00:15:08 03dolorous02 07* 0.34-a0-1528-g86320db5a4: Simplify a graffiti "happened" reason. 10(9 minutes ago, 1 file, 1+ 3-) 13https://github.com/crawl/crawl/commit/86320db5a400 00:15:08 03dolorous02 07* 0.34-a0-1529-gafc03387fa: Expand graffiti insults a bit. 10(2 minutes ago, 1 file, 12+ 2-) 13https://github.com/crawl/crawl/commit/afc03387fa21 00:24:09 <06dolorous_84348> Quick question regarding mutation interactions: if you're a demonspawn, you can get the "ignite blood" mutation that makes spilled blood sometimes erupt in flames, and since demonspawn have blood, you can also get the "spiteful blood" mutation that sometimes spawns erythrospites at the same time. Should those two mutations actually be able to coexist? 00:33:25 Unstable branch on cbro.berotato.org updated to: 0.34-a0-1527-g17997c0906 (34) 00:55:00 <04dracoomega> I don't see why your blood can't be both angry and hot. (In fact, I actually did some coding specifically to make sure that erthyrospite blood trails would burst into flame if you have ignite blood) 00:55:37 Monster database of master branch on crawl.develz.org updated to: 0.34-a0-1529-gafc03387fa 01:02:13 03dolorous02 07* 0.34-a0-1530-g2ba17d1e3b: Add another graffiti short saying. 10(3 minutes ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/2ba17d1e3bd9 01:06:14 <06dolorous_84348> I figured, but I thought I'd double-check. I must have missed the flaming blood trails bit, but it's cool, for the record. 01:36:26 03dolorous02 07* 0.34-a0-1531-g37209c7bbe: Make Xom messaging more consistent. 10(11 minutes ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/37209c7bbe29 01:49:27 04Build failed for 08master @ 37209c7b 06https://github.com/crawl/crawl/actions/runs/19781488431 01:53:33 03dolorous02 07* 0.34-a0-1532-g186dd7f27d: Add another Xom stone arch message. 10(3 minutes ago, 1 file, 2+ 0-) 13https://github.com/crawl/crawl/commit/186dd7f27d15 02:53:27 03dolorous02 07* 0.34-a0-1533-g59ad2a9c40: Fix fountain key typos. 10(65 seconds ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/59ad2a9c407b 02:56:41 03dolorous02 07* 0.34-a0-1534-g224338057b: Show graffiti on all blue/sparkling fountains. 10(20 minutes ago, 1 file, 6+ 0-) 13https://github.com/crawl/crawl/commit/224338057bfd 02:56:41 03dolorous02 07* 0.34-a0-1535-gddb11f23fc: Show graffiti on all dry fountains too. 10(14 minutes ago, 1 file, 9+ 1-) 13https://github.com/crawl/crawl/commit/ddb11f23fcdf 02:57:46 03dolorous02 07[stone_soup-0.33] * 0.33.1-11-g13c08b6794: Fix fountain key typos. 10(5 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/13c08b679474 04:31:57 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 05:05:38 Stable (0.33) branch on crawl.akrasiac.org updated to: 0.33.1-11-g13c08b6 05:38:34 Unstable branch on crawl.akrasiac.org updated to: 0.34-a0-1535-gddb11f2 (34) 08:05:31 03dolorous02 07* 0.34-a0-1536-gfe216a7524: Expand a graffiti line. 10(3 minutes ago, 1 file, 9+ 2-) 13https://github.com/crawl/crawl/commit/fe216a7524a3 08:25:52 03dolorous02 07* 0.34-a0-1537-g8b47bc0d98: Add a few more graffiti religion entries. 10(2 minutes ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/8b47bc0d9820 08:39:47 03dolorous02 07* 0.34-a0-1538-g0e53a4e0e4: Add another graffiti rumour. 10(79 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/0e53a4e0e439 08:56:11 04Build failed for 08master @ 0e53a4e0 06https://github.com/crawl/crawl/actions/runs/19785918377 09:42:51 03dolorous02 07* 0.34-a0-1539-g9d6394c9e0: Add more species-related graffiti. 10(13 minutes ago, 1 file, 16+ 0-) 13https://github.com/crawl/crawl/commit/9d6394c9e09b 10:10:41 03dolorous02 07* 0.34-a0-1540-g37c7fda3e8: Tweak a graffiti "quality". 10(7 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/37c7fda3e896 10:10:41 03dolorous02 07* 0.34-a0-1541-gcc8f42fe07: Add more unique-related graffiti. 10(2 minutes ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/cc8f42fe07c6 10:28:55 03dolorous02 07* 0.34-a0-1542-ga3d230b921: Add another graffiti section. 10(4 minutes ago, 1 file, 10+ 0-) 13https://github.com/crawl/crawl/commit/a3d230b9216d 11:15:42 <09hellmonk> Did we ever reach a decision about how to rework death ego 11:18:01 <09gammafunk> no, there were some good discussions of ideas though 11:18:20 <09gammafunk> would have to find it in the search history, it was in this channel 11:47:02 03dolorous02 07* 0.34-a0-1543-g3cc77bdd50: Sort bland names alphabetically. 