00:51:30 Fork (bcrawl) on crawl.kelbi.org updated to: 0.23-a0-4753-g3cc4eb4eda 03:23:11 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-4753-g3cc4eb4eda 05:59:00 -!- Yermak_ is now known as Yermak 07:04:09 -!- bairyn is now known as ByronJohnson 07:14:46 Discord| that's a good point for the range/magic chars vs melee! Giving the goblin summon abilities would help break LOS for certain spells and bolts. Smite targeted abilities could be a problem even with summons. If this spawns in a late dungeon or depths I'm not sure of good ways to keep god abilities from killing easily at range. I guess a lot 07:14:47 of HP? Are there monster resists/traits that can make them more resistant to smite targeted abilities? Could give the Goblin fog scrolls though I'm not overly familiar with how smart monster AI is to use them. 07:15:50 that has worse problems, like players that get pissy when they can't kill the monsters for their fog scrolls 07:16:52 lol never heard of that complaint before. 07:17:06 <10PleasingFungus> monsters can't currently use fog scrolls. 07:17:22 I suppose there's probably no way to ensure a monster is hard to kill for every kind of cheese or player build 07:17:27 <10PleasingFungus> yes. 07:17:31 well, more precisely it's playyers who get upset when monsters "waste" "loot" 07:18:01 ahh, yea def had a hobgolbin waste a pot of invis on me lol 07:18:23 <09hellmonk> This monster should be a unique, spawn a trove portal when killed and replace troves 07:18:27 <09hellmonk> imho 07:18:54 Yea thinking of it this morning, spawning a portal to a trove on death is how it would "Drop" it's loot 07:19:23 <10PleasingFungus> def agree re unique 07:27:18 <09hellmonk> Maurice rework / timeline alt? 07:30:37 Re: Unique I'd lean more towards S branches 3-4, Elf 3, Vaults, Depths and onward as spawn locations. If it spawns in Dungeon floors it'll either be impossible to kill or if it's killable in Dungeon floors it'll be a cake walk if it spawns lower. Could possibly have it "banish" itself to a lower dungeon floor if not killed in time (if that's doable 07:30:38 in crawl code) so instead of legit self banishment to Abyss, it "teleports" itself from D15 to like Depths:4 or something. 07:39:09 sounds very alt-Maurice… but also sounds a little too similar to Boris 07:40:08 or Prince Ribbit for that matter 07:43:58 Looking at uniques.des what do DEPTH: lines like this mean? `Vaults, !Vaults:$` to me that reads like Vaults and NOT Vaults which seems to contradict. 07:44:12 "Vaults:$" means the last level 07:44:17 got it 08:06:58 it might be worth reading through https://github.com/crawl/crawl/tree/master/crawl-ref/docs/develop/levels and in particular syntax.txt 08:13:28 thanks, will do! 08:16:32 <10PleasingFungus> in general, it’s good to write down your ideas, but not to force all of them into one design 08:16:36 <10PleasingFungus> elegance is good 08:17:48 <08nicolae> yeah, you have to be willing to cut ideas that don't work 08:18:31 or save them for a different design, maybe. putting too many good ideas onto the same monster/unique generally makes a bad monster/unique 08:18:44 <08nicolae> /dev/null is full of vaults that i had worked on for some time but couldn't get working and so i just cut them and saved the idea for later 08:19:07 <08nicolae> pan rune bazaar was originally an encompass vault but i couldn't think of how to fill out the rest of the level in an interesting way 08:20:07 <08nicolae> the trick i guess is to figure out the central idea and work with that 08:20:14 Sure, I get that. So far it doesn't seem like a crazy idea. Fast, evasive unique with summons on a self banishment timer that when killed spawns a trove vault portal. 08:20:31 the only part of that idea that doesn't already exist as a mechanic is the self banishment timer 08:21:03 <08nicolae> that might take some doing but probably doable 08:24:02 certainly, and I'm excited to try it! Self banishment is already an ability a-la Lugonu "Enter the Abyss" just have to create a mechanic in which the unique is "forced" to cast it after some time. But timers also exist on things like ice cave/volcano vaults so that's no new ground either. 