00:01:15 Stable (0.14) branch on crawl.s-z.org updated to: 0.14-b1-21-g62f6471 00:04:18 -!- Quashie_ has quit [Ping timeout: 240 seconds] 00:04:32 Unstable branch on crawl.s-z.org updated to: 0.15-a0-55-g79b3258 (34) 00:04:51 -!- Kaput has quit [Quit: leaving] 00:10:58 -!- Mad_Wack has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…] 00:13:38 -!- buzzykins has quit [Ping timeout: 240 seconds] 00:16:18 -!- eb has quit [] 00:28:11 -!- rast has quit [Quit: ChatZilla 0.9.90.1 [Firefox 27.0.1/20140212131424]] 00:28:30 -!- rast has joined ##crawl-dev 00:36:29 ancient lich (16L) | Spd: 10 | HD: 27 | HP: 86-132 | AC/EV: 20/10 | Dam: 2013(drain) | 07undead, 10doors, evil, spellcaster, see invisible | Res: 13magic(immune), 05fire, 02cold++, 10elec, 09poison+++, 12drown, 04rot+++, 13neg+++, 13torm | Vul: 08holy++ | XP: 7493 | Sp: b.cold (3d37), paralyse, sum.greater demon, animate dead, iron shot (3d44), 04esc:teleport self / b.draining (3d33), slow, invisibility, throw icicle (3d37), crystal spear (3d48) / b.fire (3d37), confuse, haste, b.draining (3d33), sum.greater demon, 04esc:banishment / mystic blast (3d30), b.cold (3d37), invisibility, animate dead, o.destruction (9d20) | Sz: Medium | Int: high. 00:36:29 %??ancient lich 00:36:35 need more spells 00:40:02 -!- ayutzia_ is now known as ayutzia 00:42:57 -!- BirdoPrey has quit [Quit: BirdoPrey] 00:43:59 -!- demiskeleton has quit [Quit: Leaving] 00:44:19 -!- ayutzia has quit [Quit: Page closed] 00:45:00 -!- N78291 has quit [Quit: null] 00:47:45 -!- UncertainKitten has quit [Quit: Connection reset by pier.] 00:49:38 -!- wat1 has quit [Ping timeout: 240 seconds] 00:55:16 -!- NomadJim has quit [Read error: Connection reset by peer] 01:02:41 -!- Thalfon has quit [Quit: Page closed] 01:05:46 -!- nooodl has joined ##crawl-dev 01:06:23 -!- Keanan has joined ##crawl-dev 01:14:15 -!- archaeo_ has joined ##crawl-dev 01:15:15 -!- archaeo has quit [Ping timeout: 245 seconds] 01:16:01 -!- rast has quit [Read error: Connection reset by peer] 01:16:14 -!- rast- has joined ##crawl-dev 01:16:15 -!- rast- is now known as rast 01:18:08 -!- wat2 has quit [Ping timeout: 268 seconds] 01:18:43 -!- SkaryMonk1 has joined ##crawl-dev 01:19:42 -!- SkaryMonk2 has quit [Ping timeout: 255 seconds] 01:27:25 -!- mong has quit [Remote host closed the connection] 01:28:41 -!- ebarrett has quit [Quit: I quit] 01:31:03 -!- NomadJim has quit [Read error: Connection reset by peer] 01:32:49 -!- archaeo_ is now known as archaeo 01:33:42 -!- Keanan1 has joined ##crawl-dev 01:34:52 -!- Keanan2 has joined ##crawl-dev 01:35:27 -!- Keanan has quit [Ping timeout: 255 seconds] 01:37:54 -!- Keanan1 has quit [Ping timeout: 240 seconds] 01:39:55 -!- nooodl has quit [Quit: Ik ga weg] 01:44:43 -!- ontoclasm has left ##crawl-dev 01:48:49 -!- PleasingFungus has quit [Ping timeout: 240 seconds] 01:53:08 -!- raskol has quit [Ping timeout: 240 seconds] 01:54:58 -!- st_ has quit [Ping timeout: 252 seconds] 01:55:14 -!- wat1 has quit [Ping timeout: 240 seconds] 01:56:12 -!- Crehl has quit [Quit: qqqqqqq] 01:56:50 -!- icantfindaname has quit [Client Quit] 01:57:00 -!- neddim has quit [Remote host closed the connection] 02:00:39 -!- tksquared has quit [Ping timeout: 255 seconds] 02:04:48 -!- notcluie has quit [Ping timeout: 245 seconds] 02:11:25 -!- NomadJim has quit [Read error: Connection reset by peer] 02:13:38 -!- wat2 has quit [Ping timeout: 240 seconds] 02:14:52 -!- PoopBridge has quit [Ping timeout: 268 seconds] 02:14:53 -!- [1]PoopBridge is now known as PoopBridge 02:15:13 -!- wat1 has quit [Ping timeout: 240 seconds] 02:16:02 -!- Rewans has quit [Quit: Page closed] 02:16:37 Stable (0.14) branch on crawl.beRotato.org updated to: 0.14-b1-21-g62f6471 02:23:19 Unstable branch on crawl.beRotato.org updated to: 0.15-a0-55-g79b3258 (34) 02:27:51 -!- Patashu has joined ##crawl-dev 02:29:24 -!- lobf has joined ##crawl-dev 02:45:10 -!- Quashie has quit [Quit: Leaving] 02:45:38 -!- tabstorm has quit [Ping timeout: 245 seconds] 02:46:04 -!- lobf has quit [Quit: lobf] 02:48:21 -!- omnirizon has quit [Ping timeout: 255 seconds] 02:55:52 -!- Patashu[Zzz] has joined ##crawl-dev 02:55:52 -!- Patashu has quit [Disconnected by services] 02:59:13 -!- SkiChan has quit [Quit: Page closed] 02:59:47 -!- Cabadath is now known as Red_Buttons 03:01:51 -!- Mateji has quit [Ping timeout: 255 seconds] 03:03:12 -!- gnum has quit [Ping timeout: 255 seconds] 03:06:21 -!- wat2 has quit [Ping timeout: 255 seconds] 03:07:08 -!- wat1 has quit [Ping timeout: 240 seconds] 03:11:48 -!- Elynae has quit [Client Quit] 03:12:12 -!- Netmonmatt has quit [Ping timeout: 268 seconds] 03:15:59 -!- rast has quit [Read error: Connection reset by peer] 03:16:22 -!- rast has joined ##crawl-dev 03:18:55 -!- SkaryMonk2 has joined ##crawl-dev 03:19:58 -!- SkaryMonk1 has quit [Ping timeout: 240 seconds] 03:21:03 -!- Zicher has quit [Ping timeout: 245 seconds] 03:21:19 -!- Zicher_ is now known as Zicher 03:26:30 -!- simmarine has quit [Quit: Leaving] 03:27:22 -!- jeanjacques has joined ##crawl-dev 03:32:38 -!- Red_Buttons is now known as Big_Red 03:33:07 -!- Big_Red is now known as redsquadronleade 03:34:00 -!- redsquadronleade is now known as Hobgoblin 03:34:22 -!- Hobgoblin is now known as thro 03:35:58 -!- scummos| has quit [Quit: Konversation terminated!] 03:37:15 -!- Keanan2 has quit [Quit: Leaving.] 03:38:32 -!- bmfx has quit [Ping timeout: 252 seconds] 03:41:13 -!- ussdefiant has quit [Ping timeout: 240 seconds] 03:41:43 -!- Kalir has quit [Quit: CADENZA-CLASS CLOCKWORK KNIGHT--OPERATION TERMINATED] 03:44:41 -!- Ragdoll has joined ##crawl-dev 03:44:55 -!- thro is now known as PyroDash 03:55:33 -!- Zicher has quit [Quit: Page closed] 03:59:23 -!- conted has quit [Ping timeout: 245 seconds] 04:08:24 -!- bonghitz has quit [Remote host closed the connection] 04:11:08 -!- NomadJim has quit [Read error: Connection reset by peer] 04:11:34 -!- wheals has quit [Ping timeout: 265 seconds] 04:16:48 -!- tali713 has joined ##crawl-dev 04:31:54 -!- dpeg has joined ##crawl-dev 04:32:22 -!- minqmay has quit [Remote host closed the connection] 04:32:42 -!- dpeg_ has joined ##crawl-dev 04:32:48 -!- dpeg has quit [Client Quit] 04:34:33 Hi! I would like to suggest (again?) that teleportation traps have limited uses. The idea is old, and originally to prevent teletrap luring. However, it would also rid us of having to be more careful about teletrap placement, as in 04:34:37 crawl.develz.org/tavern/viewtopic.php?f=8&t=11641 04:36:38 -!- gnum has quit [Ping timeout: 240 seconds] 04:38:04 -!- wheals has joined ##crawl-dev 04:55:11 -!- scummos| has quit [Ping timeout: 268 seconds] 05:04:47 -!- dtsund has quit [Ping timeout: 255 seconds] 05:05:38 -!- ChongLi has quit [Ping timeout: 240 seconds] 05:12:06 -!- NomadJim has quit [Read error: Connection reset by peer] 05:13:08 -!- ToBeFree has quit [Ping timeout: 252 seconds] 05:16:18 -!- rast has quit [Read error: Connection reset by peer] 05:16:37 -!- rast has joined ##crawl-dev 05:17:52 -!- HellTiger has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/] 05:19:04 -!- SkaryMonk1 has joined ##crawl-dev 05:19:54 -!- SkaryMonk2 has quit [Ping timeout: 240 seconds] 05:25:37 dpeg_: i think its a good idea 05:26:08 -!- radinms has quit [Ping timeout: 240 seconds] 05:26:51 but now lunch is over so cant elaborate atm ;) 05:31:53 -!- Lawman0 has quit [Ping timeout: 246 seconds] 05:33:50 -!- ToBeFree has joined ##crawl-dev 05:34:32 -!- Blazinghand has quit [Read error: Connection reset by peer] 05:36:02 -!- Red_Bucket has quit [Quit: Page closed] 05:37:33 -!- scummos| has quit [Client Quit] 05:38:50 -!- trystero_ has quit [Quit: Page closed] 05:42:01 -!- Ragdoll has quit [Ping timeout: 240 seconds] 05:42:20 -!- Ragdoll has joined ##crawl-dev 05:42:34 -!- nyfair has quit [Quit: No Pic U Say A JB!] 05:49:36 -!- Lasty1 has joined ##crawl-dev 05:53:06 -!- lessens has quit [Ping timeout: 252 seconds] 05:55:24 -!- Ququman has joined ##crawl-dev 05:59:30 -!- PyroDash has joined ##crawl-dev 05:59:40 dpeg...isn't here :( 05:59:54 huh? 05:59:59 * dpeg_ is not dead yet. 06:00:08 -!- lessens_ has quit [Ping timeout: 240 seconds] 06:01:02 -!- PyroDash is now known as Bcadren 06:01:38 Bcadren: I saw your Beogh proposal. 06:02:04 I don't think if Beogh is underpowered, though. 06:02:51 dpeg_ Yea that's what I was looking for you for. I don't think it's ready to go as is and now that I wrote it I think it might be more interesting as a different god. *shrug*. Though adding 'hymn' to an orc invocation; I'd definately push; the spellcasting stuff feels like it could go to a different god. 06:03:06 (Slightly Ot, but I have a feeling you are mostly thinking like a player when it comes to design: "when in doubt, buff it".) 06:03:31 Beogh became much more playable with dungeon-wide Recall in 0.12 or so. 06:03:34 !hs hanon12 06:03:35 1344. hanon12 the Slayer (L23 LOFi of Beogh), escaped with the Orb and 15 runes on 2014-01-19 13:42:12, with 47803653 points after 30139 turns and 6:03:03. 06:03:49 30k turns :O 06:03:59 but LO 06:05:10 Bcadren: also, how can you get rid of water walking? It's a gimmick, but also a trademark power. Everyone has to smile when they see it for the first time (thank Trog that no evangelists play Crawl yet). 06:05:30 The allies portion is less helpful in Hell, etc. because of everyone being tormentable. the Smiting...I don't think it's worth training Invo for it in most cases...you've be better off going AE of Beogh and using Airstrike; which does roughly the same thing. 06:05:54 Bcadren: we don't care about hell. If a god is not perfect for optional stuff, it's not a concern. 06:05:57 again, bcadren: hanon's run should be proof enough that Beogh's problem, if any, is in the interface, not the power level 06:06:04 plus what archaeo says 06:06:09 -!- Nunya has quit [Remote host closed the connection] 06:06:12 dpeg_ I didn't. I also didn't mention the bless orcish allies or orc gifts he gives; they are long to explain and would have muddied it up. 06:06:25 archaeo: do you think there's still grave interface worries? 06:06:49 dpeg_: hardly as grave as they're often made out to be, but it's one of the clunkiest interfaces in crawl 06:07:23 archaeo: Beogh is one of the reasons why I think about suggesting "monsters don't pick up items" for 0.15. 06:07:47 (Of course, this means that Beogh has to provide alls boosts and resistances, of course.) 06:07:53 erm, of of course course 06:08:05 beogh giving straight upgrades to equipment seems like it might play better 06:08:16 archaeo he already does that. 06:08:48 bcadren: not in the same sense as in dpeg's proposal, I think 06:08:57 Bcadren: my point is that orcs wouldn't pick up stuff whatsoever. That's obviously an interface improvement and a nerf (hence the need for Beogh compensation). 06:09:06 randomly applies an enchant weapon III or enchant armor when your allies kill something. 06:09:27 Bcadren: allies have to get access to resistances in some form. 06:09:37 bcadren: that's nothing compared to a player micromanaging warlord equipment to have a bunch of optimal tanks 06:10:03 How do you manage equipment? 06:10:25 the ctrl-t menu lets you tell your allies how to manage equipment, I think 06:10:28 and before anyone here says it: there won't be a Give or Take command 06:10:50 archaeo: yes, global behaviour (pick up items from grounds etc.) 06:10:53 it's bad enough managing one inventory :\ 06:10:54 -!- Nethris has quit [Quit: Leaving.] 06:10:56 !tell PleasingFungus I really liked the SA discussion about the -5 MR++ cloak. 06:10:56 dpeg_: OK, I'll let pleasingfungus know. 06:11:18 SA 06:11:19 archaeo: you won't get more than 52 item slots either :) 06:11:22 ? 06:11:38 bcadren: somethingawful, a popular forum 06:11:57 SomethingAwful, has a Crawl forum with juvenile/imbecile/obnoxious tendencies 06:12:06 has gotten better recently, thanks to folks like PleasingFungus 06:13:42 The Orcish Hymnal is something that Warlords currently do (enemy version is +3 Slaying bonus to orcs with lower HD) The Seraph of Holy Pan also has it. 06:13:46 -!