10(17 minutes ago, 1 file, 2+ 2-) 13https://github.com/crawl/crawl/commit/3cc77bdd5091 11:47:02 03dolorous02 07* 0.34-a0-1544-g71f7a30ef5: Add a few more bland names. 10(11 minutes ago, 1 file, 16+ 0-) 13https://github.com/crawl/crawl/commit/71f7a30ef56e 11:50:10 <04dracoomega> @Odds So, I have finally spent a couple hours going over all the math in the piety decay removal PR and there are a couple of issues. 1) Because piety gain code is extremely opaque, a few of the changes you made to reduce piety gain for some gods has (I think) actually increased it. For instance, DID_KILL_WIZARD used to give 0.7 piety for an HD 10 wizard and now gives ~0.73 for the same thing (instead of an intended ~0.56). A lot 11:50:10 of standard kill conducts have this issue to some degree, since you've increased the piety denominator, but the numerator itself is calculated by flatly subtracting from the denominator. (ie: 4/10 is less than 6/12) 11:50:15 <04dracoomega> 2) You say you determined the amount to decrease piety gain by calculating 'real piety gained' versus 'real piety lost' while getting to 6*. But something worth considering is that real piety gain was affected by the piety gain stepdown (ie: less piety above 4 and 6 stars), but piety decay was not. Accounting for 15 starting piety, and the piety gain stepdown, it takes 85 'real piety' to hit 4* and another 90 piety to hit 6*, 11:50:15 meaning the journey from 0-6 was spent with an average real piety gain of ~82% (ie: over the course of gaining 145 real piety, it took ~177 'raw' piety.). If a ~20% decrease in piety gain averages out to the same thing (assuming a piety conversion rate of ~82%) this means that piety gain over 6* (which is common in real games, but the point your testing stopped), which have a conversion rate of ~44%, will actually gain piety considerably faster than 11:50:15 they do right now. (And even piety over 4* in general will increase to a lesser degree.) (To put it more clearly, let's assume that a player gains 100 piety and loses 20 in a given span of time). Below 4*, they have lost 20% of their total piety gain, so gaining 80 piety and losing 0 is equivalent. But above 6*, that 100 gain only gives ~44 piety but you still lose 20, meaning that piety decay eats 45% of your piety. This means that gaining 35 piety 11:50:15 but losing 0 is actually a 45% increase over current rates. 12:08:07 <04dracoomega> To be clear, I'm not entirely sure how to properly address the latter issue. You could possibly get a better equivalence by increasing the piety stepdown proportionally, though it already seems fairly high above 6* 12:36:16 <11Odds> Oops re: 1) 12:37:13 <11Odds> Good point on 2). I'm not super enthused by increasing the stepdowns, not least because different gods would want different stepdowns and that adds complexity to an already complex system. I'll have a think 12:37:50 <04dracoomega> Yeah, I'm definitely not faulting you on (1). I missed it on my first pass, too. The code for this is so much harder to read than it surely needs to be ^^; 12:38:09 <11Odds> Yeah I'll up my suspicion of the code and reread 🙂 12:38:51 <04dracoomega> (Like: "I would like to rewrite and replace this with something else." but possibly not right now >.>) 12:39:04 <11Odds> I do wonder if we should just do that 12:39:32 <11Odds> There's a lot of bells and whistles. Is it really desirable that gift gods have dramatic slowdowns in piety, for example? 12:39:42 <11Odds> (Vehumet's being a particularly silly case) 12:39:56 <04dracoomega> Maybe? (It's also worth keeping in mind that there seemed to be general agreement to make gift timeouts stop eating piety also, which is yet another rate change. In that case, we wouldn't want to keep Sif's piety gain rate the same, since that was some of the objection to it in the first place. But Oka, for instance, probably doesn't need to be faster.) 