08:25:08 well, there's a difference between removing a dungeon feature on a timer (easily done in lua) and making a monster cast something on a timer 08:25:30 monster stuff like that is generally based on HP (prince ribbit, bai suzhen, etc.) 08:25:31 agreed, I meant that timers in a general sense are not new ground 08:27:45 <08nicolae> why not fake it as a duration and when it expires it just says "so-and-so casts themself into the trove dimension!" 08:28:04 <08nicolae> like a really long enchantment basically 08:28:26 <08nicolae> although it would be fun to silence the monster when the timer goes off and it's just standing there like "what da fuck" 08:28:49 That's an idea! Silence as a strat to keep it around is also fun! 08:30:03 is the `Discord|` tag from some bot? Is this chat duplicated/mirrored in discord somewhere?> 08:30:11 <08nicolae> so my first guess: have a really long enchantment that's activated when you first see the trove unique, that way the game will keep track of the turn count for you even off-level (i think), and then when it expires, set a flag that says "timer's up, get the hell outta here" and then if the flag is true and the monster can cast, it casts the spell, that way you could list the spell in the monster description 08:30:19 <08nicolae> also yes there's a bot bridging the discord channel and the irc channel 08:31:38 most of the devs are on the discord side these days 08:31:43 Which Discord channel is that because that would be easier since I already have it up. 08:31:49 devs only 08:32:05 aww sad day 08:32:33 <08nicolae> i don't know offhand why the discord is devs only if the dev channel is bridged to a freely-available irc channel, but that's how it is 08:32:45 :shrug 08:33:07 ^ I didn't expect that to work lol but it'd be funny if the bot does it for me on the discord side 08:33:09 <08nicolae> only devs can be trusted with emoji reacts and quote replies 08:33:30 <08nicolae> sadly it did not, you need colons on both sides 🤷 08:33:40 depennding on irc client you can have that work 08:33:43 🤷 08:33:59 * geekosaur uses emoji.py for hexchat 08:34:19 <08nicolae> that showed up as a shrug on this end 08:34:20 <08nicolae> nice 08:34:51 03mgdelmonte02 07https://github.com/crawl/crawl/pull/2669 * 0.29-a0-1138-g351304271c: test-remove-freetype 10(28 seconds ago, 2 files, 0+ 4-) 13https://github.com/crawl/crawl/commit/351304271cf0 08:35:04 I should make a PR against it, I made it a bit more reliable/less likely to misfire 08:39:02 03advil02 07* 0.30-a0-3-gc9032a9cac: fix: unbreak change email/pw 10(2 minutes ago, 2 files, 22+ 22-) 13https://github.com/crawl/crawl/commit/c9032a9cacfc 08:39:08 <10PleasingFungus> iirc there are moderation concerns re discord 08:39:20 <06advil> I think it is dev+server admin only because that drastically simplifies thinking about moderation 08:39:20 <06advil> yes 08:39:52 <06advil> namely, we don't have to do it (on top of the minimal irc moderation that is necessary) 08:40:21 <08nicolae> ah, that'll do it 08:49:31 03mgdelmonte02 07https://github.com/crawl/crawl/pull/2670 * 0.27-a0-1063-ge22e23c99c: initial 10(27 hours ago, 1 file, 26+ 0-) 13https://github.com/crawl/crawl/commit/e22e23c99ce6 08:49:31 03mgdelmonte02 07https://github.com/crawl/crawl/pull/2670 * 0.29-a0-1095-g4a2f35b117: Merge branch 'feature/add-lua-stack' of https://github.com/mgdelmonte/crawl into feature/add-lua-stack 10(51 seconds ago, 0 files, 0+ 0-) 13https://github.com/crawl/crawl/commit/4a2f35b117b5 08:53:08 Unstable branch on crawl.kelbi.org updated to: 0.30-a0-3-gc9032a9cac (34) 08:58:50 <08nicolae> scroll of plants 09:11:58 Unstable branch on crawl.akrasiac.org updated to: 0.30-a0-3-gc9032a9cac (34) 09:15:49 <10PleasingFungus> hm 09:16:15 <10PleasingFungus> interesting 09:27:56 <05kate> I'm very wary of the idea of a monster/unique that's entirely built around fleeing, since we have a whole bunch of historical experience that fleeing monsters are mostly just pretty obnoxious 09:29:30 <05kate> and Maurice being the closest current comparison isn't a good sign to me since I think the only reason Maurice still gets to exist is historical inertia, he's not a design that would be likely to be added today 09:31:14 <05kate> (but if there's a way to make it play well somehow then definitely reworking/replacing Maurice sounds good) 09:31:28 Hmm, I see. Well this idea is from the D3 mechanic and that thing runs away but it also "takes a break" occasionally so you can catch up if you had to fight some mod to get to it. 09:31:28 My plan is to give it the ` { SPELL_BLINK_AWAY, 15, MON_SPELL_WIZARD | MON_SPELL_EMERGENCY },` spell like boulder beetles and then various summoning spells. The theory being it would summon things to defend itself then blink away in emergencies. 