- rchandra has joined ##crawl-dev 06:14:06 seraph (13A) | Spd: 10 | HD: 25 | HP: 185-237 | AC/EV: 10/20 | Dam: 25, 10 | 08holy, 10weapons, 10items, 10doors, fighter, see invisible, fly, !sil | Res: 13magic(immune), 04fire+++, 10elec, 03poison, 04rot, 13neg+++, 08holy | XP: 5776 | Sp: minor healing (2d12) | Sz: Medium | Int: high. 06:14:06 %??Seraph 06:14:27 It's some weird passive; not a spell. 06:15:57 Seraphs version works on Holy monsters only; Warlord's works on Orcs, only. (not Ogres or Troll if he spawned with some). 06:16:32 Beogh isn't underpowered, but he is weaker than Yred, who is similar. 06:18:17 And Yred doesn't have the micromanagement (only gifts he gives that can wear armor and use equipment are ghouls and Profane Servitors; most his stuff can't) 06:19:28 as I tried to indicate, the micromanagement will be addressed (if we're lucky, in 0.15) -- I note that your proposal does nothing about orc ally interface 06:19:57 I don't think that Beogh needs further sweeping changes. Warlord chanting should work on allies, that is true. 06:21:53 As far as -too weak or too strong goes- I don't think there are any gods that are too weak (Save, maybe Sif)...for too strong...Chei and Lugonu. 06:22:18 why would there be a desire for gods to have a consistent "strength"? 06:23:40 True enough. I mostly think Beogh needs a buff because he and Yred are so similar; yet Yred has some excrutiatingly powerful invocations as well as the perma-allies and Beogh gets...just smiting (which is basically Airstrike that ignores AC). 06:27:25 Bcadren: Beogh does not need a buff, read the log of the character archaeo listed above. And what you suggest as buff on the forum is grossly overpowered: you cast Haste and 27 orcs get hasted!? 06:28:57 it's worth noting that the highest Yred score, in comparison, falls at 566 on the high score list 06:29:40 -!- Lasty1 has quit [Quit: Leaving.] 06:30:01 though that may have more to do with the popularity of Yred among turncount speedrunners 06:30:16 lack of popularity* 06:30:30 idk how to read the log. 06:31:51 I guess if there's one thing Beogh has on Yred; it's that you can get the allies much more quickly. (takes less time at high piety to wind up with powerful ones) 06:32:08 -!- BlackrayJack has quit [Ping timeout: 240 seconds] 06:32:14 Or rather. I don't know how to get IRC to display the log so I can read it. I'm sure I could read it if you did though. 06:32:42 http://kr.dobrazupa.org/morgue/trunk/hanon12/morgue-hanon12-20140119-134212.txt 06:33:34 -!- Patashu[Zzz] has quit [Ping timeout: 240 seconds] 06:39:36 dpeg_ curious? do you think haste is the only one of those that's too powerful? boosting charms to AoE might come up as another god because it seems like an interesting enough effect that people would want it if it was balanced right. [As for Beogh; I do stand by reducing the Piety cost of smite (3 seems high for something so relatively weak) and the hymnal invoke.]; though the rest of it seems like it needs to be rebalanced and prob 06:42:20 -!- NomadJim has quit [Read error: Connection reset by peer] 06:46:51 bcadren: you would have to put together some cherry-picked list of charms that you could use with the ability, with all kinds of messy special cases 06:48:37 and from what I can tell from hanon's log, hasting orcs one-by-one is already extremely powerful 06:49:34 Well the special cases are mostly because the player version of the spells doesn't work for monsters as is. Like there's no monster version of Ozo's Armor or any transmutation spell. And it's not really cherry picked; I mostly excluded ones that wouldn't do anything and grabbed boosts from all spellschools. 06:53:05 Bcadren: no, haste is just one glaring example 06:53:59 Crawl Light has the following neat effect: potions you drink affect nearby allies. Also smells of feature creep but it much, *much* less contentious than spells. 06:54:47 Some of the +/* Berserk artifacts affect all allies and that's like Might + Haste all. 06:55:12 ??sword of jihad 06:55:12 sword of jihad[1/1]: +12,+10 eudemon blade of holy wrath, berserkitis, rN+, EV+3, and stealth-. The berserkitis triggers on 3% of attacks. In 0.14+, going berserk (by any means) while wielding this sword will cause all natural-holiness allies in LOS to go berserk as well. 06:55:16 archaeo: cool idea on fleeing rune! 06:55:29 thanks dpeg_ 07:02:10 If their equipment micromanagement was removed; sharing weapon brand and resistances might balance the loss (and completely circumvent 'bad interface'). 07:04:48 -!- notcluie has quit [Ping timeout: 245 seconds] 07:06:32 -!- lessens has quit [Ping timeout: 268 seconds] 07:09:34 picking up daughter from school, later 07:09:50 Bcadren: yes, that would be one option 07:10:42 -!- archaeo has quit [Quit: Page closed] 07:11:37 -!- Sgeo has quit [Read error: Connection reset by peer] 07:15:08 -!- DrKe has quit [Ping timeout: 240 seconds] 07:16:25 -!- rast has quit [Read error: Connection reset by peer] 07:16:49 -!- rast has joined ##crawl-dev 07:18:21 -!- SkaryMonk2 has joined ##crawl-dev 07:19:38 -!- SkaryMonk1 has quit [Ping timeout: 240 seconds] 07:20:10 -!- nooodl has joined ##crawl-dev 07:21:41 -!- debo has quit [Quit: debo] 07:33:59 -!- conted_ has quit [Quit: Page closed] 07:46:22 -!- NomadJim has quit [Read error: Connection reset by peer] 07:47:01 -!- Mad_Wack has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…] 07:50:46 -!- Lightli has quit [Ping timeout: 245 seconds] 07:58:58 -!- wat2 has quit [Ping timeout: 240 seconds] 07:59:18 -!- DrKe has quit [Ping timeout: 240 seconds] 08:01:15 -!- trystero_ has quit [Quit: Page closed] 08:01:53 -!- DrinkMachine has joined ##crawl-dev 08:01:53 -!- DrinkMachine has quit [Changing host] 08:01:53 -!- DrinkMachine has joined ##crawl-dev 08:05:38 -!- notcluie_ has quit [Ping timeout: 245 seconds] 08:10:38 -!- DrKe has quit [Ping timeout: 240 seconds] 08:12:31 -!- SkaryMonk has quit [Ping timeout: 268 seconds] 08:13:32 -!- ToBeFree is now known as Guest12345 08:13:45 -!- Vaporware has joined ##crawl-dev 08:13:47 -!- Guest12345 is now known as ToBeFree 08:14:46 -!- Vaporware has quit [Client Quit] 08:15:15 -!- debo has joined ##crawl-dev 08:15:46 -!- SkaryMonk has joined ##crawl-dev 08:31:28 -!- SkaryMonk2 has quit [Quit: Leaving.] 08:33:42 LordofAtlantis (L10 MfFi) ERROR in 'tileweb.cc' at line 212: Socket write error: Resource temporarily unavailable (D:8) 08:33:43 Heini (L5 HOFi) ERROR in 'tileweb.cc' at line 212: Socket write error: Resource temporarily unavailable (D:3) 08:33:44 olXor (L12 GrFi) ERROR in 'tileweb.cc' at line 212: Socket write error: Resource temporarily unavailable (Lair:4) 08:33:45 iafm (L27 DECj) ERROR in 'tileweb.cc' at line 212: Socket write error: Resource temporarily unavailable (Zig:1) 08:33:46 undersol (L5 SpVM) ERROR in 'tileweb.cc' at line 212: Socket write error: Resource temporarily unavailable (D:3) 08:47:16 -!- SurpriseTRex_ has quit [Quit: SurpriseTRex_] 08:52:26 -!- NomadJim has quit [Read error: Connection reset by peer] 08:54:55 -!- Cryp71c has joined ##crawl-dev 08:58:12 -!- CKyle has joined ##crawl-dev 08:58:46 -!- tensorpudding has quit [Ping timeout: 268 seconds] 08:59:56 -!- Matejii has quit [Ping timeout: 252 seconds] 09:03:33 -!- User is now known as Xenobreeder 09:08:06 -!- st_ has joined ##crawl-dev 09:19:38 -!- rast- has joined ##crawl-dev 09:20:01 -!- rast has quit [Ping timeout: 240 seconds] 09:20:15 -!- rast- is now known as rast 09:26:17 -!- gammafunk has joined ##crawl-dev 09:26:18 Chris7 (L16 DDAr) ERROR: range check error (-1 / 80) (Shoals:5) 09:26:52 !crashlog Chris7 09:26:53 No milestones for Chris7 (crash). 09:27:27 !crashlog Chris7 09:27:28 1. Chris7, XL16 DDAr, T:39952 (milestone): http://dobrazupa.org/morgue/Chris7/crash-Chris7-20140404-142616.txt 09:30:17 old_target with a -1 x value 09:30:20 wonder how that happened 09:31:33 my guess is something related to open sea since it is in shoals and open sea tends to cause trouble 09:32:03 -!- NomadJim has quit [Read error: Connection reset by peer] 09:33:27 oh, negative value even in the dist object 09:33:59 -!- Zicher has quit [Quit: Page closed] 09:34:00 -!- Bloax has joined ##crawl-dev 09:34:19 that comment mentions the dist object 'will be initialized by direction()' 09:35:34 oh 09:35:40 well that is an easy crash to reproduce 09:35:44 -!- Staplegun is now known as sgun 09:35:48 just x beyond the edge of the level 09:35:50 yeah 09:35:52 and then v 09:35:53 was thinking that would be it 09:36:00 looks like that comment isn't correct 09:36:06 and the target property of the dist 09:36:08 is getting used 09:36:29 wow, weird things happen if you do that on a non-shoals level 09:36:37 e.g. the lmove variable at direction.cc:817 09:36:49 probably due to random initialization 09:36:52 it might be in-range 09:36:56 yet a totally weird location 09:36:58 would be my guess 09:37:02 so you get innacurate info? 09:37:15 if I xv beyond the level boundary on D:1, it doesn't crash 09:37:24 I just get a blank screen with a single ! in the upper left corner 09:37:37 -!- Tux[Qyou] has quit [Ping timeout: 268 seconds] 09:37:38 -!- reaverb has joined ##crawl-dev 09:37:42 that behaves like other descriptions 09:38:08 oh, I get that for places in-range too 09:38:13 that I can't see 09:38:16 yeah it's probably looking up some weird location 09:38:21 oh 09:38:27 this is with lang_sgrunt 09:38:36 haha 09:38:37 I should change that and see whether that is causing the ! 09:40:09 -!- Keanan has joined ##crawl-dev 09:40:13 -!- Keanan has left ##crawl-dev 09:41:43 yeah, I can see where the value is getting used 09:41:51 hrm 09:43:18 yeah, I think initializing that lmove variable's target 09:43:48 ah, I see actually, the key should be disabled 09:43:51 that's really the problem 09:44:06 or rather the large problem 09:45:01 okay, I wasn't actually outside the level boundary when I got the single ! description 09:45:17 (and in non-grunt-mode, it is a single . instead) 09:45:30 but that still seems like a poor description for a square that isn't in your map memory 09:45:47 but yes I guess there are two separate issues here, the crash and this description 09:46:31 I'm rebuilding master so it will take a bit for me to test 09:46:47 but you get that crash for xv on the open sea? 09:46:58 -!- blackcustard has quit [Quit: being stupid, sunstruck, and Dead, flew into the rocketing FIN.] 09:48:08 -!- Vidiny has quit [] 09:48:19 no, beyond the open sea 09:48:26 the open sea is fine 09:48:56 I also get it beyond the level boundary on other levels, but you have dig to get close enough to the level boundary to xv to it on normal levels 09:49:29 ah, yeah that's the problem then 09:49:39 it's allowing the command on past boundary 09:49:46 yeah I just got same crash 09:50:05 should just disallow the v command on past the map boundary 09:50:08 well, ideally we wouldn't leak the information of where the boundary is 09:50:26 and it would show up the same as squares that the player doesn't have in their map memory 09:50:32 -!- Xenobreeder has quit [Read error: Connection reset by peer] 09:51:37 -!- reaverb has quit [Ping timeout: 240 seconds] 09:52:05 yeah; the command should be disabled for anything not in map memory 09:52:08 -!- Rewans has quit [Quit: Page closed] 09:52:29 -!- reaverb has joined ##crawl-dev 09:52:48 hrm, I don't actually have time to fix this right now, but the bug is diagnosed at least 09:53:19 can you make a patch for this? otherwise we should maybe get a bug report for it 09:55:42 I'm looking at the code myself now, if I can't figure it out then I'll put it on mantis 09:57:39 why is summoning AI combined with glass still so terrible 09:57:39 Ragdoll: You have 1 message. Use !messages to read it. 09:57:48 debug_stethoscope 10:00:10 elliptic: probably a simple check in describe_target() that does an early return if the cell isn't in the viewport is enough 10:00:40 yeah, I just found describe_target() and will try adding stuff 10:01:29 viewport isn't what you mean I think 10:01:37 -!- Morg0th has quit [Quit: leaving] 10:01:44 -!- MaxFrost has quit [Remote host closed the connection] 10:01:50 it should just do nothing if the square isn't in map memory or is out of bounds 10:02:38 well I'm talking in terms of what method you want to use to test that; perhaps it's not crawl_view.in_viewport_g, no 10:03:21 -!