00:17:15 Liquefaction fails when cast on stairs (https://crawl.develz.org/mantis/view.php?id=3086) by MarvintheParanoidAndroid 00:21:15 -!- psyshvl has quit [Remote host closed the connection] 00:22:32 hi! 00:23:00 Keskitalo: did you look at the map I emailed? 00:27:19 Aborted spell castings can still wake up monsters (https://crawl.develz.org/mantis/view.php?id=3087) by MarvintheParanoidAndroid 00:32:21 Huh, people have made more suggestions for holy monsters. 02:27:45 moin 02:32:19 whyyyy do functions exist which take a reference and set it and then return a boolean to say if it set that variable or not? 02:33:46 monster teleport moves a monster out of the grid 02:33:47 due: http://en.wikipedia.org/wiki/Semipredicate_problem 02:34:31 due: see section "Multivalued return", second bullet 02:34:36 sorear: okay; more simplistically, why are we using these functions when there doesn't apear to be any need? 02:34:56 it takes a coord_def and returns a boolean if it set the coord_def. 02:35:41 It would make more sense to return a coord_def of (-1, -1) or (0, 0) rather than doing complicated references. 02:36:10 coord_def blah = function(); if (blah.origin()) // do something else 02:36:27 rather than coord_def blah; if (!function(blah)) //do something. 02:36:44 anyway 02:37:04 monster teleport takes a monster out of the grid, finds it a new random location and if found, moves it there... 02:37:05 -!- Textmode has quit [Ping timeout: 240 seconds] 02:37:17 then it puts the monster into mgrd(mons->pos()) and continues as normal. 02:37:28 *it does nothing with a false value returned from the random placement function*. 02:37:37 is this a bug? intentional? 02:38:09 If the random placement function ever couldn't find a valid spot for the monster, it would presumably just put it back into its old position. 02:38:16 Is this intentional behaviour or a mistake? etc. 02:40:04 I am going to rewrite it to just silently fail teleportation if it can't find a suitable location. 02:41:45 putting it back into the original position seems like silently failing the teleportation 02:41:58 exactly 02:42:07 there's no comments to say that this was intentional behaviour 03:02:43 -!- Ashenzari has joined ##crawl-dev 03:02:43 -!- The topic of ##crawl-dev is: Home of the Vi-Kings! | Logs: http://tozt.net/crawl | People with +v have commit access. | Please keep general crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Ashenzari source: http://github.com/doy/crawlbot | Long stuff to pastebin.ca please. 03:03:25 would it be possible to balance weapon pluses working like slaying? 03:03:39 or extra base damage even 03:04:09 because this is why nobody knows which weapon to use 03:07:35 slaying is damn powerful 03:08:13 this would greatly shift the power from randarts to regular weapons -- even of weak base types 03:08:25 or, if we went that way, nerf slaying 03:08:41 I'd try that new weapon wield screen idea first 03:09:09 obviously you'd reduce generation odds 03:09:17 of ew scrolls 03:09:52 and curtail generation of stupid +11 clubs on jessica 03:10:44 Trog has mad +dam bias. 03:10:52 the only way to get +15 gear 03:11:31 seriously? nice 03:11:33 heck, I've got +9,+15 giant spiked clubs from him myself 03:11:39 giant spiked knife 03:11:40 the absolute max 03:11:45 ??penguin 03:11:46 b0rsuk[6/7]: y - the -3,+15 bardiche of the Penguin (weapon) {god gift, speed, +Blink rElec AC+6} 03:12:01 right 03:12:05 ah, penguin had only -3 acc... but still 03:12:50 Oka has a -dam malus, so his randarts suck. 03:16:52 you could alternatively make slaying work exactly like +dam works now, and increase the magnitude of slaying on items by 1.5 or some such to compensate 03:17:55 going the other way is better imo 03:18:46 a bonus to base damage is more intuitive than a flat increase 03:19:16 well, flat increase is used by lots of other stuff currently also 03:19:21 might, at least 03:19:28 I guess pain and elec and such are a bit different 03:19:47 and most new players seem to find the flat 1dX increase more natural, afaict 03:20:34 I'm not sure what I think about the idea, but the way weapon plusses works is definitely conceptually simpler 03:22:43 is it actually +1dx or is it random2(x) 03:22:53 or does it just add the weapon enchantment onto the max damage and roll that 03:23:07 as one roll 03:23:35 ??weapon damage 03:23:35 weapon damage[1/3]: Approximately: 1d(base*strength_bonus + slaying)*skill_bonus + 1d(to_dam) + various bonuses. Weapon {brand} is applied after monster AC reduction. 03:23:47 if that is to be trusted, it is actually +1dx 03:24:03 on the other hand, that says "approximately" 03:24:13 I seem to recall it is actually pretty accurate though 03:24:19 I left out all the random +-1 factors 03:24:34 henzell entries are only useful if they are short 03:26:44 why does "s" not work for splitting hunks using add -p? 03:26:58 it looks like it is actually random2(x+1) 03:27:04 which is more sensible anyway 03:27:28 right, that 03:27:54 tgwiglaf: if slaying directly affected base damage, I would be more inclined to agree with you about it being more intuitive, but it isn't 03:28:07 it should 03:29:09 phrase it this way: a bonus to base damage is more intuitive than a flat roll tacked on 03:29:54 do you also find might nonintuitive? 03:30:01 also, doesn't it being random2(x+1) mean that higher enchanted weapons have a different distribution? 03:30:10 (which is weird) 03:30:28 might is unintuitive in that it has no status light 03:30:48 in this context, though, no, since might is an enchantment and not a property of the weapon 03:30:49 you mean it being a flat increase rather than affecting the max damage? 03:30:58 yeah 03:31:15 that means that higher enchanted weapons have less damage variance, yes 03:31:49 seems like it should be like 1d((base+slaying)*str*skill + enchantment) 03:31:51 I'll note that slaying is also not a property of the weapon 03:32:13 it's a property of the captain's cutlawss 03:32:17 or whatever 03:32:21 (cutlass) 03:32:33 okay, but that's a special case :) 03:32:39 :P 03:33:02 I was referring to the +dam 03:33:12 the enchantment I mean 03:33:17 not slaying bonuses 03:34:49 yes, but you want to make weapon plusses and slaying be equivalent and different from might 03:35:11 and I'm saying that "might is not a property of the weapon" isn't much of an argument when slaying isn't either :P 03:35:20 hrmm 03:35:49 I want to make the damage enchantment add to base damage 03:36:04 if you'd rather, slaying could work like the enchantment does now :P 03:36:12 just so everyone is confused 03:37:05 I wonder, if shop generation is converted to be done using vaults, will some odd behaviour that's worth preserving be gone? 03:37:51 what behavior? 03:38:05 I don't know, hence I ask 03:38:17 shops can be placed within vaults, as far as I know. Is it okay if they no longer can be? 03:38:39 that's probably a good thing I think 03:38:44 there are several vaults with fixed shops 03:38:57 sorear, I'm talking about random shop generation 03:39:01 Zaba: bazaars 03:39:08 randoml y putting things into vaults has the potential to mess with them 03:39:13 Zaba: random shop generation has a highly clusterish distribution 03:39:13 due, what about them? 03:39:17 better if it's not done often imo 03:39:33 Zaba: 5 is much more likely than (chance of 1)^5 03:40:41 do serial vaults still exist? 03:40:54 my current reimplementation uses serial vaults to place shops 03:40:58 right 03:41:08 you could preserve the distribution pretty well 03:41:09 sorear, are you referring to _num_shops? 03:41:26 What's wrong with current shop generation? 03:41:47 due, if it can be adequately reimplemented using a more general-purpose mechanism, why shouldn't it be? 03:42:34 if the behaviour is identical, sure. 03:42:49 apart from not placing shops in vaults, it should be the same 03:43:03 Well, why wouldn't you be able to place shops in vaults? 