09:35:23 <08nicolae> i think the fact that it will time itself out eventually might make the fleeing unnecessary, it might prefer to stay at range and summon at you, but i don't think it needs to straight up avoid combat. you would basically have to decide "do i want to deal with it now or pass up the fabulous prizes" 09:35:27 <08nicolae> like a timed portal that hates you 09:35:43 <10PleasingFungus> kate: i agree re history, but also treasure goblins are great 09:36:18 <10PleasingFungus> and i wonder if the feeling is different when there’s just one fleeing thing in the game + it has a reward beyond xp 09:36:25 I gave it the `M_MAINTAIN_RANGE` property as well which I� _believe_ should keep it trying to stay at range and summon at you 09:36:39 <08nicolae> i believe that's what that property does, yep! 09:37:24 <08nicolae> also i know it's fun to drop references to other things you like, lord knows i've done it plenty, but sometimes you can trim bits off the edges that don't quite fit and come up with a new flavor 09:37:47 <10PleasingFungus> extremely good to do so, even 09:37:53 This is the first cut of it, haven't added the summon spells yet: 09:37:53 ``` 09:37:54 { 09:37:54 MONS_TREASURE_GOBLIN, 'g', COLOUR_UNDEF, "Treasure Goblin", //1 09:37:55 M_UNIQUE | M_WARM_BLOOD | M_SPEAKS | M_SEE_INVIS | M_UNBLINDABLE | M_GENDER_NEUTRAL 09:37:55 | M_FLIES | M_WEB_IMMUNE | M_FAST_REGEN | M_NO_EXP_GAIN | M_NO_POLY_TO | M_MAINTAIN_RANGE, //2 09:37:56 MR_RES_ELEC | MR_RES_POISON | MR_RES_FIRE | MR_RES_COLD | MR_RES_NEG | MR_RES_MIASMA | MR_RES_ACID 09:37:56 | MR_RES_TORMENT | MR_RES_PETRIFY | MR_RES_DAMNATION | MR_RES_STICKY_FLAME | MR_RES_VORTEX | MR_RES_STEAM, //3 09:37:57 0, MONS_GOBLIN, MONS_GOBLIN, MH_NATURAL, WILL_INVULN, //4 09:37:57 { AT_NO_ATK, AT_NO_ATK, AT_NO_ATK, AT_NO_ATK }, //5 09:37:58 0, 2000, //6 09:37:58 2, 12, MST_BLINKER, false, S_SCREAM, //7 09:37:59 I_HUMAN, HT_LAND, 20, DEFAULT_ENERGY, //8 09:37:59 MONUSE_OPEN_DOORS, SIZE_SMALL, MON_SHAPE_HUMANOID, //9 09:38:00 {TILEP_MONS_IJYB}, TILE_ERROR //10 09:38:00 }, 09:38:01 ``` 09:38:01 It's _probably_ too hard to kill like that but it's also a treasure trove so it shouldn't be easy 09:38:01 <10PleasingFungus> uh 09:38:09 That formatting might not transfer at all 09:38:13 <08nicolae> it did not 09:38:14 <10PleasingFungus> probably post as a link to a github gist 09:38:16 <10PleasingFungus> or similar 09:38:19 will do 09:40:22 <06advil> I wonder if we should have an explicit webtiles server changelog 09:40:34 <06advil> I've sort of been doing that in issues anyways 09:41:57 https://gist.github.com/Static-Flow/2ea4e6a3a7ba78bd24d4e9bb8d6816c7 09:44:31 Webtiles server stopped. 09:44:41 Webtiles server started. 09:45:07 Essentially, resist everything, infinite will, high health, summons baddies that are issues for both casters and melee dudes, with an emergency blink. No idea of AC/EV values since I'm not super well versed in how those calculations work. All that is open to changes, improvements. I'm not married to any of it. Just trying to make something that's 09:45:08 challenging to kill since it's dropping a treasure vault. 09:45:39 <08nicolae> i don't know much about the monster defs but off the top of my head: if you're not planning on having it generate in Spider it probably doesn't need M_WEB_IMMUNE. it probably doesn't need all those resistances, there aren't very many player sources of damnation and vortex, and even if it's supposed to be hard to kill it is nice to have a trick up your sleeve. 