- PleasingFungus has joined ##crawl-dev 10:03:38 -!- ussdefiant has joined ##crawl-dev 10:04:41 -!- gammafunk has quit [Quit: leaving] 10:05:30 hm, checking in_bounds isn't quite right because then you can't examine the open sea 10:06:42 I guess map_bounds 10:12:29 -!- rossi has quit [Quit: WeeChat 0.4.3] 10:14:18 -!- gnum has quit [Ping timeout: 255 seconds] 10:18:01 03elliptic02 07* 0.15-a0-56-gc5940bd: Fix issues with xv-ing unknown squares. 10(5 minutes ago, 1 file, 2+ 0-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=c5940bd0d397 10:18:02 03elliptic02 07[stone_soup-0.14] * 0.14-b1-22-ge8f5117: Fix issues with xv-ing unknown squares. 10(5 minutes ago, 1 file, 2+ 0-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=e8f511767850 10:18:38 huh, why did chei announce the cherry-pick 10:19:07 I thought chei just said "Cherry-picking one commit to stone_soup-0.14." or some such 10:19:10 -!- NomadJim has quit [Read error: Connection reset by peer] 10:19:51 and also I thought commits to branches were generally not announced on ##crawl too? 10:20:00 or am I just confused, I do have a bad memory 10:25:13 -!- notcluie has quit [Ping timeout: 245 seconds] 10:27:27 I recall cherrypicking too. 10:28:52 -!- iasov has quit [Quit: ZNC - http://znc.in] 10:30:31 <|amethyst> oh, hm 10:30:32 |amethyst: You have 2 messages. Use !messages to read them. 10:32:21 <|amethyst> Chei announces commits to trunk and stable branches to ##crawl 10:32:47 <|amethyst> as for the cherry-pick thing, it seems I disabled that for some reason but I cannot remember why 10:32:50 ah 10:32:52 <|amethyst> let me check the git log 10:33:31 -!- demiskeleton has quit [Quit: Leaving] 10:34:01 btw, http://crawl.develz.org/wordpress/0-14-tournament 10:34:55 -!- Cheibriados has joined ##crawl-dev 10:34:55 -!- The topic of ##crawl-dev is: Crawl Development | Logs: http://s-z.org/crawl-dev/ | People with +v have commit access. | Please keep general Crawl-related chit-chat to ##crawl. | Dev wiki: http://crawl.develz.org/wiki | Long stuff to pastebin.ca please | Immortal Warwalrus and Crazy Yiuf forever. 10:34:56 -!- The topic of ##crawl is: Check out new European server at ??clan! | Play Crawl online now: type ??cao, ??cdo, ??cszo, ??clan, or ??webtiles for instructions | http://crawl.develz.org | FooTV: termcast.develz.org | ##crawl-offtopic: You got your gender in my peanut butter! Xom thinks this is hilarious. 10:35:38 -!- DrKe has quit [Ping timeout: 240 seconds] 10:38:58 <|amethyst> !tell Grunt as for 0.13.2 on sf.net, what's your account? 10:38:59 |amethyst: OK, I'll let grunt know. 10:39:34 <|amethyst> elliptic: Anything I need to take care of on the servers for tournament prep? 10:40:14 <|amethyst> elliptic: I'll be out of town when the tournament starts; hopefully there will be some internet access, but at the conference last year there were networking problems :/ 10:40:20 |amethyst: I don't think so, everything is pretty well established now :) 10:40:38 I'll test the scripts on CSZO before then to make sure 10:40:52 |amethyst: smelenchuk 10:41:19 <|amethyst> Grunt: okay, you have admin access now 10:41:23 <|amethyst> I think 10:41:23 |amethyst: thanks! 10:41:27 * Grunt goes to check. 10:41:45 If there are any last things to put into 0.13 (does that xv thing crash in 0.13?) this would be a good time. 10:42:02 it might 10:42:17 I didn't check how old the problem is 10:42:34 (oh good, I still remember my sf password) 10:43:21 (currently building 0.13 to check) 10:43:29 Zicher? 10:43:32 <|amethyst> sorry I've been mostly incommunicado; I had a job interview on Monday, and we've been hosting NCUR (undergrad research conference) this week as well 10:43:58 Hi Nappy - there were several things I wanted to ask you, but now I can't really remember most of them :( 10:44:03 <|amethyst> BTW, this is checked on our project metadata page: 10:44:12 <|amethyst> This project incorporates, accesses, calls upon or otherwise uses encryption software with a symmetric key length greater than 64 bits ("encryption"). This review does not include products that use encryption for authentication only. 10:44:18 <|amethyst> Is that actually true? 10:44:29 o_O 10:44:35 I don't think so? 10:45:46 Napkin: the one I remember for certain related to the possibility of having another mantis admin or two or three (so that we can, say, update the Crawl versions and/or do other Mantis bits and maintenance). 10:46:18 I think I was also going to ask you about Debian packages and the state of trunk builds (I think you were working on at least one or the other). 10:46:23 ...and sorry if I'm flooding you with questions :( 10:46:59 definitely, the "Hi Nappy" sentence ran past me, not nick-hilighting 10:47:14 -!- Netmonmatt has quit [Ping timeout: 240 seconds] 10:47:19 (todo don't call you Nappy when you're actually here) 10:47:24 how about private messages? i'll have time tomorrow, i think 10:47:27 |amethyst: Maybe it's triggered by the RNG? bh said he designed it resist cryptoanalysis. 10:49:09 -!- DrinkMachine has quit [Quit: Leaving] 10:50:31 -!- _D_ has quit [Quit: leaving] 10:54:27 -!- Akien has joined ##crawl-dev 10:55:07 -!- radinms has quit [] 10:55:07 -!- NomadJim has quit [Read error: Connection reset by peer] 10:56:46 sigh 10:57:13 -!- ussdefiant has quit [Ping timeout: 240 seconds] 11:00:38 -!- notcluie has quit [Ping timeout: 245 seconds] 11:01:57 -!- rast- has joined ##crawl-dev 11:03:40 elliptic: good job with the new banners, though I object to xom's new color scheme :P 11:04:05 :( 11:04:12 Wensley: Grunt is responsible for the new banner images :) 11:04:38 -!- RZX has quit [Ping timeout: 240 seconds] 11:04:42 Grunt: did you use the factory?? 11:04:46 -!- randomizr is now known as RZX 11:04:49 -!- rast has quit [Ping timeout: 240 seconds] 11:04:51 Wensley: yes!! 11:04:54 hooraaaay 11:05:02 -!- rast- is now known as rast 11:05:05 (apparently I remember how to use Inkscape better than elliptic) 11:05:16 well done 11:05:33 (I should probably go and fix the kerning on the Heretic banner at some point) 11:05:35 politician is really good. like the color scheme, like the kerning 11:05:38 yeah I was gonna say 11:05:44 heretic is looking pretty spread out :P 11:05:52 (that was the first one I did btw <_<) 11:06:07 -!- alefury has joined ##crawl-dev 11:06:49 the rainbow banner :( 11:07:13 Stable (0.14) branch on crawl.lantea.net updated to: 0.14-b1-22-ge8f5117 11:08:14 -!- mizu_no_oto has quit [Quit: Computer has gone to sleep.] 11:08:18 -!- DrKe has quit [Ping timeout: 240 seconds] 11:09:27 are there any tournament rules concerning lua bots? 11:09:39 "don't do something stupid" 11:10:54 anything specific you had in mind? 11:11:53 -!- notcluie_ has quit [Ping timeout: 245 seconds] 11:11:55 Grunt: banner image updates accepted whenever of course 11:12:07 elliptic: How hard would it be to put xw, qw, and the other bots into a robot clan? 11:12:23 nooodl: 1) A lua bot may not injure a human being or, through inaction, allow a human being to come to harm. 11:12:24 reaverb: that was sort of the plan already 11:12:33 elliptic: Cool 11:12:41 nooodl: 2) A lua bot must obey the orders given to it by human beings, except where such orders would conflict with the First Law. 11:13:01 nooodl: 3) A lua bot may not injure its own kind and defend its own kind unless it is interfering with the first or second rule. 11:13:04 elliptic: i was worred about what if people get the Speed Demon banners "cashybrid style". but that falls under "stupid" 11:13:23 reaverb: last tourney xw and rw were in a clan (http://dobrazupa.org/tournament/0.13/clans/xw.html) 11:13:41 qw would have asked to join too but I didn't get qw running in tourney until after clans were frozen 11:14:22 elliptic: https://dl.dropboxusercontent.com/u/5494313/banner_lugonu1.png 11:14:32 -!- Lawman0 has quit [Ping timeout: 246 seconds] 11:14:45 elliptic: I like how xw just randomly got a sniper banner. 11:14:49 -!- morik has quit [Ping timeout: 240 seconds] 11:15:10 nooodl: yeah, I'm not that concerned about using lua bots to "cheat" at banners, because banners are supposed to just be fun accomplishments 11:15:13 -!- notcluie has quit [Ping timeout: 245 seconds] 11:16:09 nooodl: I'd be more concerned about people using lua bots to try to get the first win for extra tourney points on their main account and such 11:17:15 elliptic: for the record, the view thing does exist in 0.13, and your fix applies cleanly, so... 11:17:17 but the unassisted bots aren't good enough to do this with any high probability, and it is a little hard to draw a line between hybrid play and normal autofight stuff 11:18:10 Grunt: sounds like a reasonable inclusion then 11:18:50 Cherry-picked 1 commit into stone_soup-0.13 11:19:22 also, we can't really detect players doing unfair things like having multiple people playing on the same account to rack up a ton of wins anyway 11:20:02 reaverb: he was choosing combos specifically to try to get that banner :) 11:20:12 -!- Blazinghand has quit [Read error: Connection reset by peer] 11:20:22 elliptic He 11:20:38 s/He/Ha 11:23:08 -!- notcluie_ has quit [Ping timeout: 245 seconds] 11:23:23 -!- ainsophyao has joined ##crawl-dev 11:23:55 Grunt: images updated 11:24:13 -!- ainsophyao has quit [Remote host closed the connection] 11:25:39 -!- Foamed has quit [Read error: Connection reset by peer] 11:25:59 -!- lobf has joined ##crawl-dev 11:29:05 03Grunt02 07[stone_soup-0.13] * 0.13.1-39-g42362f2: The final 0.13.2 changelog update. 10(7 minutes ago, 1 file, 2+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=42362f24509b 11:29:05 Cherry-picked 1 commit into stone_soup-0.14 11:31:55 -!- bmfx has quit [Quit: leaving] 11:32:44 famous last words? 11:32:50 -!- Kaput has quit [Quit: leaving] 11:33:59 %git stone_soup-0.14 11:33:59 07Grunt02 * 0.14-b1-23-gcb5fb12: The final 0.13.2 changelog update. 10(12 minutes ago, 1 file, 2+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=cb5fb12736ae 11:34:13 Cherry-picked 1 commit into master 11:34:15 %git stone_soup-0.13 11:34:15 07Grunt02 * 0.13.1-39-g42362f2: The final 0.13.2 changelog update. 10(12 minutes ago, 1 file, 2+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=42362f24509b 11:34:15 ah, right, want that in 0.14 11:34:28 -!- Kaput has quit [Client Quit] 11:34:34 ...I forgot to push the tag, didn't I <_< 11:35:17 %git stone_soup-0.13 11:35:17 07Grunt02 * 0.13.2: The final 0.13.2 changelog update. 10(13 minutes ago, 1 file, 2+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=42362f24509b 11:35:21 ...much better. 11:35:23 * Grunt restarts the build. 11:36:58 -!- HellTiger_NB has quit [Ping timeout: 240 seconds] 11:38:33 -!- Garhauk has quit [Ping timeout: 245 seconds] 11:42:16 -!- PleasingFungus has quit [Remote host closed the connection] 11:42:43 -!- PleasingFungus has joined ##crawl-dev 11:46:03 -!- notcluie has quit [Ping timeout: 245 seconds] 11:50:48 -!- minmay has joined ##crawl-dev 11:51:56 -!- Cedor has joined ##crawl-dev 11:52:13 -!- CKyle has quit [Quit: CKyle] 11:52:38 -!- HellTiger_NB has quit [Ping timeout: 240 seconds] 11:54:55 -!- notclui has quit [Quit: Page closed] 11:55:46 Why is it that if you hit Shift + Backspace on the name character screen it produces a y with an Umlaut? 11:57:56 If you have that as your name close crawl and reopen it changes to O with an accent and degree symbol. 11:58:04 Is this a Glitch? 11:59:43 -!- wheals has quit [Quit: Leaving] 11:59:57 -!- rast- has joined ##crawl-dev 12:00:17 -!- CKyle has joined ##crawl-dev 12:02:25 -!- rast has quit [Ping timeout: 240 seconds] 12:02:32 -!- rast- is now known as rast 12:03:11 -!- NomadJim has quit [Read error: Connection reset by peer] 12:04:25 Stable (0.14) branch on crawl.akrasiac.org updated to: 0.14-b1-23-gcb5fb12 12:07:10 -!- CKyle has quit [Quit: CKyle] 12:07:16 Stable (0.13) branch on crawl.akrasiac.org updated to: 0.13.2-0-g42362f2 12:07:33 -!- Bcadren has quit [Quit: Depression is merely anger without enthusiasm] 12:11:03 Unstable branch on crawl.akrasiac.org updated to: 0.15-a0-57-g3f1ec01 (34) 12:11:42 -!- Pepe has quit [Ping timeout: 252 seconds] 12:13:23 -!- dpeg_ has quit [Quit: Lost terminal] 12:14:02 wheee, locale fun 12:17:53 -!