03:43:16 shops in vaults have the potential to be quite amusing 03:43:24 because the vault placement mechanism won't overwrite existing vaults 03:43:29 I've had a shop in the miasma cloud on the undead swamp:5 vault 03:43:49 elliptic, I don't think you can get shops in swamp these days. 03:44:01 oh, true, but the idea still holds :) 03:44:16 Zaba: okay, I say again. Why wouldn't you be able to place shops in vaults? 03:44:46 Your serial vaults are going to be 1-square definitions with a random shop type? 03:44:47 due, shops don't get _randomly_ generated in vaults 03:44:59 due, with 'any shop', yes 03:45:30 plus a serial vault that places either num_shops() of those, or a bzr_entry and num_shops() - 1 of those 03:45:33 Builder shouldn't place shops in vaults already. 03:45:38 due, why not? 03:45:55 it only checks MMT_NO_SHOP, and there are only a few vaults that are appropriately tagged to have that mask 03:46:08 Eh, it's just odd. 03:46:25 Probably because shop generation is so rare that people don't tag it like that. 03:46:40 But anyway. Are you saying that you wouldn't be able to do KFEAT: / = any shop and then have a / symbol in a vault? 03:46:46 ...no 03:46:50 Or are you saying that we would no longer have shops being randomly generated in vaults? 03:46:57 yes! 03:47:08 ... 03:47:13 Oh, you said "within" :/ 03:47:16 which is what I have said with _emphasis_ not all that long ago 03:47:16 I read it as "in". 03:47:24 he's saying that the mechanism to place shops would no longer overwrite vaults 03:47:31 Okay I'm cool with that. 03:47:33 But it was a bit unclear. 03:48:05 It was something like "Shops can be placed within vaults currently. Is it okay if this is no longer possible?" 03:48:09 I read the "within" as "in". 03:49:09 the entry with the boring beetle should have the boring beetle removed 03:49:27 it's confusing for no reason 03:50:11 or, boring beetles could just stop giving messages; there's no real need (unlike with slurps) 03:50:19 -!- TGWiglaf has quit [Quit: Page closed] 03:51:49 so.. is it okay if we no longer get random shops generated in vaults that are already present on the level? 03:52:00 or would removing such an occasional gem be too game-breaking to be even considered 03:52:20 "occasional" about sums it up. 03:52:32 I don't see how removing it could be problematic; at least for vault builders, it's not expected behaviour. 03:56:32 hmm.. 03:59:42 Hm, agrid isn't being initliased on the start of ggames AGAIN. 03:59:51 -!- syllogism has joined ##crawl-dev 04:08:39 03zaba * ra8579851cca3 10/crawl-ref/source/ (dat/des/builder/shops.des dungeon.cc): Convert random shop generation to serial minivaults. 04:08:40 03zaba * r163785cb0e9d 10/crawl-ref/source/ (branch-data.h branch.h l_dgnlvl.cc): Remove shop_chance from branch-data.h. 04:12:19 -!- Zaba has quit [Ping timeout: 276 seconds] 04:15:03 -!- Zaba has joined ##crawl-dev 04:18:34 kilobyte: I'd make lamassu but I'm too lazy. 04:18:44 03due * rec93436a95f2 10/crawl-ref/source/ (12 files in 2 dirs): Finally complete shedu! 04:22:43 Clamping effect might be a bit odd, butt meh. 04:39:44 Do we have an ophan tile? 04:41:08 -!- Mu_ has joined ##crawl-dev 04:51:57 SPELL_RESURRECT can go. 05:06:14 03due * r400956faf1e3 10/crawl-ref/source/dat/database/monspell.txt: Add blessed toe casting message for Heal Other. 05:06:16 03due * r492d7ca41217 10/crawl-ref/source/dat/descript/monsters.txt: Tweak Ophan description, fix a typo. 05:06:18 03due * r94874e660d7f 10/crawl-ref/source/dat/descript/monsters.txt: Fix Shedu descriptions. 05:06:21 03due * r3f17aa6699b3 10/crawl-ref/source/mon-data.h: Tweak Ophan colour, too. 05:06:22 03due * rda128bc42133 10/crawl-ref/source/ (dat/database/monspell.txt mon-cast.cc): Cast messages for all holy spells; tweak cast look-up. 05:06:23 03due * r44366bd62b25 10/crawl-ref/source/ (mon-cast.cc mon-spll.h): Some minor holy clean-ups. 05:11:33 * greensnark resurrects due. 05:12:33 03zaba * r0daedcb047f9 10/crawl-ref/source/mon-death.h: Add #ifndef/#define MONDEATH_H to mon-death.h. 05:12:34 03zaba * r77e59290c3cf 10/crawl-ref/source/ (5 files): Correct whitespace errors using checkwhite. 05:14:30 -!- monky has quit [Quit: hello] 05:15:24 <3 greensnark. 05:15:40 What's the point of #ifndef/define MONDEATH_H? 05:16:37 it means its contents are included only once 05:16:49 every other header does it ;P 05:17:36 wouldn't the compiler optimise that out? 05:18:11 they're only declarations, there's nothing to optimize in them 05:18:35 No 05:18:44 Wouldn't the compiler optimise the contents being included multiple times? 05:18:50 but if there was some other header that'd include mon-death.h, you'd get mon-death.h's data twice in a single .cc file after preprocessing 05:20:02 no, the preprocessor doesn't care if you include a file several times, that's why you need to take care of that using such ifndef/defines 05:21:40 hence why C++ sucks 05:24:55 Argh. 05:25:30 due: Hence why it's good to know the phases of translation. 05:25:41 due: C has the same "problem". 05:26:02 C has less features that might make you think of it as of a higher level language than it really is, though 05:26:04 I'm glad I don't work with C++; I'd kill myself. 05:26:08 Note that separation into "headers" and "source files" is just convention. 05:26:18 You can have any code in a header, really, as long as it ends in a newline. 05:26:24 Zao: Yes, I know. 05:26:41 Thus, the compiler cannot "optimize it out", as it cannot deduce if it was intentional or not. 05:26:50 Meh. 05:26:59 See awesome libraries like Boost.Preprocessor, which has header recursion :D 05:27:06 I did some C++ work in the past; it was terrible. 05:27:10 Why does everynoe have a massive hardon for Boost? 05:27:18 And, if such a hardon is obligated, why do we not use Boost? 05:27:31 Hm. Jory's "explode in blood" is maybe a bit... much. 05:27:35 due: Because it contains some libraries that make life much better for most applications. 05:27:48 We used to have snotty flamewars about people not using pimpl and about how multiple return points for functions are evil and really silly conventions about managing COM components 05:27:55 there's a stigma associated with using boost 05:27:56 C++ is painful enough to not take advantage of the brief glimmers of light :D 05:28:10 boost is awesome, but it's a giant dependency 05:28:32 Every time I google an issue there's all these idiots going "just use boost instead". 05:28:40 It really makes C++ much nicer to work with 05:28:44 Like about 200% nicer 05:28:46 Even if you bcp out something simple such as say shared_ptr, you get around three megs of headers. 05:29:09 Mostly because it encourages mutual use of libraries to make life less horribad. 05:29:23 Things that suck: people wasting time to tell other people that they shouldn't be trying to do what they're trying to do, rather than whether oor not it is possible. 05:29:45 "Can I do this?" "You should do that instead" "No, but can I do this?" "NO! YOU MUSTN'T EVER EVER DO THAT!" "..." 05:29:45 Yup. 05:29:57 "I really, really want to do this. Is it possible?" 05:30:03 "Meh, yeah." 05:30:20 #python and most # channels seem to have this issue; so do #ubuntu and a few other places. 05:30:50 they do the right thing by encouraging best practices 05:31:30 No, they piss people off by being obtuse. 05:32:18 "can I shoot myself in the foot?" "yeah, sure!" 05:32:20 Encouraging the right practices is "Yes, you can do that. This is how. But you should do this instead, because of why." 05:32:21 that's how you want it to be 05:32:22 due, anyway, if you want to do X, and you think Y is a way to do X, you should ask "how do I do X?" rather than "can I do Y?", no matter how sure you are in that Y is the right way. 05:36:20 I can't even think of an example but it got into a huge wrapping versus subclassing argument when I simply asked what the best method for subclassing something was. 05:36:21 Erm, s/of an example/the example.. 