12:40:31 <11Odds> Yeah Oka is blazing fast when not in gift range... probably too fast IMO 12:40:55 <11Odds> Though I guess the distribution of "scary" monsters changes during the games in ways I wouldn't like to guess at! 12:41:24 <04dracoomega> Oka's unique piety gain mechanics to make the rate hard to determine. Anecdotally, formicids (with their very fast levelling) can be a lot slower to hit 6* than normal species 12:41:53 <04dracoomega> (Even though it's unclear that they're especially stronger at the time) 12:41:58 <04dracoomega> 'A little' 12:43:46 <11Odds> Oka's piety gain mechanics do do something I guess - punish stealthy play that doesn't kill scary things 12:43:58 <11Odds> Can't say I've ever actually thought about this incentive as a player! 12:44:29 <04dracoomega> All of this is kind of making me wish we actually logged piety info from regular games in a more thorough way that we could do bot queries on, since there's a lot of things that are somewhat hard to construct an easy local test of, but aggregate data would be very nice for. (...y'know, assuming we also had a bot, but I'm sure at some future point we will.) 12:44:39 <11Odds> Yeah 12:45:15 <11Odds> The existing data is enough to measure the intersting seeming metric of "time to champion", but basically nothing else 12:46:29 <04dracoomega> And while that is important, you do spend most of a winning game after that point. (In fact, for a lot of the expensive capstone abilities, you're unlikely to have the invo to use them even a single time pre-Champion) 12:47:38 <04dracoomega> (I can't help but imagine a more ambitious rewriting of piety gain code to be something clearer to read, after gathering actual data on piety spent, lost, and gained, in real games. That seems like definitely not a 0.34 thing, though >.>) 12:48:07 <04dracoomega> (There's no reason to slavishly replicate anything specific about the current system, but I do like knowing these things) 12:48:53 <08o____0> (I think the reason piety isn't available in milestones is it would be a mild info leak, you could query your milestone to know the number that's normally hidden) 12:49:26 <04dracoomega> No, I know this. I assumed it would only be printed on death, like vault info stuff is 12:49:41 <08o____0> oh! yeah sorry 12:50:46 <04dracoomega> I feel in the past we did something with collecting XP stats when examining removing respawns (and wanting to know how much XP they contributed?) 12:50:47 <11Odds> Feels like this would be a lot of data in the logfile? 12:51:14 <04dracoomega> Yeah, I have no idea how feasible this is or what form it would take. 12:51:20 <04dracoomega> Just "It would be nice to know." 12:51:27 <04dracoomega> Even as a temporary thing 12:52:57 <11Odds> I guess we could log piety levels hit and abilities used. That gives quite a lot of data. Piety decay is pretty simple so doesn't really need logging 12:54:02 <04dracoomega> Decay is very calculable from time, I guess 12:54:18 <04dracoomega> The randomization really doesn't matter from a game-long perspective 12:55:20 <11Odds> Yeah, so if we have the time from 3 stars to 4 stars (or whatever), and the ability uses, we can work out the gain during that time, the decay (up to a bit of randomness), and the spend 12:55:50 <11Odds> I guess we could also track the actual decay and log it cumulatively with the piety rank transition 12:56:34 <11Odds> (reached ***. since last level: gained X, decayed Y, spent Z) 12:56:50 <04dracoomega> Hmmm 12:57:41 <11Odds> For just removing decay, I'm inclined to say this is overkill, but it would certainly make us more confident the balance isn't shifting too much 12:59:27 <04dracoomega> (It is probably overkill, yes. I just wanted to also remove gift timeout piety loss, and then it might be nice to simplify some other stuff... And really, all that data might not be necessary, but 'staring at this with more concentration' certainly would be.) 13:06:10 <04dracoomega> Like, it's actually kind of unclear how much piety is lost to gifts. They take 75% of the piety you gain while you're on gift_timeout, but you are by no means guaranteed to be given another gift immediately afterward, so "What percentage of the time worshipping Sif are you actually on timeout or not?" 13:08:00 <04dracoomega> (If you have 140 piety, there is something like a 10% chance to be given a gift from Sif when you gain a raw piety point) 13:09:03 <04dracoomega> So I guess you tend to gain ~6-7 real piety before going on cooldown again (though somewhat less due to decay, at present) 13:10:18 <04dracoomega> (And then the timeout is ~43 piety) 13:12:01 <11Odds> I'm confused, where is that 10% coming from? 13:12:28 <11Odds> Oh I see the line 13:13:38 03dolorous02 07* 0.34-a0-1545-g7a4a1a89be: Fix errant newlines in oklob plant quote. 