- LexAckson_ has joined ##crawl-dev 12:18:30 -!- mizu_no_oto has quit [Ping timeout: 255 seconds] 12:25:27 -!- zxc232 has quit [Quit: Leaving.] 12:26:47 -!- dpeg has joined ##crawl-dev 12:27:32 -!- Kalir has quit [Changing host] 12:28:07 -!- dtsund has joined ##crawl-dev 12:30:01 -!- zxc232 has quit [Ping timeout: 240 seconds] 12:30:03 So we have a 0.13.2 now. \o/ 12:31:58 -!- rossi has quit [Quit: WeeChat 0.4.3] 12:32:43 -!- raskol has quit [Quit: quit] 12:33:51 yay! 12:33:51 dpeg: You have 2 messages. Use !messages to read them. 12:46:53 -!- HDA has quit [Ping timeout: 245 seconds] 12:48:49 Crawl Light has the following neat effect: potions you drink affect nearby allies. Also smells of feature creep but it much, *much* less contentious than spells. 12:49:03 Only for Beogh followers in good standing and piety, and only for orcish allies 12:49:33 dtsund: thanks for correction 12:49:50 we agree your solution is much better than spell sharing, right? 12:51:06 Haven't seen the original proposal yet 12:51:18 Forum, or dev wiki? 12:53:11 -!- NomadJim has quit [Read error: Connection reset by peer] 12:54:08 |amethyst: one thing I noticed while updating the downloads page is this link: http://crawl.develz.org/main/0.13.1.txt 12:54:24 |amethyst: do you know if there's a way to get a 0.13.2.txt there? 12:54:36 -!- Hailey has quit [Ping timeout: 252 seconds] 12:54:40 -!- simmarine has joined ##crawl-dev 12:55:42 dtsund: forum, Beogh proposal 12:57:13 Ah, it's in Yiuf's corner, that's why I couldn't find it 12:57:57 "3 was a little ridiculous for something weaker than Lesser Destruction" 12:58:09 Dude, those invocations don't exist in a vacuum 12:58:21 Lesser Destruction needs to be spammable because that's what Makhleb does 12:59:22 -!- KurzedMetal has quit [Read error: Connection reset by peer] 12:59:36 -!- wheals has joined ##crawl-dev 13:00:17 Yeah, boost sharing is just bad. 13:01:08 Even putting aside balance concerns, the interface and discoverability issues alone would make it a bad idea. 13:01:29 dtsund: yes, of course 13:02:03 ...and then I get to Messiah. Good grief, a horde of followers spamming Fire Storm? 13:02:29 :) 13:02:30 ...and then I get to Messiah. 13:02:34 you get to https://www.youtube.com/watch?v=fXWNsNd0zBE what 13:03:20 The rationale behind potion sharing was to make it less likely that a single bad encounter could utterly destroy your entire army 13:04:38 It might be feasible to tone down reinforcements with that in place, since you shouldn't need them as much with proper play 13:05:33 true 13:05:42 Can I assign a macro to shift+1 by any chance? 13:05:44 oops 13:05:46 wrong window 13:06:05 Well if there's a recall ability. 13:06:09 Why isn't there a dismiss one? 13:06:38 I think that actually got proposed at some point. 13:06:53 Bloax: because it's not a good idea 13:07:04 even if you don't want all followers around you, collecting them is good 13:08:05 -!- Garhauk has quit [Quit: Page closed] 13:08:18 -!- Flun has quit [Quit: Page closed] 13:12:21 -!- CKyle has joined ##crawl-dev 13:13:21 -!- simmarine has quit [Read error: Connection reset by peer] 13:13:33 -!- Kellhus has quit [Ping timeout: 245 seconds] 13:14:18 Why does Crawl use "const foo_type &bar" in function arguments so much? 13:14:56 -!- simmarine has joined ##crawl-dev 13:15:48 -!- LexAckson_ has quit [Quit: Page closed] 13:17:26 because things are mutable by default in C++, and you need to tell the compiler that you aren't going to change the thing by marking it with 'const' 13:18:09 Zaba: Yes, I mean as oppose to "foo_type bar" 13:18:44 because that makes a copy 13:18:45 -!- LexAckson has joined ##crawl-dev 13:19:01 & means it's a reference, so it's not copied 13:19:43 Yes, but const means that you can't change it anyway, so there's no point to the copy. 13:20:03 no, it does -not- make a copy 13:20:11 it'd make a copy if it were 'foo_type bar' 13:22:04 Zaba: "foo_type bar" does make a copy. "const foo_type &bar" does not. However, the "const" means you can't modify the referance anyway, so there's no effective differance between a referance or a copy. 13:22:17 performance, stack usage 13:22:22 there is, memory usage 13:22:37 especially if it's a big structure 13:23:08 -!- tabstorm has quit [Ping timeout: 245 seconds] 13:23:24 geekosaur: I'm very confused by people concerned about stack or memory usage on a program as lightweight as Crawl. 13:23:38 lightweight!? 13:23:45 Lightweight?? 13:23:50 well, you can just consider it a good coding practice 13:24:03 (and it's a shame that it's so syntactically cumbersome, but such is C++) 13:24:39 reaverb, I maintain a private build of crawl for OS X because the standard build disables optimization due to limitations of Apple's default compiler suite. The result of that disabled optimization is PAINFUL 13:25:01 geekosaur: I compile myself and I've never noticed Crawl lagging, ever. 13:25:16 On Mac OSX 13:25:19 yes, it's not lagging, thanks to the const & everywhere :P 13:25:38 it does lag some. I do admit this machine is fairly busy all the time so pauses are more noticeable 13:26:38 Lightweight?? 13:26:44 seriously a new level in a build based on apple gcc 4.2 with no optimization often takes close to a minute to load 13:27:15 it'd be far worse if crawl copied large structs instead of using const references 13:27:29 * Grunt reads a scroll of remove curse. Grunt's keyboard glows for a moment. 13:27:52 in any case this is pretty basic C++ best practice 13:28:08 heck, it's a best practice I know and I am emphatically not a C++ programmer 13:28:27 * reaverb wonders if Crawl will be moved to another language eventually. 13:28:36 ??fcrawl 13:28:36 fcrawl[1/5]: http://bpaste.net/raw/0mfJJgUDZ89kDH6gMIT0/ 13:28:55 Lightweight?? 13:28:58 -!- notcluie has quit [Ping timeout: 245 seconds] 13:29:30 heh 13:29:32 Lightweight?? 13:35:45 -!- nooodl_ has joined ##crawl-dev 13:37:15 -!- nooodl has quit [Disconnected by services] 13:37:17 -!- nooodl_ is now known as nooodl 13:37:59 -!- nooodl is now known as nooodl_ 13:38:05 -!- nooodl_ is now known as nooodl 13:38:29 -!- LordSloth has quit [Remote host closed the connection] 13:39:08 -!- SurpriseTRex has quit [Ping timeout: 240 seconds] 13:43:54 -!- wheals has quit [Quit: Leaving] 13:46:53 -!- DrinkMachine has joined ##crawl-dev 13:48:15 fcrawl ♥ 13:51:58 Maybe I should land multisprint early in 0.15 and work on μcrawl properly. 13:53:59 Grunt: you mean have crawl become even more bloated, following in the footsteps of μtorrent? 13:54:10 *rimshot* 13:54:29 caught in the act 13:54:38 μcrawl is my hypothetical sprint with the same dungeon structure as the main game only every branch is one level. 13:55:01 by the way 13:55:13 where's AnActualZiggurat-sprint :q 13:55:24 There's one in the multisprint code! 13:55:30 (I'll probably need to modernise it) 13:56:31 -!- xFleury has joined ##crawl-dev 13:57:29 fr: rechargeable evocable that has a scroll of fog effect 13:57:51 We have a fog evoker. 13:57:54 ??cloak of the thief 13:57:54 cloak of the thief[1/1]: the +2 cloak of the Thief {+Fog Dam-3 SInv Stlth++}. The fog costs 2MP and is as difficult as berserk rage to evoke successfully. 13:58:51 yes, it was amazing, best item in-game for extended IMO, but I imagine my chances of ever finding it again are quite low 13:59:23 Grunt: the punchline here is to have it function like the elemental evocables with the effect of +fog 13:59:40 so like a multi-use scroll of fog 13:59:47 that you can't spam 13:59:56 -!- rast has quit [Read error: Connection reset by peer] 13:59:57 (unlike the thiefcloak) 14:00:09 -!- rast has joined ##crawl-dev 14:01:26 -!- wheals has joined ##crawl-dev 14:17:44 -!- Mad_Wack has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…] 14:21:37 -!- mizu_no_oto has quit [Quit: ["Textual IRC Client: www.textualapp.com"]] 14:21:54 -!- Ketsa has quit [Quit: WeeChat 0.3.8] 14:22:38 -!- Nivim has quit [Ping timeout: 240 seconds] 14:22:49 -!- Nivim has joined ##crawl-dev 14:23:22 -!- ackack has joined ##crawl-dev 14:26:28 -!- nooodl is now known as fcrawl 14:28:40 -!- LexAckson_ has joined ##crawl-dev 14:28:56 -!- LexAckson has quit [Quit: Page closed] 14:30:32 -!- Nivim has quit [Ping timeout: 268 seconds] 14:32:50 -!- LexAckson_ is now known as LexAckson 14:33:07 -!- ldf2866 has joined ##crawl-dev 14:33:33 -!- Rolland has quit [Ping timeout: 245 seconds] 14:34:04 -!- ldf has quit [Quit: ldf] 14:34:19 -!- ldf2866 is now known as ldf 14:39:18 I'm tempted to code it, and submit it as a patch, if I'm not the only one who thinks it's a fantastic idea. 14:44:13 -!- dpeg has quit [Quit: Lost terminal] 14:47:09 -!- wheals has quit [Quit: Leaving] 14:47:43 -!- notcluie has quit [Ping timeout: 245 seconds] 14:49:21 why not just a wand of fog? 14:49:32 -!- dtsund has quit [Quit: dtsund] 14:49:51 oh nm 14:49:55 the spam issue 14:50:31 Fogstone 14:50:47 smoke crystal 14:50:57 smokestone 14:50:59 xFleury: Rewrite the ability code to be more sane and include the fog evocable as part of the patch. 14:51:10 i dont like consumables that do the same thing as other consumables 14:51:22 wand of healing? 14:51:23 also cloak of the thief really shouldnt exist 14:51:26 wand of healing is bad 14:51:26 That's how the macbre finger necklase got in.t :) 14:51:48 healing/haste/tele all should not both be a wand and a different consumable imo 14:51:57 and invis i guess 14:52:03 ah ok. I hadn't thought about it much 14:52:08 what is the main rationale? 14:52:12 -!- Nivim has joined ##crawl-dev 14:52:39 i dont see what the point is of having two things that do the same thing ... in the case of /hw and /haste specifically they're also bad because of how much you get 14:52:42 just the general crawl design direction of not duplicating any effect or feature? 14:53:06 (one single /haste is doing something like quadrupling the amount of consumable haste you have in that game, with a pessimistic count of ?recharging) 14:53:37 crate: I don't really equate "evocable fog v scroll of fog" as the same issue as "wand of healing v potion of healing". 14:53:39 but anyway repeatable fog is not a thing that should exist, look at GIFTDROP doing tomb for why i think that 14:53:51 i think fog works nicely as a scroll but should not be more common 14:54:02 keep it as pale drac gimmick 14:54:25 -!- dtsund has joined ##crawl-dev 14:54:44 hm. would moving the heal/haste wands to a rechargeable have any effect on the overall opinion? 14:54:55 or is that worse since it's effectively infinite 14:54:58 just slower 14:55:44 i'm not really sure how much i like the xp-recharging thing that evokers do anyway but i dont have good reasons for that 14:55:52 that wouldn't help, it would a "no-brainer" to use the rechargeable whenever you want to heal/haste and it is charged 14:56:04 xp-recharging just doesn't work for things that exactly duplicate a potion 14:56:06 hmm 14:56:23 it would suck because when you use wands of healing you usually use them a lot 14:56:44 if crawl did not have any wands, what are the chances that /hw or /haste or /tele would get added? 14:56:49 i would think pretty low 14:56:59 personally I agree with crate that duplicating a potion with a wand isn't great, however I'm not sure what the best solution is without messing up balance 14:56:59 yea 14:56:59 could buff how much it heals. but I'm starting to see better what the potential issues are 14:56:59 interesting! 14:57:31 poor DD 14:57:34 anyway repeatable fog is bad, and i think cloak of the thief illustrates why it is bad 14:57:43 since it is ridiculously abusive 14:57:57 cloak of the thief is overrated IMO 14:58:00 eh? it would make evocations skill not "useless" and a "waste of skill" in extended 14:58:11 good thing it's not useless and a waste of skill then 14:58:18 yes 14:58:42 if anything evocations is maybe too strong right now honestly but i'm not going to debate that 14:59:05 maybe that is justified since there is difficulty in actually finding evocables 14:59:10 quoted because those are indeed quotes from some players who I don't think are generally considered ones giving bad advice 14:59:33 "quotes from some players" without even saying who they are isn't very persuasive here, fyi 14:59:47 I *think* one is from minmay, but don't quote me on that 15:00:15 i had several evocations-focused wins without nemelex before elemental evokers got added or rods got reworked, and evo is unquestionably stronger now than it was then 15:02:18 anyway having an xp-charged evocable that duplicates the effect of scroll of fog is not something that will happen, for the same reasons I said earlier about why xp-charged heal/haste is bad - xp-charged evocables need to produce unique effects 15:09:19 -!