05:36:21 There were two people who were all "omg NEVER SUBCLASS" and some other people who were "subclass if necessary". The first set told me to rewrite everything; the second told me how to do what I wanted to do. 05:36:35 Hence why I much prefer google. 05:36:49 Yay less sanctimonious arseholes! 05:40:13 03due * rabec47e9177a 10/crawl-ref/source/ (9 files in 4 dirs): New unique: Jory, in the Tower of Silence; new monster enums. 05:42:15 -!- elliptic has quit [Ping timeout: 240 seconds] 05:42:43 03due * re5c4acac7e9c 10/crawl-ref/source/dat/database/quotes.txt: Fix raven quote (Sunfall). 05:43:11 03due * r87dfcf8a558d 10/crawl-ref/source/mon-stuff.cc: Properly make bog mummies curse equipment. 05:44:07 is "jory" a reference to something? Juri or yuri would feel better I think 05:44:11 at least as an american 05:44:28 Juri looks odd and yuri makes me think of bad fan fiction. 05:44:29 OG17: "yuri" has erotic connotations. 05:44:30 So Jory it is. 05:45:24 yuri's a pretty widespread name, strange to rule it out because of what it means in a single language 05:45:28 why jory anyway, though? 05:46:38 Because it's the Cornish spelling of Jure and I like Cornish names. 05:47:09 jure? 05:47:14 Yes. 05:47:21 Jure Grando. 05:47:25 Look it up on Wikipedia. 05:47:40 just did 05:47:52 grando is a pretty great name though 05:49:26 Sounded too Marxian, I thought. 05:52:07 Rats! Laboratory rat tiles (https://crawl.develz.org/mantis/view.php?id=3088) by coolio 05:56:49 due, were you planning on making leda slow monster movespeed instead of total speed? Going to be weird if a monster casts it and slows itself and other monsters more than the player 05:57:00 a difference is weird anyway 05:58:27 hmm. 06:02:48 -!- galehar has joined ##crawl-dev 06:04:20 !tell jpeg I will detach the command tab. But I'm doing it in a generic, flexible and optionable way, so it might take a while. 06:04:20 galehar: OK, I'll let jpeg know. 06:21:49 -!- blackpenguin has quit [Ping timeout: 260 seconds] 06:50:05 hrm. 06:50:19 -!- coolio_mc has joined ##crawl-dev 06:53:35 at the moment, _place_altars has a (c / 5^(n - 1))% chance to place the n-th altar (c% for 1st, (c / 5)% for 2nd, etc.). 06:54:09 c depends on the branch (20 in Orc, 10 in snake/forest/dwarf, 8 in D/elf, 5 in lair/slime/vault/crypt, 1 in zot) 06:55:01 converting what _place_altar does to a vault is pretty trivial, but preserving the slim chance (4% in orc, 2% in snake/forest/drawf, 1% in D/elf/lair/slime/vault/crypt) of placing a second altar on a single level is not 06:56:00 * Zaba ponders 06:57:31 I _don't_ want to look like somebody who sacrifices existing features for code quality, but, objectively, does such a slim chance of a second altar really matter? 06:58:17 Zaba: As a gut feeling, unpredictability is good. 06:58:36 -!- MarvinPA has quit [Quit: Page closed] 06:58:55 Zao, what is unpredictable here? 07:03:17 Zaba: seems to me the not only is the change small, but it's not even wrong. I don't see how having rarely a second altar on the same level can be a bad thing. 07:03:28 s/the/that/ 07:03:57 I'm not saying it's a bad thing 07:04:06 and if it increase the code quality, then +1 07:05:28 if _place_altar is (naively) converted to a minivault, said placing of a second altar will be gone 07:13:04 hrm.. 07:23:24 -!- Zaba_ has joined ##crawl-dev 07:26:24 -!- Zaba has quit [Ping timeout: 255 seconds] 07:52:54 -!- galehar has quit [Quit: disconnect] 08:02:32 -!- Zaba_ is now known as Zaba 08:39:14 -!- TGWiglaf has joined ##crawl-dev 09:12:25 -!- OG17 has quit [] 09:13:20 -!- OG17 has joined ##crawl-dev 09:14:24 -!- OG17 has quit [Read error: Connection reset by peer] 09:19:54 -!- OG17 has joined ##crawl-dev 09:25:30 -!- TGWiglaf has quit [Quit: Page closed] 09:28:02 -!- TGWiglaf has joined ##crawl-dev 09:40:11 -!- Textmode has joined ##crawl-dev 09:49:05 -!- TGWiglaf has quit [Quit: Page closed] 09:53:24 -!- TGWiglaf has joined ##crawl-dev 10:02:36 -!- blackpenguin has joined ##crawl-dev 10:11:45 -!- TGWiglaf has quit [Quit: Page closed] 10:42:31 -!- psyshvl has joined ##crawl-dev 10:47:55 03j-p-e-g * r68fee44eb422 10/crawl-ref/source/ (72 files in 4 dirs): Add hellbinder sigil tiles for the demon wizlab. 10:49:49 03zaba * rd95ef9a50856 10/crawl-ref/source/dungeon.cc: Remove the unused 'vault_only' argument from _place_specific_stair, and reorder some functions. 10:49:51 03zaba * ra51ef52322e5 10/crawl-ref/source/ (dat/des/variable/altar.des dungeon.cc): Convert _place_altar(s) into a minivault. 10:49:52 03zaba * rcdf65f0f765c 10/crawl-ref/source/ (branch-data.h branch.h): Remove altar_chance from branch data, as it is no longer used. 10:59:38 -!- Cryp71c has joined ##crawl-dev 11:00:28 -!- Cryp71c has quit [Client Quit] 11:00:56 -!- Cryp71c has joined ##crawl-dev 11:14:53 -!- enne_ has quit [Quit: *poof*] 11:15:08 -!- enne has joined ##crawl-dev 11:23:05 -!- psyshvl has quit [Quit: stuff] 11:46:23 -!- monky has joined ##crawl-dev 11:53:20 03j-p-e-g * r0647c905dcef 10/crawl-ref/source/rltiles/dc-floor.txt: In dc-floor.txt, set %sdir dc-dngn/floor and change paths accordingly. 11:53:25 03j-p-e-g * ref0560ae8fe1 10/crawl-ref/source/rltiles/dc-wall.txt: And the same for walls. 12:12:07 -!- elliptic has joined ##crawl-dev 12:26:14 -!- psyshvl has joined ##crawl-dev 13:18:58 03j-p-e-g * r0ec58c456f82 10/crawl-ref/source/ (tag-version.h tags.cc): Reinitialise dungeon tiles if TILE_WALL_MAX changed between versions. 13:19:39 We srsly need to just save tile names into files to avoid that sort of workaround. :( 13:19:55 Though, <3 jpeg for doing something about it. 13:27:22 -!- psyshvl has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.6/20100625223402]] 13:28:10 enne: especially when a tiles save gets loaded in console 13:28:58 There are some separate issues there. We almost need to save some of the tile info (like special tiles from vaults, etc...) even when playing the console version. 13:29:09 ...so that going back and forth isn't such a lossy process. 13:29:16 (Although, who are these players going back and forth?) 13:30:36 this is not a big issue now, but it will be if nettiles ever get off the ground 13:57:28 Finnish theme channel is showing The seven samurais, directed by Akira Kurosawa in its authentic form 13:57:31 ... without subtitles 13:57:36 -!- dpeg has joined ##crawl-dev 13:57:42 !seen Zaba 13:57:42 I reckon that's very... useful for 99% of the populace 13:57:42 I last saw Zaba at Wed Dec 29 13:26:24 2010 UTC (6h 31m 18s ago) quitting with message Ping timeout: 255 seconds. 13:58:34 felirx: are they doing that intentionally? I'd call that just a mistake everyone can make. 13:58:53 probably a malfunction of some sort 13:59:42 and I guess they're not as well-staffed as major channels 14:00:24 I think they speak in arhaic samurai Japanese though 14:00:28 can't understand much :( 14:00:56 but oh well, I have no TV in places where I lived for 14 years and counting, so I'm not exactly knowledgeable about TV channels 14:01:35 watched precisely one movie the whole year, too 14:01:51 I haven't properly watched tv in like 5 years 14:01:57 only when I'm at my dad's like now 14:07:41 -!- jpeg has joined ##crawl-dev 14:09:57 new Antaeus tile (https://crawl.develz.org/mantis/view.php?id=3089) by co 14:35:17 kilobyte: I am not sure what to do with the level_type mails. Did the replies give us a clue? 14:40:07 dpeg: not really, Zaba noticed that this change wouldn't remove all special cases, and it is obviously true 14:40:17 it still would remove quite a number of them 14:41:56 is the plan to get rid of level_types entirely? 