.16__
  • Rethink ammo/launcher brands --- interactions between launcher & ammo brands are complex & unintuitive (which brands stack? which conflict?). Possibly launcher & ammo brands should be switched to non-overlapping sets (launchers -> strict buffs, ammo -> situational effects?), or perhaps either launchers or ammo should lose their brands entirely. __Not going in 0.16__
  • Strategic consumables --- move these out of the inventory? (ID game issues?) __No specific proposal/implementation yet, not going in 0.16.__
  • Spellbooks --- make these tactically relevant in some way (a reason to carry books around), or turn them into strategic items (per above)? __No specific plans yet, not going in 0.16__
  • Ground item interactions --- let players universally interact with (wield, read, quaff, etc) items on the ground? (remove annoyances from full inventories). __Not going in 0.16. Possible for 0.17 (MarvinPA): ground item interaction/menus similar to angband.__
  • <del>Auto-ID enemy-held weapon brands</del> --- let players respond to the threat of dangerous brands without getting hit by them repeatedly, allowing for more interesting tactical combats. (also possibly allows for adding more dangerous brands?) __Done!__
  • ?EW --- The removal of enchant weapon scroll failure was definitely a boost for player sanity, but has also resulted in significant weapon enchantment inflation - most players now seem to have +9 weapons before they enter the Depths. Reducing generation even further is one option, but I'd rather experiment with making higher enchantment levels cost more scrolls in a deterministic fashion. __Was discussed but nothing agreed upon, not going in 0.16__
  • <del>triple swords</del> __Done! thanks, doy__
  • <del>Unidentified wands</del> --- evoking these should use up more charges (say 1d4, with no effect if the wand doesn't have sufficient charges for all of them). Rationale: less trivial wand identifying, sometimes players may use ?id on early wands to save charges.
  • <del>Rings of regeneration</del> --- shift to amulet slot, to compete with guardian spirit/gourmand/faith and discourage swapping after every fight? __The ring of regen became the amulet of regen in 0.16.__
==== Gods ====
  • <del>Gozag</del> -- Fix/rethink bribe branch, duplicate, wrath. (//dpeg// for 0.16). __MarvinPA reworked G wrath and duplicate is disabled for now. Gozag will make 0.16 and further work on duplicate and other abilities will happen in 0.17.__
  • <del>Ideas for fixing Gozag Wrath from MarvinPA and PleasingFungus</del> __See above.__
    • Remove the thing where he takes part of the gold you find.
    • Remove the thing where he turns found gold into stones.
    • Let him send monsters at you in non-bribeable branches.
    • Replace blurry mouth with the curse of midas (probably only for potions, since scrolls are absuable in stupid ways).
    • Remove goldification on pickup.
  • <del>Ru</del> --- Test & balance. __Ru is considered tested and reasonably balanced, thanks to Lasty for the tremendous work.__
  • <del>Nemelex</del> --- Left in a half-fixed state; should probably be re-reworked. __Nem got a big infusion of new cards thanks to 78291 and tenofswords. Possible plans for 0.17 include further card balancing and rework and possibly moving deck usage to the ability menu, but these are not specific nor agreed upon proposals.__
  • Qazlal --- Wrath probably wants a second pass. <del>(particularly the 'lava' effects, which help the player in most situations...)</del>. __Temporary lava is gone; probably someone should review the rest at some point in 0.17.__
  • Beogh --- Reconsider early & late-game balance. (Probably should be made stronger in one of them - currently neither fish nor fowl.). \ //dpeg//: allow prayer over orc corpses to either give 20% chance for piety vs 80% for resurrection (allied orcs) or 80% chance for piety vs 20% chance for resurrection as an ally (hostile orcs). Resurrected orcs keep their name and "had a gift" tag (if any). The name is appended, such as "Beowulf the Revenant" etc. \ //dpeg//: A Beogh interface toggle that llows to switch between four modes for orc company: 0 (none), 3, 12, all. Regardless of option, hostile orcs keep converting. (Perhaps 3 should take best fighter, best wizard, best priest.) If an orc dies, implicit call for replacement to keep the number up. __These proposals are ideas for 0.17, but won't be in 0.16__
  • Lugonu --- Replace corpse piety with altar desecration piety (now that wraths are more gradual). Start small, i.e. low wrath and small piety gain. __Lucy is unchanged in 0.16. Further removal of corpse sacrifice will likely happen in 0.17, but the altar desecration needs agreement about details.__
  • Yredelmnul --- If abandoning the god, destroy all zombies (rather than making them hostile). __This didn't make the 0.16 feature freeze, but should be easy to do in 0.17.__
==== Monsters ====
  • <del>Spell slot rework --- casting frequencies, wizard/priest/natural flags.</del> done
  • Uniques --- more xp for many of them, higher incentive to kill by better loot. For numbers and more ideas see https://crawl.develz.org/tavern/viewtopic.php?f=8&t=11276 __This didn't make 0.16. No specific proposals yet.__
    • TODO: Possible further tweaks to unique branch spawns. (Boris, Xtahua esp.)
