00:00:21 the random2(you.piety) > 120 && one_chance_in(4) checks should be brought back 00:01:11 yep 00:01:16 I just just reading through that commit 00:01:20 and came to the same conclusion 00:01:36 so that would be why piety gain was so bad; that logic is broken and gifts are way too frequent 00:02:22 good catch; I think I kept seeing that "automatic gift at 5*" behaviour too and it seemed a bit strange but I assumed it was a coincidence 00:02:24 I believe what happens is that as soon as you get to ***** piety, you get a gift, and then gift timeout stops you from raising piety for a while 00:02:28 exactly 00:02:31 and then when gift timeout ends, you immediately get another gift 00:02:50 so it's basically impossible to build up piety beyond the very edge of ***** 00:02:50 those gifts will prevent piety gain, then when you overcome it you just get another gift, hence never gain piety past the breakpoint 00:03:06 cool, well that's an easy fix and solves that mystery! 00:03:13 I'm not sure if we blame ebering or alexjurkiewicz on that one 00:03:30 tbf that whole bit of logic could have used refactoring and was confusing 00:03:39 elliptic: did you want to make a patch for that or should I? 00:04:52 why don't you, I'm not really set up on this computer to push commits 00:05:18 will do and push in a bit 00:05:48 i think you can blame me. It looks like I moved the check further down 00:05:48 still find it hard to follow this logic even now 00:06:18 i guess it figures out both if you get a gift and what gift type at the same time 00:06:41 yes that logic is confusing 00:06:48 I'm going to refactor it to be a bit more clear 00:09:07 alexjurkiewicz: I actually forgive you for the trog piety bug, what I don't forgive you for is the indent at religion.cc:1367 00:09:18 for that a quokka will be punished 00:27:34 but their smiles! 01:20:58 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-3001-g9d8a61a446 02:59:17 03gammafunk02 07* 0.25-a0-835-gb37b1c2: Fix Trog weapon gifting (elliptic) 10(24 minutes ago, 1 file, 22+ 21-) 13https://github.com/crawl/crawl/commit/b37b1c29c382 03:11:52 Unstable branch on crawl.beRotato.org updated to: 0.25-a0-835-gb37b1c2 (34) 03:32:53 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-835-gb37b1c29c3 (34) 05:48:46 I might self nominate here 05:49:19 !learn add big bug Can't reach 6* piety with Trog for 5 months (b37b1c2) 05:49:19 big[2/2]: bug Can't reach 6* piety with Trog for 5 months (b37b1c2) 05:49:27 !learn del big[2 05:49:28 Deleted big[2/2]: bug Can't reach 6* piety with Trog for 5 months (b37b1c2) 05:49:33 !learn add big_bug Can't reach 6* piety with Trog for 5 months (b37b1c2) 05:49:33 big bug[1/1]: Can't reach 6* piety with Trog for 5 months (b37b1c2) 05:49:47 !learn mv big_bug big_bugs[2] 05:49:47 Syntax: !learn mv SRC[x] DST[y] or SRC[x] DST or SRC DST 05:49:52 !learn mv big_bug[1] big_bugs[2] 05:49:52 big_bug[1] -> big bugs[2/2]: Can't reach 6* piety with Trog for 5 months (b37b1c2) 08:45:19 Could it be that we have the same problem with Sif? Except that Sif runs out of gifts eventually. 09:08:10 ??big bugs 09:08:10 big bugs[1/2]: Every monster generating awake on D:1 (19abc6c) 09:10:48 !learn add big_bugs shock serpent retaliation attacks didn't check rElec for 4 years after their big revamp in 2014 09:10:49 big bugs[3/3]: shock serpent retaliation attacks didn't check rElec for 4 years after their big revamp in 2014 09:10:51 Yermak: that's possible 09:14:11 Yermak: Wise Wolf did an analysis of the Sif piety in Discord, I have copy pasted it here if you want to check: https://github.com/crawl/crawl/issues/1238 09:14:43 looks like it's not technically broken, but should be tuned to not gift that often 09:16:59 and I'm not sure if Trog was broken at the time he did the analysis, so he may have compared it to the broken Trog logic? 