14:41:56 jpeg: You have 1 message. Use !messages to read it. 14:42:00 !messages 14:42:01 (1/1) galehar said (8h 37m 41s ago): I will detach the command tab. But I'm doing it in a generic, flexible and optionable way, so it might take a while. 14:43:02 !tell galehar No sweat! Detaching isn't urgent, and we definitely need it configurable. Thanks! :) 14:43:02 jpeg: OK, I'll let galehar know. 14:44:12 jpeg: my idea, at least. But it's not level_type which I care about, it's mostly about allowing more levels for types other than "main dungeon" 14:45:03 sorry, I still don't understand 14:45:08 it doesn't matter if we fix all comparisons and keep it multidimensional, or if we flatten it to just branch/depth 14:45:52 ah, you mean levels as in multi-level portal vaults 14:46:03 for some reason I thought you were talking about subtypes 14:46:04 the problem is: you can have either LEVEL_DUNGEON:branch:depth, or just LEVEL_PANDEMONIUM 14:46:15 yeah 14:46:29 why is that bad? 14:46:53 I don't care if it becomes BRANCH_PANDEMONIUM:3 or LEVEL_PANDEMONIUM:some_branch:3 14:47:35 you want to allow going back and forth in Pan or Ziggurats? 14:48:14 not necessarily... but allowing using DEPTH: and such in Pan and Ziggurats would make things easier 14:48:35 Zigs have depth hacked in Lua, poorly 14:48:51 it always irked me that you can't access the descriptions for non-Dungeon levels from ?/, but that's probably a minor point 14:49:54 jpeg: but it also follows from lack of consistency 14:50:03 kilobyte: should discuss on IRC when Zaba is aroun 14:50:04 d 14:50:25 yeah, I don't fully understand what he means 15:11:59 -!- psyshvl has joined ##crawl-dev 15:25:02 hey, co is not just drawing tiles, he also makes great maps 15:38:43 -!- Cryp71c has quit [Quit: Leaving] 15:57:56 damn it I can never figure out cross-compiling for windows :( 15:59:48 -!- galehar has joined ##crawl-dev 16:00:12 evening 16:00:12 galehar: You have 1 message. Use !messages to read it. 16:00:17 !messages 16:00:17 (1/1) jpeg said (1h 17m 15s ago): No sweat! Detaching isn't urgent, and we definitely need it configurable. Thanks! :) 16:00:28 galehar: hi! 16:00:41 hi! 16:00:46 hi guys! 16:09:47 Zet: do you mean compiling from Linux for Windows? 16:10:00 yes 16:10:37 you need to have mingw32 installed 16:11:38 on Debian, it comes in two versions: gcc-mingw32 and mingw32 16:12:13 both work, you probably don't care which one you want 16:12:36 I'm sure Red Hat and Gentoo have "mingw32" somewhere in the name 16:13:50 then, you compile with: make debug CROSSHOST=i586-mingw32msvc 16:14:07 (or WIZARD=y instead of debug, or with TILES=y as well, etc) 16:18:18 dpeg: I did it :D Only think I would like to keep is "easy_quit_item_prompts". Sorry for the delay. 16:19:13 kilobyte: since when does the debian version of mingw32 produce usable binaries? Is the problem gone we once had? 16:20:07 Zet: I had install MinGW from here: http://www.nongnu.org/mingw-cross-env/ 16:20:12 quite easy though 16:20:45 Guardian Serpent Corpse can be zombified... (https://crawl.develz.org/mantis/view.php?id=3090) by Twilight 16:21:14 Napkin: hmm, I kind of forgot what the problem was... but compiles from my box work correctly 16:21:17 gcc-mingw32 16:21:18 dpeg: *Only _thing_ I would like to keep is "easy_quit_item_prompts". I will never understand why people would like it to behave like "more" instead of "less" 16:22:17 lemme re-check both console and tiles (will take ~5 mins) 16:22:19 the binaries build with debian's mingw on cdo (lenny) had problems on quitting a game.. it crashed before the inventory could show 16:22:40 yes, now I recall that... 16:23:03 something like that.. I remember installing the 64bit version of mingw from sarge.. then 32bit from lenny.. and and and 16:23:31 compilation always worked - but somehow the resulting binary always had some problem 16:23:48 uhm... sarge and mingw64? I am kind of sure it was packaged only very recently. 16:24:16 puuh, I don't remember the details anymore 16:24:32 crash on quit? Is this related to #3060? 16:25:02 i'm just happy we got it working with the stuff from http://www.nongnu.org/mingw-cross-env/ 16:25:29 no, galehar 16:25:38 Napkin: thanks 16:25:48 the crashs we had were not debuggable and it was happening in the Console version too 16:25:59 thanks for waiting, dpeg :) 16:26:04 Napkin: you set easy_quit_item_prompts to off? 16:26:14 yes, always 16:26:26 interesting 16:26:42 space to see second page.. space to see top again.. i'm paging it 16:27:11 also, using cursor-down to check if i'm at the bottom.. which would leave the inventory without that option to false 16:28:40 ok 16:28:46 also a reason i hate to have to use "more" on AIX.. just scrolling one more line down and pager is closed.. grr.. have to open it again ;) 16:30:24 in Crawl, I've mostly got that problem when browsing my notes 16:30:25 imagine pressing page-down once too much in windows notepad and it closes because of that :-O 16:30:36 hi jpeg 16:30:40 hi! 16:30:58 :) 16:31:35 Napkin: Has your dad seen the detached tabs yet? (provided the resolution matches the requirements, of course) 16:32:08 only in a screenshot galehar posted here which I sent him via email ;) 16:32:08 03j-p-e-g * r16aacef1caa1 10/crawl-ref/source/ (dungeon.cc mapdef.cc tags.cc): Store tile name for vault-defined monsters. 16:32:11 03j-p-e-g * r4af2d1954426 10/crawl-ref/source/mapdef.cc: Also store vault-defined monster tile names in console builds. 16:32:31 i'm keeping him at stable - less updating work for me ;) 16:32:51 time to get 0.8 out the door then :) 16:32:56 yes! :D 16:33:04 i will give him rc of 0.8 though 16:33:14 rc? 16:33:15 he's a pretty thorough beta tester ;) 16:33:20 ah! 16:33:24 the upcoming release candidate 16:33:28 Hi! 16:33:35 Hey Eino :) 16:33:36 Keskitalo! 16:33:42 How's the family? 16:33:50 Good evening/night, Napkin, jpeg 16:33:56 Had any sleep lately? ;) 16:34:01 jpeg: great and awful :) 16:34:04 hehe 16:34:32 the mother is back in the hospital for a serious inflamation, but she's going to be fine 16:34:56 Keskitalo: child is at home? 16:35:00 She should be back home on Friday 16:35:03 Hi dpeg! 16:35:09 Keskitalo: congrats, btw 16:35:15 Yeah, she's with me, we're doing great 16:35:25 :) 16:35:26 jpeg: Thank you :) 16:35:53 We have my mother here helping out and also had some friends over yesterday 16:36:07 And I'm getting just enough sleep :) 16:36:10 Hang in there, Keskitalo! :) 16:36:26 sleep - reminds me - off to bed! 16:36:30 g'night o/ 16:36:35 Good night! 16:36:40 Nacht! 16:37:21 I as just saying yesterday when I looked up commit mails, it's awesome how many changes and iproveents there have been in some four days or so 16:37:29 or whatever, a week actually 16:37:34 yeah, time for UP2 :p 16:37:37 I ave no sense of time of course :) 16:37:45 jpeg: Hey, actually! 16:38:06 We will get a team of testers in this spring 16:38:36 on the usability testing course 16:39:48 Has anyone read this? https://crawl.develz.org/wiki/doku.php?id=dcss:usability_project:immortal_warwalrus_results 16:39:59 this was the usability testing course team we had last year * 16:40:15 on top of the project course that we did with felirx and others 16:40:36 sadly, their report was in Finnish and I lacked the stamina to finish the translation to English 16:41:26 I recently had a look through the UP wiki page, and was happy to be able to cross off several issues 16:41:27 but we have another team lined up this spring, and we could test the new tutorial and see if some of the learning difficulties were fixed 16:41:37 jpeg: Did you update the page? Awesome! 16:42:18 about the tutorial, one thing that bothers me is that I feel you don't get enough time to actually try out the new commands 16:42:34 it's pick up, wield, wear, hit, fire, zap, cast, ... 