10(6 minutes ago, 1 file, 2+ 3-) 13https://github.com/crawl/crawl/commit/7a4a1a89beb0 13:13:42 <04dracoomega> (Really, in a 'new' system for this, we could randomize a gift timeout to be actually how long we wanted it to be until the next gift and not have these other piety/chance-based functions afterward, probably) 13:14:20 <11Odds> Yeah. There's no way these bells and whistles are the right thing 13:14:58 <11Odds> In that there's no way anyone has tuned this finely enough to need the number of parameters we have for piety 13:18:54 <04dracoomega> I mean, I think a bunch of the code was made with some 'vaguely reasonable-sounding' principles like "God gifts you more benefits, the more piety you have" and "Stronger monsters give more piety in a non-linear way" and the net combination of all these things ended up being tuned to something that plays vaguely reasonably over the years. But the specific implementation chosen was surely not precise at any point. And, again, I am not 13:18:55 adverse to changing the net effect. It's just that sometimes it's easy to change it by a much larger amount than expected, which would have real consequences. 13:19:35 <04dracoomega> Like, I don't think lategame Oka needs 50% more piety or whatever, for instance ^^; 13:21:33 <04dracoomega> But like, the core of the Sif book gifting code probably existed before Sif even had things to spend meaningful piety on 13:22:19 <04dracoomega> (Of course, you also gained piety by a different means entirely do. I'm sure the numbers have been nudged in various ways over the years.) 13:30:19 <11Odds> So my maths suggests the most current-system-preserving parameters would be: Fast piety decay gods: - Piety multiplied by 0.86 - Stepdowns go from (2/3, 2/3) to (0.61, 0.58) Slow decay gods: - Piety mult 0.92 - Stepdowns to (0.64, 0.62) 13:33:54 <11Odds> And that if you adjust only piety gain, matching time-to-champion, then: - Time to 4* goes up ~5% - Time from 4 to 6 goes down ~5% - Gain at champion goes up ~20% 13:52:18 <11Odds> https://docs.google.com/spreadsheets/d/1hl_GW4mYg7r5uP32kKjk0R_7mgQzt6wgzBWxDOX4gGU/edit?gid=0#gid=0 for how I'm getting those numbers. I think the best match we can do with a single stepdown number is roughly the "simplified" set of parameters in there. 13:55:28 <04dracoomega> Ooo, fancy 14:02:08 <11Odds> * @param god_is_good Whether this is a good god. * (They don't scale piety with XL in the same way...?) I do love the comments in the codebase which convey confusion I too am feeling. 14:03:06 <04dracoomega> Yes, there's a bunch of opaque stuff there 14:04:13 <04dracoomega> Still trying to wrap my head around all the parts of that chart you linked. Not quite sure what you mean by 'piety only' versus 'simplified' (It may be worth pointing out that you don't need to gain 160 piety to hit 6*, since you start at 15 upon first joining. Not sure how much effect this has or not.) 14:05:30 <11Odds> Yeah sorry it's not that well explained - "piety only" is a set of parameters (for piety gain + stepdowns) where we only change piety rates, and "simplified" is a middle ground where we change stepdowns but all to the same value 14:06:06 <11Odds> Oh right, missed that 15 piety... 