09:53:35 it looks like the chance is probably too high there even if it isn't 100%, yeah 09:56:12 also I had forgotten that piety gain is actually still technically possible under gift timeout, just really slow (especially with trog's piety decay being one of the faster ones) 09:58:15 it looks like that analysis is comparing with pre-breakage trog logic (which was similar to oka logic) 09:59:06 anyway I think it would be pretty safe to add a one_chance_in(2 or 3 or 4 or something) check on top of the current random2(piety)>100 and see what it feels like 10:26:41 elliptic! 10:26:44 hi 10:26:55 hey 10:28:35 !seen reaverb 10:28:35 I last saw reaverb at Sat Feb 15 06:49:26 2020 UTC (9w 5d 7h 38m 59s ago) quitting, saying 'Quit: Leaving'. 10:30:20 I think at some point a year or two ago someone may have tried to check sif for outright bugs along those lines (because this question comes up regularly), but that isn't really any guarantee it still doesn't have any 10:34:02 isn't this the kind of thing an integration test could catch? 10:34:55 e.g. use wizmode to repeatedly do things that trog favours, and assert that piety rises 10:43:46 New branch created: all_item_subtypes (1 commit) 13https://github.com/crawl/crawl/tree/all_item_subtypes 10:43:46 03Aidan Holm02 07[all_item_subtypes] * 0.25-a0-834-gc1d408b: Add all_item_subtypes() 10(21 hours ago, 6 files, 49+ 11-) 13https://github.com/crawl/crawl/commit/c1d408b9e858 10:58:01 aidanh maybe, I think the testing logic could be pretty complicated exactly because of its stochastic interaction with gifting 10:58:13 Old item conversion in unmarshallItem() leads to multiple item stacks 13https://crawl.develz.org/mantis/view.php?id=12242 by aidanholm 10:58:18 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-834-gc1d408b: Add all_item_subtypes() 10(21 hours ago, 6 files, 49+ 11-) 13https://github.com/crawl/crawl/commit/c1d408b9e858 10:58:18 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-835-gf7581c1: Remove code for old potion types 10(2 days ago, 5 files, 86+ 533-) 13https://github.com/crawl/crawl/commit/f7581c18f35a 10:58:18 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-836-ga796a89: Remove is_blood_potion() 10(2 days ago, 6 files, 4+ 78-) 13https://github.com/crawl/crawl/commit/a796a8957db6 10:58:18 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-837-g337fb8b: Remove more old potion code 10(2 days ago, 7 files, 8+ 228-) 13https://github.com/crawl/crawl/commit/337fb8b4285f 10:58:18 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-838-g1c2ce65: Remove an unused parameter 10(2 days ago, 3 files, 9+ 12-) 13https://github.com/crawl/crawl/commit/1c2ce655efd9 10:58:52 so I'm not immediately sure where that specific automation project (vs. manual testing) lies in https://xkcd.com/1205/ 10:59:15 heh 10:59:32 well, what if it wasn't stochastic? 10:59:48 e.g. what if you could hook the RNG in tests 10:59:51 well, now that I say that I'm not sure if that xkcd chart is 100% the right way to think about automated testing 11:00:04 ah, it's true that seeding might help here 11:01:02 i don't mean seeding, i mean "in this test, make all rng calls from should_trog_give_a_gift() return 100" 11:01:13 right, also possible 11:01:26 i think seeding would be too brittle, and not unit-testy enough 11:01:42 (very hand-wavey, i know) 11:02:53 btw, potions of blood *are* supposed to be gone, yeah? 11:03:27 I think the multiple stack thing is known and basically considered harmless, if you drop and pickup the stacks combine 11:04:06 it happened with food too 11:04:21 right, i remember that 11:04:30 absolutely bananas 11:04:32 https://crawl.develz.org/mantis/view.php?id=11328 11:04:58 or lacking in bananas 11:10:40 well, yeah :( 11:17:01 ebering: !blood *is* supposed to be removed, right? 