16:42:41 Anyway, I'll mandate that they hand over their report in English - that way it can be almost directly submitted to the wiki and get use 16:43:26 with the immortal warwalrus report, I feel it hit the language barrier and if I recall correctly I and felirx are practically the only ones who really even know about it, and I'm not sure if felirx has the document either 16:43:27 ... and if you want to go back to try again, you have to go through this horribly lengthy introduction about movement 16:43:46 ... with autoexplore being constantly interrupted by the tutorial messages 16:44:01 I feel like I dropped the ball on that one, but in retrospect the report being in Finnish is what the proble is 16:44:26 jpeg: Yes, there's no chance to really reinforce the learning with repetition 16:44:27 don't we have more Finns on the team? feels like there are dozens of you 16:44:47 hehe :) 16:45:30 I'm thinking about modifying the tutorial yet again, separate the main fighting options (same as in hints mode: melee, magic, ranged) 16:45:54 the imortal warwalrus report was interesting in two things; it had testers with no roguelike background, so they had kind of harsher, more modern expectations 16:45:55 ... so the player can try all of them in one session, or do them in several sessions, if they'd prefer that 16:46:03 the other thing was that it replicated a lot of our findings 16:46:13 jpeg: Yes, this is a great idea in my opinion 16:46:22 I was going to suggest this sort of "episodising" 16:46:35 heh, even without a "proper" report, your findings were extremely helpful 16:46:46 oh yes, I remember 16:46:54 huge buzz on our kind of ad-hoc initial findings report 16:47:21 then the final report, which was a huge amount of work, didn't get as much "bang for the buck" so to speak 16:47:24 in fact, two days ago rereading the wiki page prompted me to add the new clickable action/command tab 16:47:33 \o/ 16:47:38 that was an awesome addition 16:47:43 haven 16:47:59 't tried it yet, but.. the game is probably very close to being entirely playable by mouse! 16:48:25 -!- Textmode has quit [Ping timeout: 276 seconds] 16:48:50 the only big menus left are the help screen (!) and making the ability screen clickable too 16:48:54 hi jpeg! 16:49:01 jpeg: Do you think we should test stable release (0.7.1 tutorial) or test a 0.8 build? 16:49:14 due: hi! 16:49:30 Keskitalo: if you do it in March, take 0.8. 16:49:34 Keskitalo: depends on the state of the game when you get around to it 16:49:52 testing the 0.7.1 version is kind of interesting "academically" because you could tell how much impact UP team had in a short time 16:50:00 true 16:50:11 then again 0.8 if it's in a good state, yeah, that would probably benefit the game developent ore 16:50:16 *more 16:50:18 but I want feedback on the tiled tabs! 16:50:24 haha, i make so many typos 16:50:44 elly: violetj: jpeg needs to be flagged as a committer in chanserv now that ey is on IRC 16:50:53 due: Could you please have a look over the hellbinder vault, just to check I didn't break anything? 16:51:00 jpeg: ! Sure! 16:51:05 sorear: can't, someone's hogging my nickname :( 16:51:17 sorear: and I refuse to change 16:51:18 jpeg: I don't know yet, I heard from the course person in November so i dont' know the schedule 16:51:19 sorear: was already done 16:51:37 jpeg: if needed, detached tabs could be backported to 0.7 16:51:48 jpeg: wait, does someone else have your nick registered? 16:51:56 galehar: Not worth it for the sake of testing i say 16:51:58 galehar: too big of a change, I'm afraid 16:52:01 jpeg: if so, join #freenode and politely request that it be dropped, as it's expired 16:52:05 jpeg: then you can register it 16:52:09 -!- pointless_ has joined ##crawl-dev 16:52:11 galehar: we generally try to avoid that 16:52:23 Hi pointless_ !! 16:52:25 elly: really? 16:52:30 jpeg: really really 16:52:31 ok 16:52:34 how do you know if a nick has expired? 16:52:35 16:51:17 -NickServ(NickServ@services.)- Last seen : Jul 23 01:55:17 2010 (22 weeks, 5 days, 20:56:00 ago) 16:52:38 hi 16:52:40 jpeg: they're eligible for drops after 60 days + 1 week of inactivity per year of age 16:53:07 jpeg, dpeg, felirx, galehar, let's have e-mail about the testing thing 16:53:11 of course, jpeg is a very attractive nick :) 16:53:15 Keskitalo: yes 16:53:15 elly: Then I don't think this one is. 16:53:22 due: jpeg certainly is 16:53:22 dpeg: :D 16:53:25 I'll quite possibly just mail on c-r-d, but I expect replies then :) 16:53:27 png is free! 16:53:34 due: 22 weeks is, um, a lot of weeks :) 16:53:35 Keskitalo: yay! 16:53:37 It's been registered 5 years but only inactive 22 weeks? 16:53:47 due: yes - 60 days + 1 week per year of age, so 60 days + 5 weeks 16:53:50 60 days = 8 weeks * 5 = 40 weeks 16:53:54 which is about 13 weeks, which is less than 22 16:54:01 oh 16:54:06 due: 60 days + (1 week per year of age), sorry 16:54:15 right :) 16:54:16 hehe :) 16:54:32 dpeg: what do you think of the implementation of leda instead of the one previous discussed? 16:54:33 due: I haven't tested any of the, but nice work on the new monsters and other stuff too 16:54:38 everyone has been awesomely busy 16:54:46 and lots of patches and graphics and vaults too 16:54:54 Keskitalo: Shedu are done! 16:55:07 yes! it's my favourite holy monster idea 16:55:17 and it got TWO nice tiels from contributors :D 16:55:27 where will the holies actually appear? zig? 16:55:30 Easier optionis to split them into 1 shedu and 1 lamassu. 16:55:37 Then we can use both tiles :D 16:55:44 due: That could be fun. 16:56:07 due: mirror one of them and have them face each other :) 16:56:16 :D 16:56:29 they could be like dowand and duvessa, and providea slightly different threats to different character builds 16:56:41 jpeg: you should go get that nick so I can add you to the access list 16:56:42 or i don't know 16:57:02 feeding and diapers i think, back soon 16:57:06 well, their main threat is melee; you have to either kill them both very quickly 16:57:16 Basically they're difficult to kill andkeep dead :) 16:57:31 elly: how, where? I'm completely clueless about IRC... 16:57:36 jpeg: /join #freenode 16:57:44 and then? 16:57:45 jpeg: then ask politely for the nick 'jpeg' to be dropped so you can use it 17:00:23 due: I think it's okay. The main points are that the spell should have a drawback to the caster and not require targeting. Both are met. 17:03:02 well I'm not liking sprint-4 very much 17:05:05 Blash, I said I wasn't going to do any crawl coding this morning... too many other pet projects to work on! 17:05:47 pointless_: why? 17:06:01 due: only a little Crawl :) 17:06:10 I seem to get one shotted by acid spit a lot 17:06:16 gah 17:06:19 I broke the imeplemtnation already 17:08:19 Was anyone able to look over my implementation of area centres and check it for memory leakage/speed? 17:08:20 pointless_: do you use invisibility? 17:08:49 not particularly, do I have to play En? 17:09:18 pointless_: the ring! 17:10:35 jpeg: 31 tiles? O_o Wow! You are amazing! <3 17:10:40 elly: there, all mine! Mwahahaha! :D 17:10:53 you know I never realized that was a fixed object, and I've been trying to buy ID scrolls for it 17:11:04 due: I had practise with the krakens :) 17:11:12 :) 17:11:36 pointless_: you should have used it a few times :) 17:11:46 jpeg: you're added to the access list 17:12:01 if I want to store vault-defined tile name strings, should I simply use CrawlHashTable? 17:12:21 I should have used it to kill all the oklobs 17:12:25 elly: yay, thanks! <3 17:13:46 pointless_: well, you'll get glow. But without Inv it seems hopeless. 17:13:50 -!