- Hirsch_ has quit [Quit: Page closed] 15:10:23 -!- Amnesiac has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/] 15:11:53 -!- yalue has quit [Quit: Leaving] 15:13:40 -!- Lasty has quit [Read error: Connection reset by peer] 15:14:44 -!- wheals has joined ##crawl-dev 15:15:03 -!- Lasty has joined ##crawl-dev 15:17:43 -!- notcluie has quit [Ping timeout: 245 seconds] 15:19:32 -!- bonghitz has quit [Quit: Leaving...] 15:22:32 -!- rossi has quit [Ping timeout: 252 seconds] 15:23:53 -!- NomadJim has quit [Read error: Connection reset by peer] 15:26:25 -!- wheals has quit [Quit: Leaving] 15:39:38 -!- Ladykiller69 has quit [Ping timeout: 240 seconds] 15:40:33 -!- mizu_no_oto has quit [Ping timeout: 255 seconds] 15:40:40 -!- UncertainKitten has joined ##crawl-dev 15:41:20 -!- Tedronai has quit [Quit: leaving] 15:42:35 -!- LexAckson has quit [Quit: Leaving] 15:47:00 -!- LexAckson has joined ##crawl-dev 15:48:16 -!- caleba has quit [Quit: caleba] 15:50:14 -!- PleasingFungus has quit [Quit: ChatZilla 0.9.90.1 [Firefox 31.0a1/20140403030201]] 15:51:16 -!- wheals has joined ##crawl-dev 15:51:45 -!- raskol has quit [Quit: quit] 15:52:11 -!- caleba has quit [Client Quit] 15:55:38 -!- notcluie_ has quit [Ping timeout: 245 seconds] 15:59:16 -!- LexAckson has quit [Read error: Connection reset by peer] 15:59:48 -!- mk83 has quit [Ping timeout: 245 seconds] 16:00:07 -!- rast has quit [Read error: Connection reset by peer] 16:00:21 -!- rast has joined ##crawl-dev 16:02:02 -!- Cedor has quit [Ping timeout: 246 seconds] 16:03:51 -!- NomadJim has quit [Read error: Connection reset by peer] 16:04:29 -!- Rewans has quit [Quit: Page closed] 16:06:13 -!- Krakhan has quit [Changing host] 16:09:57 -!- Patashu has joined ##crawl-dev 16:10:04 -!- brainwrinkle has quit [Quit: Leaving] 16:10:13 -!- notcluie has quit [Ping timeout: 245 seconds] 16:12:13 -!- gnum has quit [Read error: Connection reset by peer] 16:12:56 -!- Ragdoll has quit [Read error: Connection reset by peer] 16:17:00 -!- Mad_Wack has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…] 16:17:58 -!- ackack has quit [Ping timeout: 240 seconds] 16:18:10 -!- Arrhythmia has joined ##crawl-dev 16:18:11 -!- dpeg has joined ##crawl-dev 16:18:59 -!- Arrhythmia has quit [Client Quit] 16:26:32 -!- Lasty1 has joined ##crawl-dev 16:28:08 -!- raskol_ is now known as raskol 16:30:56 -!- Lumpydoo has quit [Quit: Page closed] 16:31:25 -!- Aryth has quit [Quit: KVIrc 4.0.4 Insomnia http://www.kvirc.net/] 16:32:37 <|amethyst> !tell Bcadren re shift-backspace: that sends the DEL character, code point 255, which in Windows code page 1252 is ÿ 16:32:38 |amethyst: OK, I'll let bcadren know. 16:34:49 <|amethyst> !tell Bcadren err, wait, del is 127... 16:34:49 |amethyst: OK, I'll let bcadren know. 16:36:33 <|amethyst> !tell Bcadren as for the mojibake, that sounds like your terminal not working in UTF-8 mode or something 16:36:34 |amethyst: OK, I'll let bcadren know. 16:36:54 -!- ffffff has quit [Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com )] 16:37:24 -!- PleasingFungus has joined ##crawl-dev 16:38:56 -!- xFleury has left ##crawl-dev 16:39:08 exit and restart triggering it makes me wonder if the file holding the last combo is not in utf8 for some reason 16:39:34 written in one encoding, read back in another 16:39:40 or something like that 16:41:56 <|amethyst> hm, FileLineInput (used for reading the startup prefs) supports different encodings based on the presence of a BOM 16:42:11 (IMO defuse the BOM) 16:42:16 %git HEAD^{/BOM} 16:42:55 07ontoclasm02 * 0.14-a0-2431-g1d11b85: Drop the BOM 10(9 weeks ago, 1 file, 1+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=1d11b85fce16 16:42:55 <|amethyst> the thing is, when writing it back out there shouldn't be a BOM 16:43:20 <|amethyst> %git 47f8aa00 16:43:20 07|amethyst02 * 0.12-a0-1657-g47f8aa0: Use logical punctuation quoting. 10(1 year, 3 months ago, 1 file, 2+ 2-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=47f8aa00355a 16:43:26 <|amethyst> (that one's further down) 16:43:36 <|amethyst> %git b6a86b 16:43:37 07|amethyst02 * 0.13-a0-2979-gb6a86b3: What do you mean I can't say "BOM" on an airplane? 10(7 months ago, 1 file, 1+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=b6a86b3bbb58 16:43:59 -!- NomadJim has quit [Read error: Connection reset by peer] 16:44:22 -!- Patashu[Zzz] has joined ##crawl-dev 16:44:22 -!- Patashu has quit [Disconnected by services] 16:47:49 -!- caleba has quit [Quit: caleba] 16:55:26 -!- mizu_no_oto has quit [Quit: Computer has gone to sleep.] 16:58:08 How do I compile a "DEBUG_DIAGNOSTICS" build? (for -mapstat) 16:58:28 make debug i think does all of the DEBUG_* 16:58:43 -!- Chris7 has quit [] 17:00:00 03elliptic02 07* 0.15-a0-58-gfb3cfe9: Improve display of Nemelex wrath on Ctrl-O. 10(5 minutes ago, 1 file, 7+ 2-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=fb3cfe924d6a 17:00:00 Cherry-picked 1 commit into stone_soup-0.14 17:00:55 -!- NomadJim has quit [Read error: Connection reset by peer] 17:01:28 wheals: It's compiling, can't tell yet if it will work. 17:02:07 <|amethyst> reaverb: check your .cflags file 17:02:20 |amethyst: I don't know what a .cflags file is. 17:02:45 <|amethyst> reaverb: when you compile crawl it creates a file called .cflags to keep track of the flags being passed to the compiler 17:02:48 Stable (0.14) branch on crawl.s-z.org updated to: 0.14-b1-24-g875c316 17:03:20 <|amethyst> reaverb: if you see -DFULLDEBUG there, it it being built with full diagnostics 17:03:40 |amethyst: Thank you, it is. 17:04:05 -!- ystael has quit [Ping timeout: 268 seconds] 17:04:24 Unstable branch on crawl.lantea.net updated to: 0.15-a0-58-gfb3cfe9 (34) 17:04:26 Also, would anything break if crawl-ref/git-hooks was renamed someting like crawl-ref/crawl-git-hooks? I spent around hour troubleshooting a while back because I mixed it up with .git/hooks 17:05:23 -!- CKyle has quit [Quit: CKyle] 17:07:12 -!- Lumpydoo has quit [Quit: Page closed] 17:07:25 <|amethyst> it might break people's existing configs if they had their .git/hooks/ stuff calling the ones in crawl-ref/git-hooks (as opposed to just having copied the files into .git/hooks) 17:07:46 <|amethyst> but other than that it wouldn't be a problem 17:08:23 |amethyst: Do you think it's worth breaking that? I know calling it through that path is what I did to make my hooks, but it would be easy to update it. 17:14:56 -!- Garhauk has quit [Quit: Page closed] 17:15:13 -!- notcluie has quit [Ping timeout: 245 seconds] 17:15:39 -!- debo has quit [Quit: debo] 17:15:53 I'm getting a crash whenever I try to run ./crawl mapstat 17:16:13 (From master) 17:16:40 Can somebody quickly check if this is on all builds? 17:16:49 ASSERT(m_layers[LAYER_TILE_CONTROL].m_regions.size() == 1) in 'tilesdl.cc' at line 228 failed. 17:17:07 <|amethyst> reaverb: re the proposed change, I'm kind of conservative about things like that, but other devs may have a different opinion 17:17:12 <|amethyst> hm 17:17:25 <|amethyst> well, that wouldn't be on a console or webtiles build for one thing :) 17:17:53 |amethyst: Yes, that's fine. I just wish the git-hooks thing was differant from the beginning. 17:17:57 <|amethyst> trying a fulldebug tiles build, but my computer is slow and none of this is ccached, so it will take a while 17:18:18 |amethyst: Yeah, I'm compiling console right now. I'll try fulldebug once it finishes. 17:18:36 <|amethyst> by fulldebug, I mean the equivalent of make debug 17:19:01 <|amethyst> except I use make FULLDEBUG=y instead 17:19:12 <|amethyst> because I'm silly 17:19:29 |amethyst: Yes, that crash was on a make debug build. 17:22:26 -!- rossi has quit [Ping timeout: 252 seconds] 17:27:54 |amethyst: Does -mapstat normally look like it's completely killing your CPU with no visibile effect for a while? Because that's what console -mapstat is doing to my computer. 17:29:24 <|amethyst> I believe so; it's been a while since I've used it 17:29:33 <|amethyst> and, yes, I get the same crash with -mapstat in tiles 17:30:04 |amethyst: Hmm, both those statements are interesting. 17:30:21 <|amethyst> * If at some point it's possible to have multiple regions 17:30:21 <|amethyst> * open while the title screen shows, the .at(0) will need 17:30:21 <|amethyst> * to be changed and saved on a variable somewhere instead 17:30:55 <|amethyst> comment to TilesFramework::update_title_msg 17:31:10 <|amethyst> I don't know tiles code at all, though... 17:32:13 -!- bonghitz has quit [Remote host closed the connection] 17:32:19 |amethyst: I don't know how pratical this is, but you might want to try the console -mapstat too. I just got a "Virtural timer expired" error for ./crawl -mapstat 1 17:34:20 o.O 17:35:03 -!- CKyle has joined ##crawl-dev 17:37:04 <|amethyst> hrm... that should only happen if WATCHDOG is turned on 17:37:04 -!- Nethris has quit [Quit: Leaving.] 17:37:28 |amethyst: Which happens for fulldebug builds? 17:37:37 <|amethyst> and WATCHDOG is only turned on by default if DGAMELAUNCH is turned on 17:38:08 -!- caleba has quit [Quit: caleba] 17:38:28 <|amethyst> which should only happen if you build with USE_DGAMELAUNCH=y 17:38:29 <|amethyst> hm 17:42:07 -!- Nightbeer has quit [Read error: Connection reset by peer] 17:42:47 -!- BirdoPrey has quit [Remote host closed the connection] 17:42:55 I don't think mapstat has ever worked properly with tiles? 17:44:08 Grunt: That would make sense, but in that case the bug is that it crashes mysteriously instead of giving an error message. : ) 17:50:17 -!- st_ has quit [Read error: Connection reset by peer] 17:50:44 -!- st_ has joined ##crawl-dev 17:52:09 -!- Lightli has joined ##crawl-dev 17:53:33 -!- Reign_ has quit [Ping timeout: 245 seconds] 17:56:44 -!- lessens_ is now known as lessens 17:57:02 -!- lobf has quit [Quit: lobf] 17:59:13 -!- lobf has joined ##crawl-dev 18:00:18 -!- rast has quit [Read error: Connection reset by peer] 18:00:35 -!- bonghitz has quit [Remote host closed the connection] 18:00:36 -!- rast has joined ##crawl-dev 18:03:46 -!- rossi has quit [Quit: WeeChat 0.4.3] 18:03:46 Should I mantis this issue or wait for somebody to do another test? 18:06:35 <|amethyst> I'm not getting any watchdog trigger 18:06:53 <|amethyst> I have gotten two "Monster orc sorcerer in stone_wall at (18, 51)" 18:07:07 |amethyst: Huh, is the -mapstat command working for you? 18:07:39 <|amethyst> I forget how mapstat is supposed to work 18:07:43 <|amethyst> it's still running 18:08:02 |amethyst: It eventually writes a file with all the generated vaults. 18:08:33 |amethyst: Also, I only got the Virtual timer error when -mapstat ended. 18:08:50 -!- debo has joined ##crawl-dev 18:09:06 -!- Vidiny has quit [] 18:09:08 <|amethyst> oh, that could be the "attaching gdb took too long" thing 18:09:40 <|amethyst> mapstat has been running for 35 minutes here (26 minutes of CPU time) 18:10:33 |amethyst: Ouch, did you run "-mapstat 1" or normal -mapstat? 18:10:41 <|amethyst> just -mapstat 18:10:46 |amethyst: Also, the crash dump had this line: "Debug build -> run gdb yourself." 18:10:54 <|amethyst> oh, right 18:10:54 <|amethyst> hm 18:11:22 -!- ackack has joined ##crawl-dev 18:11:38 <|amethyst> my debug build is kind of extra slow 18:12:08 -!- bonghitz has quit [Ping timeout: 240 seconds] 18:12:10 -!- Kellhus has quit [Quit: Page closed] 18:12:52 <|amethyst> make -j3 EXTERNAL_FLAGS="-pg -finstrument-functions --coverage" EXTERNAL_LDFLAGS="-pg -finstrument-functions --coverage" DEBUG=y NO_OPTIMIZE=y USE_PCRE=y FULLDEBUG=y 18:13:01 <|amethyst> I'm sure the profiling stuff doesn't help 18:13:35 -!- Watball has quit [Quit: Leaving] 18:14:12 -!