11:29:07 The build failed. (all_item_subtypes - c1d408b #13100 : Aidan Holm): https://travis-ci.org/crawl/crawl/builds/678629732 11:35:37 aidanh: yes 11:35:58 ok good, just checking; there was quite a bit of code left over 11:59:42 The build is still failing. (ifdefs-begone - 1c2ce65 #13101 : Aidan Holm): https://travis-ci.org/crawl/crawl/builds/678636278 12:18:35 03Aidan Holm02 07[all_item_subtypes] * 0.25-a0-834-gc905ea7: Don't mark elven and mossy runes as removed 10(78 seconds ago, 2 files, 2+ 4-) 13https://github.com/crawl/crawl/commit/c905ea7b4bb5 12:18:35 03Aidan Holm02 07[all_item_subtypes] * 0.25-a0-835-ge030b2a: Add all_item_subtypes() 10(23 hours ago, 6 files, 49+ 11-) 13https://github.com/crawl/crawl/commit/e030b2abfa3c 12:55:56 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-834-gc905ea7: Don't mark elven and mossy runes as removed 10(39 minutes ago, 2 files, 2+ 4-) 13https://github.com/crawl/crawl/commit/c905ea7b4bb5 12:55:56 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-835-g37239e2: Add all_item_subtypes() 10(23 hours ago, 7 files, 51+ 11-) 13https://github.com/crawl/crawl/commit/37239e2f2b07 12:55:56 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-836-gbc34d4a: Remove code for old potion types 10(2 days ago, 5 files, 86+ 533-) 13https://github.com/crawl/crawl/commit/bc34d4a1c8cd 12:55:56 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-837-g9d6bd0d: Remove is_blood_potion() 10(2 days ago, 6 files, 4+ 78-) 13https://github.com/crawl/crawl/commit/9d6bd0d62454 12:55:56 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-838-gae6a826: Remove more old potion code 10(2 days ago, 7 files, 8+ 228-) 13https://github.com/crawl/crawl/commit/ae6a82687a1e 12:55:56 03Aidan Holm02 07[ifdefs-begone] * 0.25-a0-839-gbe9270b: Remove an unused parameter 10(2 days ago, 3 files, 9+ 12-) 13https://github.com/crawl/crawl/commit/be9270b9c7a8 13:07:10 Unstable branch on crawl.akrasiac.org updated to: 0.25-a0-835-gb37b1c2 (34) 13:10:26 03Aidan Holm02 07* 0.25-a0-836-g827379b: Don't mark elven and mossy runes as removed 10(53 minutes ago, 2 files, 2+ 4-) 13https://github.com/crawl/crawl/commit/827379bd5bde 13:10:26 03Aidan Holm02 07* 0.25-a0-837-g13a0117: Add all_item_subtypes() 10(24 hours ago, 7 files, 51+ 11-) 13https://github.com/crawl/crawl/commit/13a0117a0c60 13:18:49 heh I wonder if there's now a way to get 16-17 runes via upgrade if you have a trunk save from the exact right time 13:20:39 maybe runes were never autoremoved 13:20:45 tbh i didn't look all that closely at the rune handling code 13:21:40 just a quick check of the CL that marked them as removed in the first place 13:21:40 s/CL/commit 13:29:50 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-837-g13a0117a0c (34) 13:54:29 03Aidan Holm02 07* 0.25-a0-838-gffd7542: Remove code for old potion types 10(2 days ago, 5 files, 86+ 533-) 13https://github.com/crawl/crawl/commit/ffd75422ce3a 13:54:29 03Aidan Holm02 07* 0.25-a0-839-g2ddb0c8: Remove is_blood_potion() 10(2 days ago, 6 files, 4+ 78-) 13https://github.com/crawl/crawl/commit/2ddb0c879219 13:54:29 03Aidan Holm02 07* 0.25-a0-840-g9ecea16: Remove more old potion code 10(2 days ago, 7 files, 8+ 228-) 13https://github.com/crawl/crawl/commit/9ecea16cc44d 13:54:29 