- psyshvl has quit [Quit: stuff] 17:13:51 jpeg: no problem :) 17:14:09 elly: I mean the instructions 17:14:10 well you can get by the first few oklobs if they don't spit at you or they miss, but it is very annoying 17:14:28 jpeg: Would it be easier to save the string names once--in a global array of strings (which could be saved and loaded so long as you're using tiles), or even saved in the map cache--and then instead store the index into that array at each location for a vault-overriden tile? 17:14:31 elly: I got my dev plus several times before, hope this time it sticks 17:14:40 jpeg: hope so :P 17:14:42 if it doesn't, poke me 17:14:45 * elly is the channel fairy 17:15:11 due: that's what kilobyte suggested (I think) 17:15:34 jpeg: Hooray, great minds think alike! :) 17:15:56 due: if I use the hellbinder vault as reference, there are a lot of tiles that are used repeatedly 17:15:59 Otherwise storing a string per-tile would seem to be like a lot of bloat and repetition. 17:16:26 due: an unusually large amount of special tiles, too 17:17:58 yeah 17:18:06 However, it's worth it to make vaults look great in tiles! 17:19:06 especially *that* great :P 17:19:30 Oh 17:20:05 dpeg, jpeg, discussed it the other night and everyone seemed to think it was a good idea; I'm replacing grey rats and their bands with porcupines: individual but slightly harder monsters than snigular grey rates (to compensate for grey rats not gneerating in bands). 17:20:15 This will remove grey rats from standard dungeon generation. 17:20:44 due: fine with me, grey rats are a bit bland 17:20:49 However, I'm *not* deleting them; this way they can be still used in vaults (Sewers, etc) and summon small mammals. 17:21:13 The latter was the most important in my mind to keep behaviour the same; I don't want to unbalance it in favour of the player by accident. 17:22:10 due: yeah, sewers, I know 17:22:41 any relation between porcupines and the DS spine mutation? 17:23:45 I'm going to implement this eventually :) 17:23:59 Small chance of getting "poked" whenever you attack them or they attack you. 17:25:34 there is one issue with grey rat ==> porcupine: a bunch of vaults use rats of all colours 17:25:45 due: ah, I just read on 17:25:46 dpeg: Yes, these vaults can still use grey rats. 17:26:04 -!- psyshvl has joined ##crawl-dev 17:26:06 hey, I just realized that a porcupine can be described as a pork (pig) of pin(e)s 17:26:06 The "bore" factor of grey rats is closely limited to how often and how many of them generate in the Lair. 17:26:12 wouldn't it be better to tweak those vaults manually and just get rid of the greys? 17:26:30 dpeg: Then they only have plain, green and orange rats. 17:26:37 dpeg: Besides: spammals. 17:26:42 dpeg: in the sewer, they serve a purpose of being "buffed rats" 17:28:17 wow, this neck-and-neck race for the Ascii Dreams vote is *really* exciting 17:28:29 ASCII dreams vote? 17:28:50 http://roguelikedeveloper.blogspot.com/ 17:29:17 T.O.M.E. and Crawl are tied for first place 17:29:26 Oh, I already voted. 17:29:29 ... T.O.M.E? 17:29:36 jpeg: I'd like to know how many of those votes are fakes :) 17:29:50 dpeg: yeah, I wonder about that myself 17:30:11 jpeg: I'm designing a murder mystery roguelike :) Randomly generated manor house, guest and plots... and then one of them is murdered! 17:30:23 Napkin: my win32 builds work correctly, including save+load+quit. So either the bug was fixed in Crawl or this version of mingw32 doesn't trigger it 17:30:29 dpeg: I mean in 2008 we somehow won with hardly anyone knowing about the poll 17:30:39 due: cool! :D 17:30:52 due: do you need a co-developer? 17:31:08 jpeg: So hopefully I get to be on next year's list! 17:31:14 due: I'm infamous for jumping ship to join small projects :P 17:31:18 Do you know Python? :) 17:31:27 not yet 17:31:55 good 17:32:00 Well, I haven't really started, but I've been reading a thesis on Bayesian filtered knowledge and random plot generation... It's all just ideas in my head. :) 17:32:01 I'd love to play a murder mystery roguelike 17:32:15 The interface will be very inspired by crawl. :) 17:33:09 due: how will it work? dialogues? collecting evidence? cutting down on suspects by killing them? 17:33:17 HAHA! 17:33:18 due: nothing wrong about that! :D 17:34:35 Basic methodology will be to generate the house, then the number of guests, the number of staff (1930s English manor house)... then use that number to generate a pool of possible suspects; each suspect will have characteristics randomly chosen (height, weight, shoe size, fingerprints(always unique), hair colour, eye colour, etc)... 17:35:17 Weapons, of course. Then randomly pick these, create a trail of "true" clues; then go through each of the other suspects and make sure that there is at least one clue that doesn't match: wrong fingerprint, wrong shoe size, wrong alibi, and so-on. 17:35:36 So yeah... collecting clues, interrogating and corroborating allibis... 17:36:10 due: sounds brilliant, I love mysteries! 17:36:21 due: sadly no one I know likes to play Clue 17:37:07 jpeg: Try "Mystery Express"! It is like Clue but done brilliantly and set on the Orient Express. :D 17:39:19 CrawlHashTable ought to be renamed at some point 17:39:30 sorear: CrawlHashTableOfDoom, clearly. 17:39:37 CrawlRBTree 17:44:57 -!- psyshvl has quit [Quit: stuff] 17:46:48 yay rebuilding tiles! 17:48:06 jpeg: Those hellbinder tiles look *awesome*. 17:48:15 :D 17:48:32 (Just looking at floor.png; have to recompile everything to see in-game.) 17:48:33 pity I won't ever get to see the vault for reals 17:48:35 I think spriggan riders 'dying' offscreen have a message leak 17:49:29 pointless_: really? 17:49:47 I got "it jumps down from its dead mount" 17:49:49 due: is the black also filtered from your floor.png? it's kinda hard to tell how the tile will look in-game like that 17:50:22 03dpeg * r7f79a64adf51 10/crawl-ref/source/dat/des/portals/sewer.des: Five new sewer maps (co). 17:50:34 jpeg: Linley must have felt the same with Pan and the hells :) 17:50:35 pointless_: Ah, fixed! 17:50:53 I think out of sight returning weapons have a message leak; or at least they did recently 17:50:53 :/ 17:50:54 back 17:51:03 so how do I kill the slake moth in sprint-IV? 17:51:08 ah well, there's always wizard mode... 17:51:20 jpeg: Yes, I've never had a wizlab myself :( 17:51:46 jpeg: This is why I like early portal vaults so much :) 17:52:01 I need to do a swamp ossuary with modified bog mummies. 17:52:18 due: "sickly" won't fit, though? 17:52:26 Keskitalo: Do you remember what UP 6.2. "Tooltip erases mouseover short description" means? 17:52:39 03due * re3df72a713fe 10/crawl-ref/source/ (areas.cc showsymb.cc spl-cast.cc status.cc): Tweak Leda's Liquefaction (OG17, others); fixes #3086. 17:52:40 dpeg: "rotted bog mummy"; to denote its lack of throw frost and so-on. 17:52:40 03due * rfe227795de4e 10/crawl-ref/source/ (8 files in 5 dirs): Replace grey rats with porcupines in most instances. 17:52:45 03due * r8b8ebcebdf77 10/crawl-ref/source/monster.cc: Fix message leak for Spriggan riders (pointless). 17:52:51 dpeg: Corpse rot and slow are thematically good; throw frost is a bit scary early. 17:52:53 jpe: that was probably my pet peeve, let me look 17:52:55 @??bog mummy 17:52:55 bog mummy (07M) | Speed: 8 (swim: 140%) | HD: 6 | Health: 31-47 | AC/EV: 1/9 | Damage: 25, 412(cold:6-17) | Flags: 07undead, amphibious, evil, spellcaster | Res: 06magic(40), 05fire, 02cold, 03poison | XP: 273 | Sp: throw frost (3d6), slow, corpse rot. 17:52:55 Keskitalo: the description or the item name, both appear on mouse-over 17:53:13 Though perhaps they are not too scary. 17:53:19 @??porcupine 17:53:19 unknown monster: "porcupine" 17:53:21 Meh. 17:53:48 due: yeah, the only portal vault I get to see more often than once in a blue moon are the Sewers, which I love 17:53:49 OG17: Okay, now it's "SlowM" and works anywhere with a solid floor! 