- NomadJim has quit [Read error: Connection reset by peer] 18:14:14 NO_OPTIMIZE doesn't help either 18:17:13 <|amethyst> make debug turns on NO_OPTIMIZE anyway 18:17:30 -!- lobf has quit [Quit: lobf] 18:17:43 <|amethyst> since -Og is fairly new 18:17:48 -!- mizu_no_oto has quit [Quit: Computer has gone to sleep.] 18:18:01 An ogre build comes into view. It is wielding a giant club. 18:18:39 -!- wheals has quit [Quit: Leaving] 18:19:51 |amethyst: made clean, recompiled, got the same Virtual timer error. 18:20:53 <|amethyst> doing just plain make debug ? 18:21:25 |amethyst: Also APPLE_GCC=y 18:21:33 that's necessary to compile on macs, right? 18:21:45 Either way, here's my cflags file: 18:21:45 http://bpaste.net/raw/uR78Figq0T1MO1CT24sn/ 18:22:57 <|amethyst> oh, hey 18:23:08 -!- nooodl_ has joined ##crawl-dev 18:23:09 <|amethyst> mg_do_build_level calls watchdog() even if WATCHDOG is turned off 18:23:29 -!- fcrawl has quit [Ping timeout: 252 seconds] 18:24:09 -!- nooodl_ is now known as fcrawl 18:24:36 <|amethyst> so I guess that means it took more than 60 seconds to generate a single level (since niters there is always 1) 18:25:36 <|amethyst> are you testing some of your own maps? 18:25:51 |amethyst: No, just trunk. 18:26:08 <|amethyst> hm 18:26:35 |amethyst: (Of course I was doing something else originally, but I switched to Trunk once I got the error) 18:27:25 reaverb: and "git diff" doesn't show anything you weren't expecting to see? 18:27:51 SamB: Nope, I'm no just master. 18:27:55 s/no/on/ 18:28:20 <|amethyst> try removing the des cache (saves/des/) just to be sure 18:28:21 -!- Kalir has quit [Quit: CADENZA-CLASS CLOCKWORK KNIGHT--OPERATION TERMINATED] 18:28:22 reaverb: so you actually ran it, right? 18:28:40 SamB: Yes, I actually recompiled and ran -mapstat 1, I even made clean. 18:28:48 I'm going to remove the des cache though. 18:28:52 by "it" I meant "git diff" 18:29:03 SamB: Yes, I ran git diff, got nothing. 18:29:11 good 18:29:18 -!- scummos| has quit [Ping timeout: 255 seconds] 18:31:02 <|amethyst> okay, with -mapstat 1 I get the same thing 18:33:00 <|amethyst> the signal comes in during _write_mapgen_stats 18:35:01 <|amethyst> which makes sense I guess, because it appears nothing pets the watchdog after map generation has finished 18:35:33 |amethyst: After regenerating the .des files, I got the responce of "Alarm clock" 18:35:45 I have no idea what this means, it could even be a success. 18:36:02 <|amethyst> that's essentially the same thing 18:36:31 <|amethyst> you'd get the virtual timer thing if UNIX is defined, otherwise "alarm clock" 18:37:05 |amethyst: Wait, I didn't recompile. I just deleted the .des files. 18:37:16 <|amethyst> hm 18:37:27 <|amethyst> run plain crawl to regen the des cache 18:37:34 <|amethyst> but anyway 18:37:38 |amethyst: I did that first. 18:37:41 <|amethyst> hmm 18:37:55 Before running the -mapstat 1 which got the "Alarm Clock" error. 18:38:09 This is bugs within bugs within bugs. 18:39:21 -!- dtsund has quit [Quit: dtsund] 18:39:43 <|amethyst> then I guess that one is the "crash dump took too long" alarm 18:40:10 <|amethyst> which appears to use alarm(120) regardless of UNIX 18:40:46 -!- xFleury has quit [Quit: ChatZilla 0.9.90.1 [Firefox 28.0/20140314220517]] 18:41:08 |amethyst: It is true the Alarm Clock error did not give a crash dump as the watchdog ones did. 18:41:25 |amethyst: So what's the solution? Make the watchdog() call require WATCHDOG to be defined? 18:41:32 -!- tkappleton1 has quit [Quit: Leaving] 18:41:40 <|amethyst> testing something 18:41:47 (in mg_do_build_level ) 18:41:56 |amethyst: Sure. 18:41:59 <|amethyst> I'm trying a call to watchdog() inside the loop in _mapgen_report_available_random_vaults 18:42:18 -!- notcluie has quit [Ping timeout: 245 seconds] 18:42:54 <|amethyst> we do want the watchdog during map-stats, since otherwise there's no way to tell the difference between "it's hanging when trying to generate this map" and "it's generating lots of maps" 18:43:15 |amethyst: That makes sense. 18:50:24 -!- Tux[Qyou] has quit [Quit: Excess flood] 18:58:20 -!- lobf has joined ##crawl-dev 18:58:33 -!- notcluie has quit [Ping timeout: 245 seconds] 18:59:21 |amethyst: Did adding the watchdog() to _mapgen_report_available_random_vaults work? 18:59:51 <|amethyst> reaverb: it's gotten much further, but I want to wait for it to finish before I say yes 18:59:56 <|amethyst> it looks like it, though 19:00:31 |amethyst: Nice. 19:02:01 <|amethyst> success! 19:03:39 |amethyst: Thank you for the help. 19:04:39 <|amethyst> (not cherry-picking this whitespace fix) 19:04:54 03|amethyst02 07* 0.15-a0-59-gf7e0886: Don't crash when mapstat vault statistics generation is slow. 10(11 minutes ago, 1 file, 3+ 0-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=f7e08864e5bd 19:04:54 03|amethyst02 07* 0.15-a0-60-g2d7d1c5: Tab is for losing weight, not indenting Crawl. 10(47 seconds ago, 1 file, 1+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=2d7d1c5946cf 19:04:54 Cherry-picked 1 commit into stone_soup-0.14 19:05:39 |amethyst: good commit message 19:06:29 How exactly would you use tab for losing weight? Isn't it a soda? 19:06:34 |amethyst: you probably want to cherry-pick both commits, for that matter. 19:06:47 ??tab[2 19:06:47 tab[2/4]: http://images.wikia.com/mspaintadventures/images/9/99/00710.gif 19:07:29 ?/tab 19:07:30 Matching terms (27): batstabbing, club_stabbing, ellitab, fast_metabolism, genetically_stable, genetically_unstable, mace_stabbing, meatstab, meatstabbing, metabolic_englaciation, metabolism, mummystab, mummystabbing, needlestab, shadow_stab, slow_metabolism, stab, stab_bonus, stabbing, stabbing_skill, stable, stabwound, sustab, tab, tabrobin, you_feel_strangely_unstable, ~tabstorm_next; entries (... 19:07:34 oops 19:08:10 ??tabstorm 19:08:10 I don't have a page labeled tabstorm in my learndb. 19:08:13 ?/tabstorm 19:08:14 Matching terms (1): ~tabstorm_next 19:10:44 -!- mizu_no_oto has quit [Quit: Computer has gone to sleep.] 19:14:00 -!- Amnesiac has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/] 19:14:23 -!- notcluie has quit [Ping timeout: 245 seconds] 19:16:22 -!- Akien has quit [Quit: Quitte] 19:16:26 -!- floatboth has quit [Ping timeout: 265 seconds] 19:19:14 -!- NomadJim has quit [Read error: Connection reset by peer] 19:20:06 <|amethyst> reaverb: The second sugarless soda 19:20:14 <|amethyst> reaverb: after Diet Rite 19:20:28 |amethyst: Ha 19:20:50 |amethyst: Wow, hmm. 19:20:58 -!- bonghitz has quit [Quit: Leaving...] 19:21:51 Is coinflip() used a lot in level gen? (Enough that an optimization would have have visible effects) 19:22:11 -!- Reign_ has quit [Quit: Page closed] 19:22:41 Cherry-picked 1 commit into stone_soup-0.14 19:22:53 %git stone_soup-0.14 19:22:53 07|amethyst02 * 0.14-b1-26-geea93df: Tab is for losing weight, not indenting Crawl. 10(19 minutes ago, 1 file, 1+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=eea93dfc6a10 19:22:53 <|amethyst> what kind of optimisation? 19:23:02 %git 0.13.0^ 19:23:02 07kilobyte02 * 0.13-b1-59-gcd28bc4: Don't process enchantments that have just been deleted. 10(6 months ago, 1 file, 10+ 7-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=cd28bc492a7f 19:23:06 %git 0.12.0^ 19:23:06 07Grunt02 * 0.12-b1-166-ga9ee58f: Fix typos in changelog. 10(11 months ago, 1 file, 3+ 3-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=a9ee58fbd0a2 19:23:09 hm 19:23:31 |amethyst: The kind that causes the RNG to be called less. (So time, mainly) 19:24:16 <|amethyst> reaverb: how would you have coinflip() call the RNG fewer than one time per call? By saving state? 19:24:43 |amethyst: Effectively yes. 19:24:59 |amethyst: Think about it, 1 call to the RNG give you 32 bits of entrophy. 19:25:12 No reason you can't use 1 call to the RNG for 32 calls of coinflip() 19:26:54 -!- mizu_no_oto has quit [Client Quit] 19:28:03 <|amethyst> hm... would have to give some thought to reproducibility, but in principle I don't see a problem 19:28:16 (And yes I'm aware this would need to be seralized to perserve reproducibility) 19:28:30 <|amethyst> better would be to make random2() also use an appropriate amount of entropy 19:28:54 |amethyst: That was my next step, but I wanted to make sure it actually worked first. 19:30:04 I don't know if keeping track of the state, how many times it's used and such would actually make it run faster, so I decided to just change coinflip() and test it with -mapstat, and you know the rest. 19:30:24 -!- buzzykins has quit [Read error: Operation timed out] 19:30:49 -!- Crehl has quit [Quit: qqqqqqq] 19:32:12 -!- jday_ has joined ##crawl-dev 19:34:07 horrible idea of the day 19:34:07 remove the haste spell 19:34:31 how bad is it 19:37:38 -!- ackack has quit [Ping timeout: 240 seconds] 19:37:38 -!- Patashu[Zzz] is now known as Patashu 19:43:11 -!- HellTiger_ has quit [Ping timeout: 268 seconds] 19:44:48 <|amethyst> reaverb: definitely talk to bh about it 19:45:08 |amethyst: Yes, that sounds like a good idea. 19:45:19 <|amethyst> reaverb: there may be a simpler way to pull the entropy directly out of the generator without going through get_uint32 19:45:39 |amethyst: Hm, I didn't think of that. 19:49:06 -!- ackack has joined ##crawl-dev 19:49:24 -!- dpeg has quit [Quit: Lost terminal] 19:50:23 -!- ldf has quit [Quit: ldf] 19:57:13 -!- ussdefiant has joined ##crawl-dev 19:59:55 -!- Basil has joined ##crawl-dev 20:00:31 -!- rast has quit [Read error: Connection reset by peer] 20:00:48 -!- dtsund has joined ##crawl-dev 20:00:48 -!- rast has joined ##crawl-dev 20:05:15 |amethyst: Looking at the code, get_uint32() is a leaf function. I doubt there's a way to get entropy from the RNG in a simply way than through that interface. 20:05:18 -!- Ququman has quit [Remote host closed the connection] 20:06:10 !tell bh Is there a way to get entropy from the RNG that isn't get_uint32()? I would like to grab random numbers smaller than uint32_t. 20:06:11 reaverb: OK, I'll let bh know. 20:07:01 <|amethyst> you're probably right 20:07:05 <|amethyst> I am not a RNG person 20:07:14 <|amethyst> I know enough about RNGs to know that I should not touch them 20:08:13 |amethyst: Yes. I'm very happy I don't actually have to touch the RNG - just it's inferface. 20:08:43 Although I am kind of sad I have to touch the seralization code. It's killed my ideas before. 20:09:43 -!- mizu_no_oto has quit [Quit: Computer has gone to sleep.] 20:10:22 <|amethyst> I don't think you'd have to serialise 20:10:33 <|amethyst> we don't serialise the RNG state currently afaict 20:10:41 -!- Kalir has quit [Changing host] 20:11:07 <|amethyst> the important thing is that starting two new games with the same seed, and taking the same sequence of actions, should produce the same result 20:11:46 <|amethyst> (two new games in exactly the same version of the code) 20:11:58 |amethyst: Really, I thought the point was that you that also applied to in progress saves. 20:11:58 reaverb: You have 1 message. Use !messages to read it. 20:12:03 !messages 20:12:04 (1/1) Basil said (1m 27s ago): Singularity still needs some work, but maybe after some more testing and feedback. For example, it currently deals 5d20 AC-piercing damage to adjacent creatures since I don't care to drum up reasonable numbers yet. 20:12:29 |amethyst: That's nice to know, should be fine than. 20:12:39 <|amethyst> reaverb: I don't think so; the reason reproducibility is important is so that, if a stress-testing bot crashes, we can repeat the crash 20:12:52 -!- Zermako has quit [Quit: good night] 20:12:56 <|amethyst> kilobyte put a lot of work into making that happen last year-ish 20:13:16 |amethyst: Ah, I thought it was also so that if a player crashed, their crash could also be reproduced. 20:13:39 |amethyst: But if it's only for stress testing bots, then it should be fine. 20:13:45 <|amethyst> that's next to impossible, since you'd have to know the exact sequence of steps they took 20:13:57 <|amethyst> but the bot is presumably deterministic given a seed 20:14:40 |amethyst: I guess I thought tyrecs could be used? The real reason makes a lot more sense. 20:15:13 -!- Nunya has joined ##crawl-dev 20:16:29 -!- Bloax has quit [Quit: Error: Bloax not found] 20:16:47 -!