09:46:33 <08nicolae> what areas of the game are you expecting to have it generate in? i'm guessing the ones where troves already generate 09:46:52 i updated the gist with the uniques.des 09:47:08 `DEPTH: Depths, Elf:3, Swamp:3-, Snake:3-, Shoals:3-, Spider:3-, Crypt, Vaults, Zot, Pan` 09:47:25 also res_torment and damnation and neg energy is generally reserved for demons or followers of yred 09:49:02 noted. Removed torment, dam, neg, and vortex 09:49:30 <08nicolae> that's a broad range of places, most uniques try to stay in a narrower range of placement: a unique who's going to be a reasonable threat in pan is going to be an absolute murder machine in the s branches, and a unique who's well-placed for the s branches is going to be a pushover in pan 09:51:12 Hmm, good point. Was trying to play to the "Randomness" that a treasure goblin could appear anywhere (again borrowing from Diablo3). With this many resists and abilities, S branches are probably too low 09:52:18 <08nicolae> like i said, you kind of have to tweak the original concept to fit the framework of a different game. i guess the better question is: what part of the game is a player most likely to want to hit up a trove, and would be willing to chase down a walking trove portal to do so? my first guess would be "where the troves are generated normally" 09:52:49 you'd need something like the mechanic you mentioned earlier, modified: its abilities scale with the branch, but if it can escape from you it respawns elsewhere with abilities scaled to the new location. which I think would be ard 09:52:52 *hard 09:53:08 otherwise it goes from being hard in s branches to a pushover in pan 09:53:18 Right because troves now can spawn even in low Dungeon floors 09:53:55 you might look at boris, who can spawn in a bunch of places, but they're all of equivalent difficulty 09:54:16 Boris depth is `DEPTH: Depths:3-, Vaults:2-, Crypt, Tomb` 09:54:35 hm, thought he could shoiw in a few more places 09:54:40 <08nicolae> well, tomb might be an outlier there 09:54:49 <08nicolae> also iirc he's special-coded to maybe generate on the orb run? c/d 09:55:17 <08nicolae> OH, right, he can regenerate on any floor after you kill him 09:55:23 I let out tomb because _in my experience_ tomb is hard enough as is and you probably are doing it at the end when troves are useless 09:55:25 <08nicolae> i guess those levels are where he shows up originally 09:55:31 ah, that's what I was thinking of 09:55:34 <08nicolae> you can tell how often i encounter boris, 09:56:15 so maybe this for TG? `Depths:3-, Vault:3-, Crypt, Zot` 10:01:35 Pan probably isn't useful because from my _limited_ extended experience you already have your kit set if you're doing pan. But a lucky TG trove in Crypt, Zot could give you the pieces you need to attempt extended. 10:02:11 sounds right to me 10:02:42 New branch created: pull/2675 (1 commit) 13https://github.com/crawl/crawl/pull/2675 10:02:42 03mgdelmonte02 07https://github.com/crawl/crawl/pull/2675 * 0.30-a0-4-gb5ad3a8f2f: Adds lua function view.get_map, which returns the entire known map. 10(2 minutes ago, 2 files, 104+ 0-) 13https://github.com/crawl/crawl/commit/b5ad3a8f2ffb 10:02:45 I think getting a trove on an emphemeral level like Pan would be mildly extremely exasperating 10:02:56 although that does make me wonder, what about a player who encounters a tg when they don't intend extended 10:02:58 lol 10:03:53 <10PleasingFungus> pinkbeast: i think the context is that the only toll is ‘kill the treasure goblin’? 10:04:05 geekosaur I imagine you'd just ignore it and let it banish itself. 10:05:03 but that's also the downside of summon spells vs just run away. If you happen to spot it, but don't want to deal with it, you kind of _have_ to because it's spawning demons at you. 10:06:34 <08nicolae> well, you can just run away long enough for its timer to go off and then "bzoop" no more problem 10:07:18 true, and summons are on their own timers right so it's not like they'll stick around forever either. 10:07:50 wonder if its summons can spawn patrolling 10:08:14 (or if that would even help) 10:09:56 is the frequency in a spell like `{ SPELL_HASTE, 30, MON_SPELL_WIZARD },` how frequently the mob has that spell or how frequently they cast it? 10:10:16 cast frequency 10:10:26 <09hellmonk> How frequently they cast it, out of 200 10:10:39 noted 10:10:55 spellbooks control how often they have it, although I think that option was actually removed some time back (no more alternate spellbooks) 10:10:58 <09hellmonk> (slightly more complex than that but 200 is "always tries to cast it" 10:12:08 so maybe we could lower the summon cast chance so the TG would only cast it the longer you spent time around it. Essentially favoring casting blink_away or haste over summons. 