17:53:54 jpeg: ah, now i remember. the proble was that when tooltip pö 17:54:03 tooltip appears, the description disappears 17:54:16 on returning and message leaking, it's still there in 0.8.0-a0-4115-g6282fd3 (I gave a monster some javelins of returning, cast spammals, drew tomb, waited, and got "The weapon returns whence it came from!") 17:54:25 leading to a too brief display of the description 17:54:46 jpeg: we just got some awesoe looking sewer from co on the tracker 17:54:53 Keskitalo: so is it fixed? 17:55:03 jpeg: I guess so! :) 17:55:11 jpeg: I like it that we got early portal vaults into the game before the poll proved their relevance =) 17:55:13 can't (be arsed) to test right now 17:55:13 monky: Should check that you can see the target too, I think. 17:55:25 Keskitalo: I just tested and pushed them?! 17:55:32 due: in my current game I finally saw my first "in-game" bazaar 17:55:35 dpeg: <3 !!!! 17:55:36 jpeg: hooray! 17:55:42 due: I don't remember whether it was one of mine though 17:56:54 oh and I also stepped away before drawing tomb so there were no gaps 17:57:50 monky: Thanks, fixed 17:57:55 Keskitalo: very long descriptions can still be partly covered by the tooltip if it's one of the leftmost items 17:57:58 monky: It always prints a message on returning 17:58:13 monky: The message is just modified depending on whether or not you can see the thrower. 17:58:19 oh 17:58:46 So now I've made it that you have to either see the thrower or the location of the throwee. 17:58:47 Keskitalo: so the tooltip placement could still be improved 17:58:53 which makes more sense 18:00:40 jpeg: all right 18:01:03 I'm in th emiddle of a tiles recompile, so hopefully I didn't break anything :( 18:01:18 Tile for the new "Lurking Horror" monster. (https://crawl.develz.org/mantis/view.php?id=3091) by co 18:01:20 Oh good, didn't. 18:01:24 03due * r735fb89997a8 10/crawl-ref/source/mon-act.cc: Fix returning weapon message leak (monky). 18:01:50 jpeg: I am in *love* with all these tiles! I haven't even finished thrashing out the abyssal horror concepts and we have tiles! :D 18:02:14 what a cute little horror 18:02:38 whoa, I switch away from my browser for a minute, and what happens? 18:02:39 Can someone link the tiles? I will never see them in play. 18:03:10 dpeg: one of these days, I'll push you down in front of my notebook and force you through a quick course of the dungeon :p 18:03:18 eek 18:03:43 -!- dpeg is now known as debegg 18:03:51 what? 18:04:07 * debegg is camouflaged now 18:04:19 lurking horror's tile is the MITD :) 18:04:26 * jpeg can't stop giggling 18:04:50 with my nick I'm probably destined to prefer the graphical interface 18:05:24 galehar: MITD? 18:05:29 http://oots.wikia.com/wiki/Monster_In_The_Darkness 18:05:42 we can probably find a use for that tile... but it's not really enough of a horror for my liking (: 18:05:53 is it a grue? 18:06:17 the mitd? 18:06:29 Oh! I remember 18:06:30 it seems like a great tile for shadows, if they don't already have a good one 18:06:34 -!- Moriasc has joined ##crawl-dev 18:06:45 I found a great set of unfortunately CC-licensed tiles... 18:07:07 http://pousse.rapiere.free.fr/tome/ 18:07:19 Mu_: maybe swap them? 18:07:46 Mu_: http://pousse.rapiere.free.fr/tome/tiles/DO/Monsters/ElementalVoid.PNG <-- that's what I envision when you say starcursed mass. 18:08:09 nice 18:09:21 http://pousse.rapiere.free.fr/tome/tiles/DO/tome-domonsterstiles.htm rather 18:10:19 They're CC-licensed. 18:12:40 OMG. Trolls with machine guns! 18:12:42 enne: I see your point. 18:14:10 ... jpeg: You are *amazing*. Those are *amazing*. 18:14:17 Mu_: Have you seen jpeg's tile version of the hellbinder sigils? 18:14:20 that void elemental reminds me a little of the vastness from dom3 http://dom3.servegame.com/wiki/Vastness 18:14:29 no, where can i see them? 18:14:29 due: the new title screen has trolls with machine guns (at least at first glance) 18:15:24 you have to give us links 18:15:26 Mu_: I'll do a screenshot 18:15:32 I want to see the machine gun trolls too! 18:15:43 and the hellbinder sigilse 18:16:00 debegg: play tiles ;) 18:16:13 I can't! 18:16:23 * jpeg whispers, "It's all part of the plan..." 18:16:32 haha 18:16:54 Heck, the '1' glyph contains more emotions than any of your tiles anyway :P 18:16:57 will screenshot 18:18:46 https://crawl.develz.org/mantis/file_download.php?file_id=1162&type=bug 18:19:03 that's trolls 18:19:14 galehar: you were a second faster than me :) 18:19:25 :) 18:19:29 Where are the machine guns? My fellow trolls look exactly like this. 18:19:45 the one on the right, with the crossbow 18:19:47 those are surely orcs :P 18:19:55 trolls can't wear gloves :P 18:20:06 You all have spent too much time watching Rambo :) 18:20:19 debegg: the right one is wielding something that's apparently a crossbow, but *looks* like a machine gun 18:20:33 sure, I saw that :) 18:20:37 http://jeah.net/jmc/sigil1.png 18:20:38 why aren't they wearing headbands anyway? 18:20:44 http://jeah.net/jmc/sigil2.png 18:20:47 http://jeah.net/jmc/sigils3.png 18:20:49 http://jeah.net/jmc/sigils4.png 18:20:51 http://jeah.net/jmc/sigils5.png 18:21:01 "Sorry, the file you requested is not here!" 18:21:02 they all 404 for me :P 18:21:07 oh meh 18:21:13 http://jeah.net/~jmc/sigils5.png 18:21:13 same here :( 18:21:16 http://jeah.net/~jmc/sigils4.png 18:21:18 http://jeah.net/~jmc/sigils3.png 18:21:20 http://jeah.net/~jmc/sigil2.png 18:21:22 http://jeah.net/~jmc/sigil1.png 18:21:28 Tilde makes aalllll the difference. 18:21:41 oh wow that's great 18:21:52 I KNOW. 18:22:03 oh, that's what the message overlay looks like? neat 18:22:07 by the way, this title screen has been proposed too: https://crawl.develz.org/mantis/file_download.php?file_id=1328&type=bug 18:22:27 jpeg: Possibly; I'm using an EEE PC with a tiny resolution. 18:22:32 beautiful 18:22:43 does the console vault have any visual fluff? 18:22:48 galehar: I don't like it - compared to the other screens it's a bit too bland 18:23:01 debegg: Yes, the sigils look great in console. 18:23:10 debegg: yellow squares where tiles got the patterns 18:23:15 * debegg is relieved. 18:23:20 :) 18:23:31 due, debegg: they look better in tiles :/ 18:23:37 May I suggest that we keep the pixelated logo for all screens? 18:24:12 debegg: what do you mean? 18:24:14 It fits an anachronist game like a roguelike. 18:24:27 the trolls had the crude logo 18:24:56 well, I think the other logo fit better with the other images. 18:25:01 03dpeg * rf51c8e834ffa 10/crawl-ref/source/dat/des/portals/sewer.des: Reduce sewer loot. 18:25:20 with the different colours and styles, mixing the logos would look weird 18:25:44 meh, it doesn't really fit the art, though it looks well with the demonic one 18:25:45 http://crawl.develz.org/wordpress/wp-content/uploads/2010/12/tiles_screenshot_splash_screen2.png 18:25:56 but what about the corporate identity? 18:26:40 ok, but we may have to at least change the colour sometimes 18:26:41 Hm, doll creator version of Jory not very nice. 18:26:42 it's been suggested to smuggle a cauldron into every splash screen 18:27:10 but that doesn't make sense and devalues the original art 18:27:12 jpeg: I loved denzi's little cauldron-on-a-sign thing. 18:27:19 maybe we can make a logo with the cauldron in the title 18:27:23 yes, I like it if this some consistency (cauldron: thematic plus "soup", same font for all screens: visual) 18:27:47 jpeg: Ouch, you're right about the silence overlay. :/ 18:28:25 galehar: you mean like google does, replace the o with a cauldron pic? 18:28:39 galehar: or let the tavern icon dangle from the C? 18:28:58 that's an idea. It could be done in any other way 18:29:07 due: it's a nice tile, it just doesn't convey the message 18:29:13 this one is even better :) 18:29:54 debegg: how about we make an implementable for drawing a logo with the cauldron or another cool icon in it 18:30:01 then we post it on the forum 18:30:08 an let users compete :) 18:30:36 there are plenty of people submitting tiles recently 18:30:47 -!