14:11:51 <11Odds> So what I'm doing is: a) Fitting the current rate of piety decay relative to gain by matching how much decay we see on the way to championing a god b) Working out the gain + stepdown parameters we'd need to match that perfectly c) Calculating the rates with a couple of other sets of parameters that don't introduce even more complexity 14:37:01 <09gammafunk> a bit late, but if this data can be summarized in a way that's either a value or a comma-separated list of values (e.g. fifsk), then it's certainly possible for logfile to have it. 14:38:49 <09gammafunk> it doesn't have to be in the logfile, necessarilly. If you can get what you need by looking at a smallish sample of games, you can make a data colleciton table for the morgue file that's trunk only and possibly just removed from trunk after you get the info you need. This is how we did XP stats for removing respawns 14:39:48 <09gammafunk> for piety logging, you could also go with a milestone approach, if that's appropriate, but then you have to be careful to not leak information; the logfile is nice because there's no chance for info leak (since it happens at the end of a game) 14:40:03 <09gammafunk> also, totally unrelated, but you love to see it cpp // FIXME: Kill this abomination - it will break! template static void _marshall_as_int(writer& th, const T& t) { marshallInt(th, static_cast(t)); } 14:40:21 <11Odds> Was the comment prophetic? 14:40:25 <09gammafunk> horrifying comment to read, especially since it gave no further explanation 14:40:32 <09gammafunk> no not so far 14:40:50 <09gammafunk> it's used for a templated function to marshall key-value pairs to the save file 14:41:45 <09gammafunk> and this templated function seems...fine? 14:41:54 <09gammafunk> cpp marshallMap(th, map.feat_renames, _marshall_as_int, marshallString); 14:42:30 <09gammafunk> not exactly clear to me why this is an abomination, but the author of the comment probably knew better than I do 14:44:02 <09gammafunk> the confusing part is actually how we marshall "vault placements" to the save file 14:44:18 <09gammafunk> I think advil understands the complexity of this a bit better, since it interacts with seed stability 14:44:39 <09gammafunk> but the main thing is we're at least not marshalling any lua/lua chunks directly to the save 14:44:54 <09gammafunk> we are doing that to the des cache, but I have save compat to guarantee a reload of that with the lua upgrade 14:45:16 <09gammafunk> interestingly the des cache uses a format very similar to save files 14:45:22 <09gammafunk> or an approach, I should probably say 14:47:41 <11Odds> There's so very much complexity going into calculating a chance of a piety gain on kill which I think is basically always between 70% and 90% 14:50:03 <11Odds> (You can drive it down to ~40% if you go around killing rats at XL27) 15:03:40 <11Odds> Bleh, it's rather hard to do things like "decrease piety gain by 10%" for on-kill piety. The existing parameters don't allow this, as the level of the thing being killed and the player are all tangled up with the constants. 15:05:00 <04dracoomega> I already knew I disliked this code, but that is only being reinforced the more I look at it today ^^; 15:05:42 <09gammafunk> %git 9cb1729a330e69a6437220670243294b013ec73c 15:05:43 <04Cerebot> gammafunk * 0.21-a0-208-g9cb1729a33: An xp_by_level XP tracking table for morgues (8 years ago, 12 files, 352+ 36-) https://github.com/crawl/crawl/commit/9cb1729a330e 15:06:10 <09gammafunk> @Odds an example of adding a morgue tracking table for trunk, if that's helpful 15:06:32 <11Odds> Yeah. I’m super unkeen to introduce any more parameters, and increasingly feel the way to go is to log things then simplify then change 15:06:51 <11Odds> Thanks! 15:07:09 <09gammafunk> I think what you'll be adding would be a bit simpler than this was 15:07:32 <11Odds> (Calling it a night now, will take a look tomorrow) 15:07:39 <09gammafunk> but there are in fact a few annoying parts you'll have to write, including marshalling/unmarshalling data to saves 15:08:02 <11Odds> Yeah I’d thought that bit would be needed 15:08:36 <11Odds> But have zero idea how morgues work so this is very helpful 15:11:04 <04dracoomega> It's worth remembering that the value of piety for a given god is a sort of 'closed system', so in a way the rate of gain can also be directly counterbalanced in many cases by adjusting piety costs of things. (Which is to say that I am also broadly in favor of this being simpler, and some wrinkles in that can be compensated for in other ways.) 15:12:44 <11Odds> Hmmmm… the time to get to thresholds also matters 15:13:11 <04dracoomega> The thing is, the basic principle involved of "Kills of stronger monsters are worth more piety, and 'stronger' is defined in some relationship to your current XL" does actually make plenty of sense. Some of the problem