10:13:19 but again this might not be a problem at all if just walking away from it it solves any issues of "i don't want the treasure goblin loot" 10:58:31 for uniques.des is a higher weight _more_ likely to spawn or less? 11:40:34 <10PleasingFungus> more 11:41:43 makes sense 12:19:49 interesting question, looking through trove gen code (and from experience) you can get useless troves. Would it make sense to keep that same mechanic or would you expect killing the loot goblin to _always_ provide something useful (like oka/trog gifts)? To steal from Diablo 3 it's a toss up on what you might get. I'm not sure what the player base 12:19:49 would _expect_ from killing something like this. 12:20:07 <10PleasingFungus> what do you mean by 'useless'? 12:20:25 Like you're a melee char and the trove is a library and vise versa 12:20:44 that's the "Gamble" with troves is you might give it 10 tele scrolls and not get anything useful inside 12:22:44 but with a Treasure Goblin players are risking a splat trying to kill it and if they did all that and didn't get anything from it that could be frustrating so should TG generate a trove that is always focused towards them. So while you might not need the books/armor it's because you have better stuff not because it's not needed for your build. 12:25:12 In D3 killing a TG always nets you some gold but its def possible to kill it and not get any gear worth using. The difference there is you can find multiple TG and also replay the same char multiple times which you can do neither of in this version which makes me lean towards skewing the trove to be "useful" loot for the player. 12:28:32 <09gammafunk> @ebering am I clear to start setting up t website today or did you want to work on that 12:33:00 <12ebering> Go for it 12:33:05 <12ebering> Rip time to make any changes 12:33:15 <12ebering> Do any you feel good about 12:38:00 <09gammafunk> I'll do setup for now and we can mull over the circulated suggestions of banning Me for relevant banners and possibly for the combo category bonus 12:38:02 <10PleasingFungus> a library for a melee char is a great incentive to consider hybridizing 12:38:08 <10PleasingFungus> likewise a great weapon for a 'caster' 12:38:23 <10PleasingFungus> crawl is a game about adapting to what you find 12:39:38 <10PleasingFungus> the character isn't trading anything for this trove, so if anything, the bar for 'usefulness' should be lower 12:40:41 <10PleasingFungus> you may have noticed that other parts of dcss have a fairly broad definition of 'useful' - excluding stuff like helmets for felids, but including books for """melee chars""", etc. it's worth taking some time to understand why before you try to do something drastically different 12:41:22 The `nicolae_trove_salty` trove might do well then 12:41:22 `KITEM: d = acquire weapon good_item / acquire armour good_item / \ 12:41:23 any jewellery randart w:5 / randbook w:5` good mix of everything 12:41:25 ugh 12:41:36 `KITEM: d = acquire weapon good_item / acquire armour good_item � any jewellery randart w:5 / randbook w:5` 12:44:01 `trove_garden` trove could also work 12:44:15 that one has potions as well 12:53:30 New branch created: pull/2676 (3 commits) 13https://github.com/crawl/crawl/pull/2676 12:53:30 03mgdelmonte02 07https://github.com/crawl/crawl/pull/2676 * 0.30-a0-4-g25d6f42ac8: Adds more info to the lua `monster` object. 10(34 minutes ago, 6 files, 85+ 6-) 13https://github.com/crawl/crawl/commit/25d6f42ac81f 12:53:30 03mgdelmonte02 07https://github.com/crawl/crawl/pull/2676 * 0.30-a0-5-gaafc82b47f: fix 10(15 minutes ago, 1 file, 1+ 2-) 13https://github.com/crawl/crawl/commit/aafc82b47f09 12:53:30 03mgdelmonte02 07https://github.com/crawl/crawl/pull/2676 * 0.30-a0-6-g265b19ff19: fix2 10(4 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/265b19ff198e 12:53:55 <08nicolae> why not just make a custom trove map that's only placed by the trove goblin, that way you can have more control and make it as generic as you like 12:55:37 Doable as well. To get a POC working I was going to start with spawning one that already exists. Trying to read through the code to see how to even do that action of "TG dies, spawn trove portal". I found where to add a section in mon-death.cc to catch TG death but trying to solve the portal gen part. 12:56:33 CAO new install of 0.29 doesn't seem to be letting me save a RC file.