- ortoslon has joined ##crawl-dev 18:31:30 crash+assert on taking stairs (https://crawl.develz.org/mantis/view.php?id=3092) by ortoslon 18:32:01 galehar: +1 18:32:12 someone uploaded splash screen variants that each had the logo's cauldron somewhere in the front 18:32:25 but they were so awful they've been removed from Mantis again 18:33:09 also denzi signatures look out of place on splash screens 18:33:40 ortoslon: we can't simply remove them, can we? 18:33:57 you can ask denzi 18:34:17 I don't mind having the artist's signature 18:34:33 think of them as title screen, not as splash screens :) 18:34:50 galehar: nor do I, it's not like it's extravagant or something 18:35:16 anyway, I'm going to bed 18:35:18 bye! 18:35:36 -!- galehar has quit [Quit: disconnect] 18:39:13 jpeg: Does adding new flags for backgrounds mess stuff up/ 18:45:50 I'm not sure what happens if we add more of them 18:46:11 it looks like there's space to add one after silenced... 18:46:29 it goes 0x01, 0x02, 0x04, 0x10... 18:46:41 Oh, but the next ones are mutually exclusive... bah. 18:46:42 which file is this again? 18:46:50 enum.h 18:46:58 I didn't read the comment though 18:47:39 yeah, mutually exclusive was the only way I could squeeze all of them in 18:47:59 because I'm afraid to break something by adding more bits than we can fit 18:48:12 What's it stored as? 18:48:52 no idea 18:49:02 all I did was copy and paste 18:49:49 I guess we could try it and see what happens 18:50:05 due: what are you trying to add? 18:51:09 Just another flag for liquefaction... only if I can make the effect look right though. 18:52:14 it might help if some of the flags could be decoupled but we're comparing against TILE_FLAG_MASK all over the place 18:52:17 -!- MarvinPA has joined ##crawl-dev 18:52:33 Yeah. I'm making some ripple/wave overlay tiles. 18:52:52 tileidx_t 18:52:58 The mask needs to fit into that... 18:54:08 tileidx_t is an unsigned int. 18:54:20 Yeah. 18:54:45 So, 32 bits. 18:55:33 If you want to refactor the removal of flags to give every cell a stack of tiles, I wouldn't complain. :) 18:57:38 0x8000000 = 28 bits, right? 18:58:36 Which means that we've space for another four flags! Hooray! 18:59:21 What flags are you adding? 18:59:25 (Out of curiosity.) 18:59:30 Just a flag for the liquefaction effect. 18:59:35 IFF it actually looks good. 18:59:40 If it doesn't I'll implement it some other way. 18:59:47 -!- casmith789 has quit [Ping timeout: 240 seconds] 18:59:57 due: it is? I thought 4*8 = 32, how do you arrive at 28? 19:00:21 I never understood this bit/byte mathematics 19:01:05 Converted 0x8000000 to binary then counted the digits ;) 19:01:23 each hex digit is 4 bits 19:01:35 However 19:01:52 Anyway 19:02:02 DungeonCellBuffer::pack_background gets passed the location of each square, correct? 19:02:17 but: TILE_FLAG_KRAKEN_SW = 0x80000000, 19:02:36 that's 8 digits, right? 19:02:38 jpeg: Ah. Yes. 19:02:44 -!- casmith789 has joined ##crawl-dev 19:02:49 Off-by-one error ;) 19:02:53 :) 19:03:00 void DungeonCellBuffer::pack_background(int x, int y, const packed_cell &cell) 19:03:12 That x, y corresponds to grd(x, y), yeah? 19:03:32 I think it corresponds to view space, not grid space. 19:03:36 Ah. 19:03:39 (Maybe?) 19:03:43 yeah, it's converted somehow 19:04:07 If we could translate it to grid space we could dump the flag application in tile_apply_properties, dump the flags, and just ccheck silenced(grid_space(x, y)), yes? 19:04:10 here: const coord_def ep = grid2show(gc); 19:04:19 there's also show2grid 19:04:24 That would remove... 19:04:41 TILE_FLAG_BLOOd, TILE_FLAG_MOLD, TILE_FLAG_SANCTUARY, TILE_FLAG_SILENCED and TILE_FLAG_HALO... I think. 19:04:59 due: I suppose you could, but it'd be one more thing that will have to be un-done if anybody ever gets around to net-tiles. 19:05:13 enne: Bah ,true. 19:05:29 wasn't the plan to decouple display and internal stuff? 19:05:34 yes, that 19:05:55 Yeah, exactly. So, if you inquire into the internals in the display routines, then you only make the coupling worse. 19:05:59 nights! 19:06:12 -!- debegg has quit [Quit: leaving] 19:06:35 Yeah, I wasn't really thinking about nettiles. 19:07:00 So... extend packed_cell, perhaps? 19:07:45 some of those "background" tiles are actually overlays of some kind (ray, oor) 19:08:21 I don't really understand the code well enough; decouple tile_flags into tile_foreground_flags tile_background_flags? 19:08:28 Oh. They already are! 19:08:35 ... and I'd be immensely happy if anyone comes up with an alternative way to keep track of kraken transitional tentacle tiles 19:08:58 yes, I'm wondering if it's possible to introduce a third tier 19:10:15 but I'm afraid it wouldn't even free up a lot of bits 19:11:14 enne: I'm pondering how to make clouds appear behind monsters 19:11:39 "// a foreground or background flag, or add a member to packed_cell. 19:12:06 Perhaps we should just make draw_bloody, draw_silenced, draw_moldy, etc, boolean members of packed_cell? 19:14:17 That seems reasonable. 19:14:20 Not sure how easy that is or where the packed cell is initialised; the apply_properties function works off a tileidx_t. 19:14:33 Oh! 19:14:50 It calls apply_properties(cell->bg). Easy to convert to apply_properties(cell). 19:15:35 packed_cells are marshalled, I presume? 19:16:26 in the save? I don't think so 19:16:50 That makes it easier. 19:16:52 then again, somehow we do arrive at the cache for remembered squares, so probably yes 19:16:59 enne? 19:17:53 packed_cells get recalculated from the internal format of remembered cells. 19:17:56 They're not marshalled. 19:18:09 good to know, thanks! 19:18:50 Hm. 19:19:03 My mistake; the cell referred to in apply_properties is a screen_cell_t. 19:21:52 -!- coolio_mc has quit [Read error: Connection reset by peer] 19:22:08 Right; it then initialises the packed cell from screen_cell_t. 19:23:12 while this discussion is really fascinating, it's terribly late here and I can't really think anymore 19:23:18 time for bed 19:23:18 Yeah :) 19:23:46 tell me if you come up with anything good 19:23:50 Will do :) 19:24:25 always fun to stop by on IRC 19:24:32 though I do wonder where all the other devs are 19:24:38 * due in Australia. :0 19:24:40 *:) 19:24:41 did the tile talk scare them away? ;) 19:24:50 Either we modify screel_cell_t and pack_cell to have the same boolean flags and these can be transferred when initilising, or we can put a packed-cell in screen_cell_t. 19:25:00 But I haven't looked at the code enough to really comment yet! 19:25:11 anyway, night 19:25:16 Night! 19:25:24 -!- jpeg has quit [] 19:26:38 -!- syllogism has quit [] 19:28:42 -!- MarvinPA has quit [Ping timeout: 265 seconds] 19:31:50 Anti-trained skills (Ice) that overtake the responsible skill (Fire) are still marked as receiving a training malus (https://crawl.develz.org/mantis/view.php?id=3093) by smock 19:41:52 -!- Mu_ has quit [Quit: Defecator, may everything turn out okay so that you can leave this place.] 19:49:50 -!- pointless_ has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101203075014]] 19:53:14 -!- Textmode has joined ##crawl-dev 20:07:15 -!- ortoslon has quit [Quit: bye] 20:14:50 -!- MarvinPA has joined ##crawl-dev 20:17:48 -!- MarvinPA has left ##crawl-dev 20:17:50 -!- MarvinPA has joined ##crawl-dev 20:57:06 -!- psyshvl has joined ##crawl-dev 21:51:21 -!- psyshvl has quit [Quit: ChatZilla 0.9.86 [Firefox 3.6.6/20100625223402]] 22:06:14 -!- ortoslon has joined ##crawl-dev 22:32:56 -!- Twinge has quit [Ping timeout: 250 seconds] 22:51:25 -!- Twinge has joined ##crawl-dev 23:04:46 -!- casmith789 has quit [Remote host closed the connection] 23:07:08 -!- ortoslon has quit [Quit: bye] 23:37:22 Windows development builds on CDO updated to: 0.8.0-a0-4194-gf51c8e8 23:49:15 -!- Twinge has quit [Ping timeout: 246 seconds] 23:53:46 -!- Twinge has joined ##crawl-dev