- jeanjacques has quit [Ping timeout: 252 seconds] 20:19:25 -!- Chippums has quit [Quit: Page closed] 20:20:56 -!- SamB has quit [Read error: Connection reset by peer] 20:23:21 -!- nrook has joined ##crawl-dev 20:23:26 -!- SamB has joined ##crawl-dev 20:27:18 -!- ayutzia has quit [Ping timeout: 245 seconds] 20:28:02 -!- ussdefiant_ has joined ##crawl-dev 20:28:38 -!- ussdefiant has quit [Ping timeout: 240 seconds] 20:28:42 -!- ussdefiant_ is now known as ussdefiant 20:29:00 -!- WWWYZZERDD has quit [Quit: Page closed] 20:43:13 -!- Jdc1197 has quit [Ping timeout: 240 seconds] 20:47:53 -!- WalkerBoh has quit [Remote host closed the connection] 20:51:28 -!- nrook has quit [Remote host closed the connection] 20:59:14 -!- Kalir has quit [Ping timeout: 240 seconds] 21:02:38 -!- Xenobreeder has quit [Ping timeout: 240 seconds] 21:08:02 If I am lethally poisoned (17 -> 0) is there any chance I'll survive w/ normal health regen? 21:09:23 -!- trystero_ has quit [Ping timeout: 245 seconds] 21:10:39 Lasty1: Yes, but don't count on it. 21:10:58 reaverb: thanks. 21:11:31 Lasty1: Did you get the "lethaly posioned" message without being additionally poisoned by something? 21:11:36 In this case, no :( 21:11:49 er 21:11:58 Lasty1: You didnt' survive? 21:12:01 I got lethally poison message while trying to move 21:12:07 yeah, didn't 21:12:18 adders give a lot of poison now 21:12:50 <|amethyst> Adders gave a lot of poison before 21:13:16 <|amethyst> you just didn't find out how much until you died :P 21:13:31 I think I would have survived previously in this case -- there was a !hw. I could be wrong. 21:14:52 <|amethyst> how much did you gain from the !hw? 21:14:52 -!- gammafunk has joined ##crawl-dev 21:14:53 went back up to 26 hp 21:15:02 adder died when I was at 17 21:15:25 drank the potion before it died tho 21:16:14 <|amethyst> and it bit you after you drank the potion I imagine 21:16:21 yeah, once more 21:17:24 <|amethyst> getting freshly poisoned at 17 hp wasn't that great before the change either, particularly when you already had some poison in you 21:18:02 true 21:18:19 -!- gammafunk has quit [Client Quit] 21:19:04 <|amethyst> though perhaps you got unlucky with the 1/3 chance to get poisoned each bite 21:20:03 perhaps. No biggie. Adders happen. I asked initially in the hopes of determining if exploring was better or worse than hiding 21:20:24 <|amethyst> doesn't really matter unless you have slow healing, I think 21:20:57 Well, if you're very lucky exploring gets you a new potion 21:21:08 so if you'll die from resting, you should explore 21:21:52 <|amethyst> yeah 21:22:46 <|amethyst> perhaps @ should show negative numbers? 21:22:56 that would be helpful 21:23:14 <|amethyst> I guess that would be bad for hope though :P 21:23:27 Accurately bad, tho 21:23:29 <|amethyst> but good for rational decisions 21:24:04 I mean, showing the "you die" message is bad for hope too, but it's honest. :) 21:25:28 -!- ussdefiant_ has joined ##crawl-dev 21:26:38 -!- ussdefiant has quit [Ping timeout: 240 seconds] 21:26:51 -!- ussdefiant_ is now known as ussdefiant 21:27:39 <|amethyst> elliptic: what do you think about showing negative numbers for poisoning in @ so the player can tell when there's no hope of a rounding error saving their life? 21:28:25 <|amethyst> elliptic: s/rounding/approximation/ 21:28:50 that might be reasonable, not sure 21:29:14 <|amethyst> it would also tell you approximately how many !/hw you need 21:29:22 that's true 21:30:31 -!- Lasty1 has quit [Quit: Leaving.] 21:31:11 I guess we might as well do it, yeah 21:32:22 I think probably the best way of doing it is to add an extra parameter allow_negative to poison_survival()? 21:32:47 <|amethyst> checking where it is used 21:33:12 -!- reaverb1 has joined ##crawl-dev 21:33:26 it is used to make HP bars in multiple places 21:33:38 -!- reaverb has quit [Ping timeout: 240 seconds] 21:33:39 <|amethyst> yeah 21:34:38 <|amethyst> and poison_is_lethal() uses ! but of course that could be fixed 21:34:50 I guess people should really be using poison_is_lethal() anyway if they just want to check whether the poison is lethal 21:35:07 so changing that and just letting it return negative numbers in general might be fine 21:35:12 (and changing the HP bar code) 21:39:01 what does "0xffffffff 0x0 + 4294967295" mean in a stack trace? 21:39:06 -!- reaverb1 is now known as reaverb 21:39:49 <|amethyst> poison_is_lethal() uses ! but of course that could be fixed 21:39:56 <|amethyst> reaverb: in what context? 21:40:04 <|amethyst> reaverb: it's -1 21:40:23 |amethyst: Ok, what does "-1" mean? 21:40:29 <|amethyst> reaverb: being interpreted as unsigned, and maybe used in pointer arithmetic 21:40:41 <|amethyst> um 21:40:46 Full stack trace: http://bpaste.net/show/fp03h2uDEQYCnks0x3bc/ 21:41:25 <|amethyst> oh, that's something where the stack dumper is confused 21:41:40 |amethyst: I've written the random2() entropy conseveration code, but I've run into an odd problem. Running around and generating levels by visiting them works fine, but running -mapstats causes a "Virtual timer" crash. 21:41:46 <|amethyst> it thinks there's a frame there (below the signal trampoline), but that frame doesn't have a return address 21:41:59 <|amethyst> %git 21:42:00 07|amethyst02 * 0.15-a0-60-g2d7d1c5: Tab is for losing weight, not indenting Crawl. 10(3 hours ago, 1 file, 1+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=2d7d1c5946cf 21:42:03 <|amethyst> %git HEAD^ 21:42:04 07|amethyst02 * 0.15-a0-59-gf7e0886: Don't crash when mapstat vault statistics generation is slow. 10(3 hours ago, 1 file, 3+ 0-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=f7e08864e5bd 21:42:12 <|amethyst> did you get that one? 21:43:05 -!- CampinSam has quit [Quit: WeeChat 0.4.2] 21:43:09 * reaverb facepalms 21:43:20 <|amethyst> I thought you were here when I committed that :) 21:43:33 |amethyst: I pulled it into my master, but I didn't merge it into the branch. 21:43:44 <|amethyst> git rebase master 21:43:58 <|amethyst> rebase is better than merge IMO if you haven't pushed yet 21:44:09 |amethyst: I know the command! I just forgot to do it. : ( 21:44:17 -!- Acidburn6 has quit [] 21:44:20 |amethyst: And I also know about rebasing. 21:44:25 <|amethyst> ah, you said "merge" so I thought you meant "merge" :) 21:44:43 |amethyst: Yes, that makes sense. 21:45:04 -!- Foamed has quit [Quit: Heading to bed.] 21:49:47 -!- moose has quit [Quit: Leaving] 21:50:05 Hmm, I one hand I finally got this to work. 21:50:25 On the other hand, it looks like my "optimization" doesn't actually make -mapstat go faster... 21:50:39 <|amethyst> how much time is spend in the rng in the first place 21:50:47 <|amethyst> s/pend/pent/ 21:51:11 |amethyst: I should probably have looked at that before starting work on this ._. 21:51:30 |amethyst: Now that I think about it, a leaf function probably doesn't take much time to run. 21:52:00 <|amethyst> youah, it's 1.5 multiplications and a bunch of cheaper operations 21:52:03 <|amethyst> s/ou// 21:52:09 <|amethyst> s/yah/yeah/ 21:52:47 <|amethyst> well, one of those multiplications is 64-bit 21:52:53 <|amethyst> err, .5 of them 21:53:12 |amethyst: Oh well, even if this doesn't trun out to be worthwhile, I still learned. 21:53:15 <|amethyst> :) 21:53:53 <|amethyst> oh, actually 21:54:30 <|amethyst> get_uint32 is 1.48% for mapstat, random2 is 2.27% 21:55:24 <|amethyst> but mapstat is kind of skewed 21:55:24 <|amethyst> since #1 is _random_map_in_list at 4.36% 21:55:24 |amethyst: So there's probably some gain, but not much. Might not be worth around a 100 lines of code to get that gain. 21:55:24 %git :/ice storm 21:55:24 07tenofswords02 * 0.14-a0-2400-gb2944a8: Give pan lords rare quicksilver bolt, orb of elec 10(9 weeks ago, 1 file, 3+ 1-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=b2944a8389eb 21:55:28 <|amethyst> reaverb: there might be gain 21:56:02 <|amethyst> reaverb: if what you did is faster than a couple of multiplications and a bunch of shifts 21:56:35 <|amethyst> now you see why I build my diagnostics builds with -pg :) 21:56:40 |amethyst: Around how many shifts is a multiplication worth? 21:56:51 <|amethyst> I don't know these days 21:56:56 |amethyst: Wait, there's an option for that? Wow. 21:56:56 <|amethyst> I could have told you on a 386 21:57:14 <|amethyst> yeah, use -pg, then run gprof on the binary when you're done 21:57:25 <|amethyst> I recommend the abyss stress test rather than mapstat 21:57:32 <|amethyst> if you want to know about real crawl games 21:58:06 |amethyst: Not sure how to run abyss stress tests, is it a single command? 21:59:37 <|amethyst> ./test/stress/abyss [seed] (it uses the PID as the seed by default) 22:00:32 <|amethyst> (which runs ./crawl, so if you rename your debug binary something different, you'll just want ./crawl-debug -name "abyss" -rc test/stress/abyss.rc and maybe a seed 22:00:35 <|amethyst> ) 22:00:40 <|amethyst> but build with -pg first 22:00:48 -!- rast has quit [Read error: Connection reset by peer] 22:01:05 -!- rast has joined ##crawl-dev 22:01:55 -!- mamgar has quit [Read error: Connection reset by peer] 22:01:56 <|amethyst> hmm 22:02:23 |amethyst: I try "make debug -pg APPLE_GCC=y" but I get an error. 22:02:25 <|amethyst> one problem with that is that make debug turns off optimisations 22:02:36 <|amethyst> make EXTERNAL_FLAGS=-pg EXTERNAL_LDFLAGS=-pg APPLE_GCC=y debug 22:02:43 <|amethyst> or with debug at the beginning 22:03:33 |amethyst: So something which is slightly optimized will do better, even if the compiler can make that differance moot. Hmm. 22:03:46 Also, thanks for all the help. 22:03:58 -!- NomadJim has quit [Read error: Connection reset by peer] 22:04:24 -!- Daekdroom has quit [Quit: Saindo] 22:05:08 <|amethyst> I mean, if you want to tell whether your change will really reduce CPU usage for players, you should be compiling with optimisations, since that's how the prebuilt binaries are compiled 22:05:11 -!- Lawman0 has quit [Ping timeout: 246 seconds] 22:05:50 <|amethyst> since things like inlining can make a huge difference 22:06:11 <|amethyst> (though I guess inlining wouldn't be applicable here since it isn't in a header file, unless you use LTO) 22:06:41 <|amethyst> (and I don't know enough about LTO to know whether it would inline that kind of thing) 22:07:29 |amethyst: And of course I'm on a mac, and apple gcc isn't trusted with optimizations. (I don't think level gen will break it though) 22:07:41 <|amethyst> well 22:08:16 <|amethyst> what version of gcc is that? 22:08:34 -!- fcrawl has quit [Ping timeout: 240 seconds] 22:08:53 4.2.1 22:09:10 <|amethyst> if you can get a newer gcc or clang you can turn on optimisations 22:09:20 <|amethyst> it's apple's 4.2 that is busted 22:09:24 (todo figure out how we can move Crawl builds to clang) 22:09:29 (for OS X) 22:09:42 |amethyst: I think I already downloaded homebrew, but I don't actually know what command to access it. 22:10:11 <|amethyst> I don't know much about Mac dev tools 22:10:40 |amethyst: Ok, I'll go elsewhere for that advice. 22:10:48 |amethyst: How do I stop the abyss bot? 22:11:32 <|amethyst> I think you can just press keys 22:11:50 <|amethyst> so ^Qyes 22:12:06 <|amethyst> ah, just ^Q 22:12:23 |amethyst: Thank you 22:12:46 -!- hurdos1 has quit [Quit: Leaving.] 22:13:00 Hmm, How do I see gmon.out ? 22:13:54 <|amethyst> gprof crawl 22:14:36 <|amethyst> or you can rename the gmon.out to something else and do gprof crawl blah.gmon 22:14:59 <|amethyst> works kind of like gdb in that regard, but with gmon.out instead of core 22:15:11 <|amethyst> err 22:15:16 <|amethyst> you probably want to redirect that 22:15:20 <|amethyst> it goes to stdout 22:15:35 |amethyst: I try 22:15:39 -!- sgun has quit [Ping timeout: 255 seconds] 22:15:40 -!- Whistling_Bread has quit [Ping timeout: 255 seconds] 22:15:40 <|amethyst> and for crawl it is going to be much bigger than your terminal's scrollback :) 22:15:44 |amethyst: accidently hit enter 22:15:59 -!- Chousuke_ has quit [Remote host closed the connection] 22:16:20 |amethyst: I tried "gprof crawl" but I get a "can't open: gmon.out (No such file or directory)" error. 