� Was told ought to mention it in here.� Thx. 13:02:09 I guess I need something like `place_monster_corpse` in `mon-death.cc` but placing a portal instead of a corpse. Unless `item_def*` can also be a portal entrance but I doubt that 13:02:13 is true 13:04:35 <10PleasingFungus> unclear to me why we’re limiting selection of trove map 13:05:24 only if you do something like a figurine of a ziggurat 13:05:32 Sorry to add confusion. My plan was to start with a POC that generates _a_ trove map just to get off the ground then work on trove specifics after . 13:06:29 seems easier to tweak _what_ trove is generated than the actual trove generation from death which is a brand new thing in crawl. Every other death is an item(s) or corpse. 13:07:28 you may want to look at lua triggers 13:07:49 there's at least one wizlab which triggers things based on death of a monster iirc 13:08:05 geekosaur suppose killing the TG could drop a special zig item that spawns the trove if spawning a portal on death is too cumbersome. I will look into that! Forgot all about wizlab that does that. 13:14:27 New branch created: pull/2677 (1 commit) 13https://github.com/crawl/crawl/pull/2677 13:14:28 03mgdelmonte02 07https://github.com/crawl/crawl/pull/2677 * 0.30-a0-4-gb2babf6489: Add `crawl.stack` function to return the current Lua stack. 10(8 minutes ago, 1 file, 26+ 0-) 13https://github.com/crawl/crawl/commit/b2babf648941 13:45:27 <09gammafunk> OldBeardedGaming: Should be fixed now, thanks! 13:45:54 <09gammafunk> @advil looks like the rcfiles dir for 0.29 didn't get made, so I made it manually with the right ownership/perms relative to the existing ones 13:46:10 <09gammafunk> not quite sure why the dgl scripts didn't make those 13:46:22 <09gammafunk> or that, I should say 13:52:57 Excellent, tested and RC file saves now, thanks! 13:59:20 <09gammafunk> @ZureaL seems that cbro is down? not sure if you're working on it or not 14:19:35 <12ZureaL> Thanks @gammafunk looks like I’ve got some major hardware failure, looking into what I can do to resolve. 14:52:04 can I get a sanity check. This _seems_ like how you're supposed to setup death triggers. The new lua code is in uniques.des but eventually this would get broken out into a specific treasureGoblin.des file� https://gist.github.com/Static-Flow/2ea4e6a3a7ba78bd24d4e9bb8d6816c7 14:52:59 But if I'm understanding it right, this adds a death trigger to the TG mob that calls the treasure_goblin_died func which should place a trove entrance at the x/y cord of the dead TG 15:24:10 Unstable branch on underhound.eu updated to: 0.30-a0-3-gc9032a9cac (34) 15:37:43 <12ZureaL> Catastrophic multiple drive failures. Trying to rebuild now, but I may have to completely rebuild. This is likely going to take a while and will be painful 15:51:15 <09gammafunk> dang, sorry to hear that. Let us know if you need any help from our end 15:52:00 Gonff (L3 TeAE) Crash caused by signal #6: Aborted (D:1) 15:52:01 Nybble (L4 DgCj) Crash caused by signal #6: Aborted (D:4) 16:21:56 <12ZureaL> I've got the array back online (minus one drive) and will be restoring from last nights backup, but I lost my entire primary storage array which means I lost ALL my VM's on the host, so I'm restoring critical services first and then will restore CBR2, so it will be a few hours. I'm going to look towards buying a new server or two and move most of my hosts to some updated hardware, but that's a bit of a longer-term approach. I will 16:21:56 hopefully have CBR2 up by mid-day tomorrow at worst, still hoping for this evening. 