  • <del>MR multipliers --- remove in favor of explicit MR numbers (also round to only about 10 steps)</del> __Done.__
  • Acid --- acid should be made a proper second-class resistance (analogous to relec), with sources of acid damage in many branches (s.t. rCorr is useful in more places than Slime). Some work on this was done in 0.15, but there's still room for slightly more acid in Vaults and Depths, and significantly more in Extended. __Acid damage prevalence will be unchanged for 0.16.__
==== Spells ====
  • Transmutations --- revamp? New forms, effects (using permanent resources to power spells?); revamp/remove sticks to snakes, ignite poison, &c __Hydra form and irradiate were added in 0.16. No specific proposals for other form rework, or STS. Blade hands has been described as too strong, but there is no agreement nor any specific proposal for changing this spell.__
  • Charms --- Redesign Badcharms (stoneskin &c, rmsl &c). Standing 'least bad' option is probably 'mp/time' costs, but that would require redesign of elementalist backgrounds... 'effects removed on armour swap' seems clunky but might be viable for some charms. __No charms changes in 0.16. Highly likely that cTele (and -cTele) will be removed in 0.17. It's possible haste the spell will be removed, since it's seen as not necessary given the wand and potion, and complicates the move to using aut for score together with removing player turn tracking, but this is by no means agreed upon.__
  • Spellpower --- spellpower should matter much more for many spells. --- __No further spells were reworked to use spell power in 0.16.__
==== Tiles ====
  • <del>Floors</del> --- By and large most floors have been darkened, to improve visibility of tiles on top of them, and to establish something closer to a standard luminosity. The precise values should be tweaked, but I'd like to get as much specific feedback as possible. __Done.__
  • Deep Elf monster tiles --- Most elf casters are not differentiated by anything but color, and their colors are not consistent with other monster sets (green priests, purple wizards, and so on). There are visual motifs for classed monsters now, like a circular aura on the ground; applying these motifs and refining them will be an ongoing process. __0.17.__
  • Adding ancient tiles for player doll cosmetics --- There are a lot of tiles on Mantis (for example https://crawl.develz.org/mantis/view.php?id=3936) that look good and deserve to be in the game, even if they don't have a specific purpose. We can also consider adding these to the rotation for tiles that appear for worn equipment, but I consider that a separate issue. __0.17.__
  • Replacing tiles made by non-tile-devs --- As much as I love the current Anubis guard tile, it and its brethren (Bennu, Ushabti) could use improvement. Please add any other tile requests here. __0.17.__
==== ZotDef ====

<del>** Disable **</del>--- it's very broken in a large variety of ways and has been for a number of stable versions, so should be disabled unless somebody shows interest in diving in and fixing it. __Done.__

Clone this wiki locally