with the code isn't that it is trying to do this, but it is doing so in a remarkably maze-like way 15:14:01 <11Odds> That effect is in practise very weak I think (for non-Oka gods) 15:14:32 <11Odds> Though probably strong enough that it stops some degenerate behaviour (for the 0% of players who understand piety) 15:15:29 <04dracoomega> I am not really sure what 'degenerate' behavior it would stop, per se, but it makes sense to me at least that low-level enemies end up being worth much less. 15:16:21 <11Odds> Like, I think this effect becomes important mostly if you save up a whole load of very low level enemies absurdly late 15:16:53 <11Odds> (And even then, they are still worth half the piety of tough enemies) 15:18:38 <11Odds> But yeah, low-level-enemies giving less piety may well be a feature a piety system should have 15:21:08 <04dracoomega> The slightly different scaling factors for different types of holiness (and the fact that good god kills don't care about your XL at all, unlike most kills by most gods) are probably pretty superfluous as a level. (Or rather, I can see things like Zin's chaotic kills being worth more in a general sense than kills of most things for most gods, but presumably mostly by a single parameter of how much they value it instead of all 15:21:08 this?) 15:22:35 <11Odds> Yeah, gotta be careful with TSO etc in that one, but I agree that flattening holiness and then tweaking each god looks right 15:27:40 <11Odds> I might have a go at creating a PR with what a revised system might look like (with various unknown parameters to be fit) 15:29:11 <09hellmonk> Voting in favor of raising piety costs, this also conveniently buffs vehumet 16:41:22 Unstable branch on underhound.eu updated to: 0.34-a0-1545-g7a4a1a89be (34) 17:34:57 03dolorous02 07* 0.34-a0-1546-g8ccaf28445: Add clarifying comment. 10(25 minutes ago, 1 file, 8+ 0-) 13https://github.com/crawl/crawl/commit/8ccaf2844552 17:34:57 03dolorous02 07* 0.34-a0-1547-g8b2951e71f: Add Makhleb to religious graffiti. 10(10 minutes ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/8b2951e71ff6 20:13:08 <08nicolae> so i was looking through nemelex stuff to get some inspiration, and i have a question. nemelex's first title, "pannier", does not appear to actually have anything to do with cards or tarot. 20:14:54 <08nicolae> so: what the heck is it about 20:15:11 <08nicolae> https://en.wiktionary.org/wiki/pannier 20:58:09 gambling 21:00:26 <06dolorous_84348> I found this definition via https://www.finedictionary.com/pannier: "In the inns of court, formerly, a servant who laid the cloths, set the salt-cellars, cut bread, waited on the gentlemen in term-time, blew the horn as a summons to dinner, and rang the bell; now, one of the domestics who wait in the hall of the inns at the time of dinner. Also pannier-man." Especially given that the next Nemelex title is "jester"? 21:00:46 I believe it's the one who collects chips off the board, derived from the pannier (in the wiktionary sense) they're traditionally collected into 21:11:20 …it might be specifically the Aussie term therefor, since I'd generally expect "croupier" for what I suggested (US usage) 21:14:41 (the panopticon disagrees fwiw) 21:16:08 suppose the finedictionary one might work if taken as "setting the table" for gamblers, but da goog doesn't seem to back that one up either 21:19:18 03dolorous02 07* 0.34-a0-1548-g2a8f48977b: Add Fedhas to religious graffiti. 10(5 minutes ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/2a8f48977b1e 21:34:19 03dolorous02 07* 0.34-a0-1549-gd2d6d5144a: Add Hepliaklqana to religious graffiti. 10(67 seconds ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/d2d6d5144a7a 21:39:42 03dolorous02 07* 0.34-a0-1550-g5e2cc7e140: Add Uskayaw to religious graffiti. 10(69 seconds ago, 1 file, 4+ 0-) 13https://github.com/crawl/crawl/commit/5e2cc7e14066 21:54:40 03dolorous02 07* 0.34-a0-1551-ge87b8e010b: Add Dithmenos to religious, etc. graffiti. 10(4 minutes ago, 1 file, 6+ 0-) 13https://github.com/crawl/crawl/commit/e87b8e010b13 23:35:24 Unstable branch on crawl.develz.org updated to: 0.34-a0-1551-ge87b8e010b (34) 23:57:56 Windows builds of master branch on crawl.develz.org updated to: 0.34-a0-1551-ge87b8e010b