16:22:50 <12ZureaL> what that means is, all games from today on CBR2 will be lost. Sorry, I wish I could fix that 17:03:57 <09gammafunk> thanks for the update. I suppose one thing we might need to do is restore the logfile and milestone file from sequell's most recent backup from when before cbro went down, which would have been minutes before any outage 17:04:06 <09gammafunk> rather than servicing the copy you have from your backup 17:04:42 <09gammafunk> @ZureaL those files can be found here: https://loom.shalott.org/server-xlogs/cbr2/meta/ 17:04:56 <12ZureaL> well, I still have to completely restore from backup, but if there are game saves/states that can be copied over after I get it back up...I suppose I can entertain that 17:05:14 <12ZureaL> https://cdn.discordapp.com/attachments/747522859361894521/1010338685348220988/unknown.png 17:05:27 <09gammafunk> the only thing we'd really have is the milestone/logfile files for the installed cbr2 versions, yeah, nothing like save files, morgues, or ttyrecs 17:06:00 <09gammafunk> I'm not sure how CAO scoring and seqeull will react to seeing milestones/logfile entries from games that get resumed from backup 17:06:12 <09gammafunk> but I suppose there's nothing we can do to prevent this 17:09:01 <12ZureaL> I'm certainly open to suggestions 17:09:27 <09gammafunk> if advil is around to see this, he might have a suggestion, I think the only big concern is any breakage for CAO scoring and possibly sequell in terms of restored games generating logfile and milestone entries when both CAO scoring and sequell think they're completed 17:10:00 <12ZureaL> help me advil-one-kenobi, you're my only hope? 🙂 17:10:13 <09gammafunk> One thing I can't see any reason not to do is to restore those logfile and milestone files I linked from Sequell's copy to use over your own backups 17:10:43 <09gammafunk> those are just sequell's fetch of those files and they would be the most recent data available from e.g. minutes before the outage began 17:11:35 <09gammafunk> cursory glance at the date suggests that the files for all versions from 0.24 through trunk would need restore as they have change timestamps from today 17:11:46 <09gammafunk> at least some of the files do, even e.g. 0.24 17:12:13 <12ZureaL> yuck 17:12:57 <12ZureaL> I'm doing a full system restoration, my entire RAID got trashed so it's not like I can do a selective restore 17:12:59 <09gammafunk> looks like you have 4 files per versions (logfile, logfile-sprint, milestones, milestones-sprint) over 6 directories 17:13:25 <09gammafunk> no, you don't need to selectively restore anything, I'm saying you just need the files I linked above to replace your backups 17:13:37 <09gammafunk> of those corresponding files 17:13:54 <09gammafunk> these live in your crawl-master folder in respective version directories 17:14:50 <12ZureaL> once the server is restored, crawl starts automatically on boot, so I'll be serving the "wrong" files right out of the gate. I can shut down ngnix I suppose so nothing can get to it until I do that restore 17:15:14 <09gammafunk> yep 17:19:52 <09gammafunk> these files go in the respective directories (absolute paths): /home/crawl/DGL/crawl-master/crawl-0.24/saves/ /home/crawl/DGL/crawl-master/crawl-0.25/saves/ /home/crawl/DGL/crawl-master/crawl-0.26/saves/ /home/crawl/DGL/crawl-master/crawl-0.27/saves/ /home/crawl/DGL/crawl-master/crawl-0.28/saves/ /home/crawl/DGL/crawl-master/crawl-latest/saves/ assuming you have a standard DGL setup 17:20:24 <12ZureaL> cool, thanks! 17:42:11 Is there a way to _force_ a new mob to spawn in your game to test it out? 17:44:26 <09gammafunk> yes, &m 17:44:43 <09gammafunk> if you're not familiar with all the wizard mode commands, hit &? 17:44:49 thanks! 17:44:54 <09gammafunk> there are dozens that you'll want to use (with varying frequency) as you develop 17:45:25 <09gammafunk> there are also important wizard mode commands under X and x, hit X? and x? in wizard mode to see those 18:06:25 Crawl's wizard mode is amazingly good 18:26:09 <12ZureaL> @gammafunk okay, I've copied all the latest files from the links you provided and make sure all owned by crawl.crawl. I'm about to bring CBR2 back online 18:28:58 <09gammafunk> great, sounds good 18:31:57 <12ZureaL> lol, sneaky bastages figured out the ssh path was open, even though ngnix was down...oops 18:33:40 <12ZureaL> okay, we appear to be back up and running...may be some oddities about the 18 hour gap, but hopefully it doesn't screw too much up. Again, my apologies, multiple drive failures at the same thing wasn't something I could've planned for 18:40:14 <10PleasingFungus> no need to apologize. thanks for working so hard to keep everything running! 