03Aidan Holm02 07* 0.25-a0-841-gfee0bf6: Remove an unused parameter 10(2 days ago, 3 files, 9+ 12-) 13https://github.com/crawl/crawl/commit/fee0bf60c558 13:57:02 New branch created: remove-old-spells (5 commits) 13https://github.com/crawl/crawl/tree/remove-old-spells 13:57:02 03Aidan Holm02 07[remove-old-spells] * 0.25-a0-842-gfc7c4b4: Extract unmarshall_vehumet_spells() function 10(3 hours ago, 3 files, 83+ 23-) 13https://github.com/crawl/crawl/commit/fc7c4b41b8f0 13:57:02 03Aidan Holm02 07[remove-old-spells] * 0.25-a0-843-gb8c398b: Add removed spells to removed_spells() list. 10(71 minutes ago, 1 file, 54+ 14-) 13https://github.com/crawl/crawl/commit/b8c398bd4b4e 13:57:02 03Aidan Holm02 07[remove-old-spells] * 0.25-a0-844-g90f2b43: Remove removed spells from Vehumet's gifts on load 10(2 hours ago, 2 files, 38+ 3-) 13https://github.com/crawl/crawl/commit/90f2b439d9f4 13:57:02 03Aidan Holm02 07[remove-old-spells] * 0.25-a0-845-gb2626da: Extract unmarshall_player_spells() function 10(42 minutes ago, 2 files, 68+ 30-) 13https://github.com/crawl/crawl/commit/b2626da5d7f9 13:57:02 03Aidan Holm02 07[remove-old-spells] * 0.25-a0-846-gbf341f2: Remove player's removed spells on load 10(21 minutes ago, 2 files, 65+ 0-) 13https://github.com/crawl/crawl/commit/bf341f24497c 13:58:33 i'm pleasantly surprised how easy it was to write tests for this 14:19:45 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-841-gfee0bf60c5 (34) 14:25:53 oh cool, making tests with test marshalled data for unmarshalling 14:31:30 the nice bit is that the reader class has a setMinorVersion, so you can write unit tests for loading saves with different versions 14:38:18 -!- amalloy_ is now known as amalloy 14:46:08 Kohinoor (L1 GnVM) ERROR in 'item-name.cc' at line 672: Removed potion type 32! (D:1) 14:46:26 Kohinoor (L1 GnVM) ERROR in 'item-name.cc' at line 672: Removed potion type 32! (D:1) 14:46:52 Kohinoor (L1 GnVM) ERROR in 'item-name.cc' at line 672: Removed potion type 32! (D:1) 14:47:33 Kohinoor (L1 GnVM) ERROR in 'item-name.cc' at line 672: Removed potion type 32! (D:1) 14:49:04 ..damn 14:49:13 !crashlog 14:49:40 21891. nano, XL24 SpAs, T:79243 (milestone): https://underhound.eu/crawl/morgue/nano/crash-nano-20200422-223146.txt 14:50:11 !crashlog Kohinoor 14:50:13 No milestones for Kohinoor (crash). 14:52:05 Kohinoor (L1 GnVM) ERROR in 'item-name.cc' at line 672: Removed potion type 32! (D:1) 14:52:18 !crashlog Kohinoor 14:52:19 No milestones for Kohinoor (crash). 14:53:14 !log Kohinoor cko 14:53:15 2. Kohinoor, XL8 GnVM, T:5755: https://crawl.kelbi.org/crawl/morgue/Kohinoor/morgue-Kohinoor-20200416-182800.txt 14:53:28 thanks 14:53:34 https://crawl.kelbi.org/crawl/morgue/Kohinoor/crash-Kohinoor-20200423-185155.txt 14:55:13 looks like it's the known items menu 15:07:25 03Aidan Holm02 07* 0.25-a0-842-g81a31cf: Fix crash on opening known-items menu 10(2 minutes ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/81a31cfd23f4 15:07:47 Potronius (L7 FoGl) ERROR in 'item-name.cc' at line 672: Removed potion type 32! (D:7) 15:08:15 Potronius (L7 FoGl) ERROR in 'item-name.cc' at line 672: Removed potion type 32! (D:7) 15:08:33 Potronius (L7 FoGl) ERROR in 'item-name.cc' at line 672: Removed potion type 32! (D:7) 15:18:38 Unstable branch on crawl.kelbi.org updated to: 0.25-a0-842-g81a31cfd23 (34) 15:24:07 The build failed. (remove-old-spells - bf341f2 #13106 : Aidan Holm): https://travis-ci.org/crawl/crawl/builds/678708078 16:57:47 zewo (L11 DrIE) ERROR in 'item-name.cc' at line 672: Removed potion type 32! (D:11) 18:24:13 Unstable branch on underhound.eu updated to: 0.25-a0-842-g81a31cfd23 (34)