22:16:32 I compiled with make EXTERNAL_FLAGS=-pg EXTERNAL_LDFLAGS=-pg APPLE_GCC=y debug 22:16:52 I have a probably really dumb question. I updated some of my dev branches (been probably a month) and one of my conflicts is in ./dat/descript/quotes.txt where the Ice Storm Spell quote is not in the current REMOTE. I did a git log of ./dat/descript/quotes.txt but I don't see any recent commits that removed it. Does someone have a minute to help explain what happened? (I'm not super... 22:16:52 <|amethyst> and then you ran it 22:16:53 ...experienced with git, but I'm not a total newb) 22:17:21 sorry, I'm trying to merge the latest master into one of my branches 22:17:22 |amethyst: yes. 22:17:57 johnstein: Considering the entire spell Ice Storm was removed, it would make since the quote would be gone too. 22:18:06 <|amethyst> reaverb: I thought you said it gave you a gmon.out... 22:18:07 yea, I assume that's what happened 22:18:12 johnstein: rather than merge, have you considered rebasing the branch onto master? 22:18:16 <|amethyst> reaverb: are you in the same directory you ran it from? 22:18:16 (or is that what you're doing?) 22:18:28 |amethyst: No, it never gave me a gmon.out. I'm still in the same directory. 22:18:31 -!- rchandra has left ##crawl-dev 22:18:44 grunt: I keep having people say I should rebase. I should start doing that. 22:19:08 johnstein: git merge --abort ; git rebase master 22:19:13 <|amethyst> johnstein: johnstein 22:19:14 <|amethyst> err 22:19:17 |amethyst: |amethyst 22:19:20 <|amethyst> %git dc7c8162d 22:19:22 07|amethyst02 * 0.14-a0-2887-gdc7c816: Remove unused descriptions and quotes. 10(5 weeks ago, 50 files, 1+ 461-) 13http://s-z.org/neil/git/?p=crawl.git;a=commitdiff;h=dc7c8162d385 22:19:49 -!- nrook has joined ##crawl-dev 22:19:57 thanks |amethyst. I'll try that grunt 22:23:52 |amethyst: Is there a way to verify whether or not a build has gprof? I have a nasty feeling that for some reason the -pg flag isn't working correctly. 22:23:58 -!- clouded_ has joined ##crawl-dev 22:24:12 <|amethyst> hm 22:24:31 <|amethyst> ./crawl --version will show you compiler flags 22:24:52 |amethyst: It does say "-pg" 22:25:11 <|amethyst> also, I don't know if there's something special you need there for Macs 22:25:23 <|amethyst> might be best to google that, or talk to geekosaur or somebody 22:25:54 -!- st_ has quit [Ping timeout: 240 seconds] 22:25:57 I haven't done much with profiling 22:26:37 Wow, gprof just flatly doesn't work on macs according to stack exchange and other websites. 22:26:55 Apparently Apple has their own tool called "Shark" thought. 22:27:22 read on 22:27:29 Shark was replaced by Instruments 22:27:33 Sounds pretty fishy to me. 22:27:52 <|amethyst> geekosaur: when? because these are 10.5 build tools 22:28:03 when dtrace was added, so post 10.5 22:28:29 (also Instruments needs dtrace probes to really give you useful output) 22:28:41 -!- Netmonmatt has quit [Read error: Connection reset by peer] 22:28:55 I'm pretty sure 10.5 has dtrace ... 22:29:03 <|amethyst> reaverb: so, if you figure this out, give us a patch to fix our makefile :) 22:29:21 <|amethyst> reaverb: in the APPLE_GCC case anyway 22:29:37 |amethyst: I'm sorry, what do you mean "fix our makefile"? 22:29:48 <|amethyst> oh 22:29:50 <|amethyst> never mind 22:30:01 SamB: It does. 22:30:03 -!- omnirizon has quit [Ping timeout: 255 seconds] 22:30:05 <|amethyst> I forget we don't have anything in there about profiling 22:30:34 <|amethyst> maybe we should have a PROFILE=y variable once you figure out what to do on mac 22:30:55 <|amethyst> so that you can turn out profiling more easily 22:30:57 -!- Staplefun has quit [Ping timeout: 255 seconds] 22:31:00 <|amethyst> s/out/on/ 22:31:02 <|amethyst> when you make 22:31:16 SamB: it did but it wasn;t integrated into xcode yet 22:31:35 what is this "xcode" of which you speak 22:31:48 the apple component giving you things like compilers and headers 22:32:05 * SamB has actually heard of it; he just means he didn't really play with the IDE proper 22:32:29 the command line tools are still part of the xcode suite. (but Instruments, and for that matter Shark, are part of the IDE) 22:33:07 -!- clouded_ is now known as st_ 22:33:47 they looked like distinct apps to me! 22:33:55 geekosaur: So how exactly do I activate Shark? 22:34:07 I know I have that on my mac. 22:34:30 -!- BirdoPrey has quit [Quit: BirdoPrey] 22:35:14 -!- MaxFrost has quit [Ping timeout: 268 seconds] 22:36:05 hm, I know less about the tools on the older xcode versions. I believe it's a GUI app though, and you would run it and then open the program through it 22:36:40 geekosaur: Found it. 22:37:07 thanks. 22:37:10 I assume you were using the stackoverflow page, which gave a path for it 22:38:20 -!- Vaporware has joined ##crawl-dev 22:38:20 -!- Vaporware has quit [Changing host] 22:38:20 -!- Vaporware has joined ##crawl-dev 22:38:22 geekosaur: I have the program open, but I have no idea how to use it. 22:39:16 I don't have 10.5, I played a little with Instruments but don't know Shark, sorry 22:39:33 * SamB supposes he did actually run Xcode itself a fair amount to view documentation? 22:41:13 -!- DrinkMachine has quit [Ping timeout: 240 seconds] 22:42:18 -!- raskol has quit [Ping timeout: 240 seconds] 22:42:28 <|amethyst> reaverb: and I guess what i was saying earlier about 1.5 multiplications isn't true, because random2() itself has not only division, but also a loop 22:43:04 |amethyst: Yeah, I guess that might be where most of the optimization comes in. 22:44:53 eww 22:45:21 <|amethyst> well 22:45:22 how would there be half a multiplication, anyway? 22:46:42 <|amethyst> SamB: I guess that's the "alternating" in "alternating step generator" 22:49:13 <|amethyst> SamB: it does one 32-bit multiply in the outer LFSG, and based on the low-order bit of that, either uses another LFSG (64-bit multiply), or a sequence of shifts and xors 22:50:27 <|amethyst> s/G/R/g 22:50:45 <|amethyst> and there's a LCG added to that 22:51:06 <|amethyst> err 22:51:19 <|amethyst> I guess the one inside is actually a LCG 22:51:27 <|amethyst> like I said, I'm not a RNG person 22:55:53 -!- lobf has quit [Quit: lobf] 22:56:09 -!- reaverb has quit [Ping timeout: 255 seconds] 23:02:14 -!- reaverb has joined ##crawl-dev 23:08:25 -!- alefury has quit [Read error: Connection reset by peer] 23:13:09 -!- bh has joined ##crawl-dev 23:14:38 -!- ussdefiant has quit [Ping timeout: 240 seconds] 23:15:20 !seen reaverb 23:15:21 I last saw reaverb at Sat Apr 5 04:02:14 2014 UTC (13m 6s ago) joining the channel. 23:15:28 bh: Hey 23:15:30 reaverb: what do you want the rng to do for you? 23:16:39 bh: I was going to take less entropy from the RNG for each call of random2(), coinflip(), etc. I implemented it with just get_uint32_t(), but it looks like it isn't much faster. 23:16:57 yeah, don't bother 23:17:05 it increases complexity 23:17:46 reaverb: that might make sense if the RNG actually kept a supply of entropy around ... 23:18:46 SamB: Part of making it take less entrophy would be storing the entropy which isn't used. 23:19:05 the RNG is cheap enough that I don't think there's an advantage to lugging around an entropy pool 23:19:05 reaverb: I mean, it's just a PRNG anyway 23:19:23 so there's not really much entropy in the system to start with 23:19:49 it's no /dev/random 23:19:52 bh: Yes, I've discovered that when I tried to implement this. That's some pretty incredible coding. 23:20:18 -!- omnirizo1 has quit [Ping timeout: 240 seconds] 23:21:44 I'm probably going to drop this, at least for now. 23:22:03 -!- BirdoPrey has quit [Quit: BirdoPrey] 23:23:26 -!- Vaporware has quit [Quit: Leaving] 23:23:49 it's a pretty bad PRNG 23:24:07 It passes all of diehard, but it really shouldn't be used for anything serious 23:24:29 the problem we had with Mersenne Twister is that you could access RNG state through lua and then decode the internal state 23:24:30 bh: Oh yeah, nothing like crypto, but it serves something like Crawl really well. 23:25:59 <|amethyst> I figured you'd have that option turned off 23:26:01 <|amethyst> err 23:26:13 <|amethyst> ??|amethyst 23:26:13 |amethyst[1/9]: <|amethyst> doh 23:28:55 bh: and what's to stop anyone from attacking this one in a similar way? 23:29:25 SamB: if someone wants to cryptanalyze an alternating step generator to cheat at crawl, let 'em. 23:29:33 hehehe 23:29:43 <|amethyst> then they can publish 23:30:13 er, I thought bh just said it wasn't appropriate for any serious usage? 23:30:29 which would seem to imply that papers about it wouldn't be terribly well recieved? 23:30:32 my recollection is that there are no known attacks on alternating step generators when the LFSR polynomials are unknown 23:30:47 I have no idea about results where the polynomials are public 23:30:49 hmm 23:30:52 <|amethyst> yeah, but they're also not as well-studied as some of the other stuff 23:31:26 <|amethyst> it might be a specialty journal :) 23:31:33 it looks like the best time complexity is O(2^(2L/3)) where L is the number of bits in the shortest generator 23:32:53 <|amethyst> anyway, we don't demand cryptographic RNGs 23:33:51 MT had a fault that it was trivial to decode its internal state 23:34:03 <|amethyst> MT in general or our implementation? 23:34:07 we can't just read from /dev/urandom, because we need it to be reproducible 23:34:15 bh: I know that 23:34:23 |amethyst: If you have access to the unaltered emissions from MT, you can always decode the state 23:34:40 <|amethyst> ah 23:34:49 I was just saying trying to "conserve entropy" didn't sound like a real goal 23:35:11 <|amethyst> SamB: I think really reaverb was trying to save time by calling the RNG less 23:35:16 -!- KurzedMetal has quit [Read error: Connection reset by peer] 23:35:31 <|amethyst> SamB: but the RNG is so simple that maybe that's not such a big deal 23:35:41 since it's not like that was actual hardware-generated entropy or anything 23:35:44 |amethyst basically got it right. 23:35:49 yeah 23:35:55 it's an xorshift generator, a multiply-with-carry-generator, a LSFR and something else 23:36:07 I can reimplement it in ASM if we want it to be faster ;) 23:36:37 <|amethyst> but the raspi builds! 23:37:44 |amethyst: hah, as if it'd be x86-asm-or-nothing 23:38:20 <|amethyst> I mean, they're the ones who would need the speed-up the most 23:38:28 <|amethyst> so it should be ARM assembler! 23:38:55 hehehe 23:39:24 <|amethyst> actually, no, PPC 23:39:33 <|amethyst> for the ancient mac users 23:39:39 <|amethyst> who get -O0 anyway 23:40:38 can we please, please move on to C++11? 23:40:46 ppc32 or ppc64? 23:40:58 |amethyst: crawl isn't the sort of application that should be finding compiler bugs 23:41:30 what, only well-written apps are supposed to find compiler bugs now? 23:42:17 <|amethyst> bh: I don't know, it has had a lot of code bums for programmers 23:42:42 it's a roguelike. It isn't like crawl is doing anything terribly advanced 23:42:47 <|amethyst> in the sense of optimizing for brevity and/or 30-year-old computers 23:43:11 <|amethyst> I mean, no, we won't tickle many parallelism bugs 23:43:13 <|amethyst> oh btw 23:43:25 <|amethyst> can/does someone here do Windows builds? 23:43:37 <|amethyst> we should disable paralled DB init on windows 23:43:46 <|amethyst> s/lled/llel/ 23:44:08 -!- Kalir has quit [Changing host] 23:44:21 why's that? 23:45:31 <|amethyst> searching mantis for it 23:47:43 <|amethyst> %bug 5354 23:47:43 https://crawl.develz.org/mantis/view.php?id=5354 23:50:47 <|amethyst> I don't believe a single person has reported that bug on another OS 23:50:54 -!- reaverb has quit [Quit: Leaving.] 23:51:11 <|amethyst> well 23:51:24 <|amethyst> I mean, "random startup fail", maybe 23:51:30 <|amethyst> but all the linked reports are Windows 23:52:06 <|amethyst> and several explicitly mention the db (Cannot find data file...) 23:52:09 random startup fail? This is crawl, not Silicon Valley. 23:52:51 -!- ackack has quit [Ping timeout: 255 seconds] 23:53:53 <|amethyst> those are hardly random 23:53:54 -!- CKyle has quit [Quit: CKyle] 23:53:58 -!- tabstorm has quit [Ping timeout: 245 seconds] 23:54:43 -!- ussdefiant has joined ##crawl-dev 23:58:35 -!- lobf has joined ##crawl-dev 23:59:24 -!- caleba has quit [Quit: caleba]