18:44:12 I'm surprised this doesn't run on AWS or something. Does DCSS run on like personal server racks? 18:45:49 <09gammafunk> dcss doesn't care about the underlying hardware, if that's what you're asking 18:46:00 <09gammafunk> and a few dcss servers do run on aws 18:46:30 <10PleasingFungus> you could have this exact failure in aws with local storage 18:46:55 I was thinking more from an uptime/reliability standpoint. I don't image DCSS needs much compute power to run. I thought AWS storage had like a 99% uptime SLA 18:50:20 <09gammafunk> dcss is fairly compute heavy in that it doesn't take all that many processes to incur significant cpu load 18:50:50 <09gammafunk> but more to the point, the servers are all owned and operated by each individual server admin, and they can run it on whatever hardware they cose 18:51:06 <09gammafunk> *choose 19:01:09 <10PleasingFungus> doesn’t 99% uptime mean being down for over 3 days/year..? 19:04:56 Their SLA for EC2 says less than 99.99 but >= 99.0 so somewhere between 3 days and 52 minutes. 19:19:50 StaticFlow: If you want to pony up for AWS I daresay individual server runners will be pleased to hear. 19:20:17 haha I've def been pwned by a AWS server bill before! 19:20:40 Also I just spawned the first version of Treasure Goblin and it is OP lol 19:21:34 though I haven't tried it with a bunch of high level gear and leveled up yet. But just off the bat is spawns greater demons and mana vipers way quicker than I was expecting 19:51:25 this is almost impossible to do because there are infinite builds but it'd be cool in Wiz mode to spawn with like an "Avg" kit for a specific spot in the game. I'm trying to create a build to test fighting this thing but just spawning all the unrands and trying to build something not super OP but still kitted 20:00:04 Wizard mode can import morgues, which is imperfect but can help. 20:00:43 Oh that's a good idea! I'll just pick some mid level 20's chars I have 20:27:18 <09gammafunk> StaticFlow: You can look at sequell data to find good stats (at least as a starting point) for various characters 20:27:33 <09gammafunk> !avggamestats * recent hufi xl=20 20:27:35 <04Cerebot> Average stats for * (recent hufi xl=20) over 14 games: XL: 20, MHP: 159.86, MMP: 25.14, AC: 27.29, EV: 14.64, SH: 14.64, Str: 24.79, Int: 12, Dex: 17.14 20:27:45 <09gammafunk> helpful command in that regard 20:27:52 oh nice, super useful! 20:27:58 <09gammafunk> and if you want to know about something more tied to a specific milestone 20:28:13 <09gammafunk> !avgmilestonestats * recent hufi br.end=vaults 20:28:16 <04Cerebot> Average stats for * (recent hufi br.end=v...) over 70 milestones: XL: 24.81, MHP: 210.7, MMP: 33.71, AC: 36.96, EV: 14.69, SH: 20.73, Str: 28.56, Int: 13.5, Dex: 18.16 20:28:29 <09gammafunk> like arriving on V:5 20:28:50 <09gammafunk> as opposed to only looking at deaths, which obviously can be a worse set of stats to sample from 20:29:24 <09gammafunk> and you can add won to that second command to only look at games that went on to win (hence are more likely to be reasonably built chars) 20:30:13 nice! Just when I think I've seen all the stat commands lol 20:30:33 <09gammafunk> I actually made those commands specifically to help with playtesting 20:30:42 makes sense! 21:35:53 03PleasingFungus02 07* 0.30-a0-4-ga7cffc901d: Fix: give octopodes storm tentacles (dilly) 10(3 minutes ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/a7cffc901dbf 21:35:53 03PleasingFungus02 07[stone_soup-0.29] * 0.29-b1-3-g171032cc4a: Fix: give octopodes storm tentacles (dilly) 10(3 minutes ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/171032cc4a7c 21:51:22 Unstable branch on crawl.kelbi.org updated to: 0.30-a0-4-ga7cffc901d (34) 22:34:43 Unstable branch on crawl.develz.org updated to: 0.30-a0-4-ga7cffc901d (34) 22:55:46 Windows builds of master branch on crawl.develz.org updated to: 0.30-a0-4-ga7cffc901d 23:15:05 Unstable branch on cbro.berotato.org updated to: 0.30-a0-4-ga7cffc901d (34) 23:54:02 Monster database of master branch on crawl.develz.org updated to: 0.30-a0-4-ga7cffc901d