00:59:38 -!- indigaz26 is now known as indigaz2 03:32:38 Experimental (bcrawl) branch on underhound.eu updated to: 0.23-a0-5261-gd9800d219b 05:09:39 Unstable branch on crawl.akrasiac.org updated to: 0.34-a0-457-ge9fc4eb (34) 05:26:40 <04d​racoomega> @hellmonk Incidentally, I took a break from banging my head against spell formulas forever to look into the shield block thing we talked about in #dcss the other day (ie: whether the block limit is worth preserving). I tested myself in wizmode, with different levels of SH and block limit against anywhere between 1 and 11 attacks take in a single turn (controlling for SH at the same time, so 'buckler with 25 SH' versus 'tower shield 05:26:40 with 25 SH'; I realize the former is a bit contrived, but with things like amulets, Qazlal, mutations, and so forth all at once, it's still possible.) and my initial conclusion is that the difference between 2 blocks/turn and 4 blocks/turn was practically invisible until you had a lot of SH, and even then tended to still be modest enough that I could see an argument for simplifying it away (like, buckler versus tower at 25 SH was still only ~15% more 05:26:41 damage against being attacked 6+ times in one turn). However, removing the block limit altogether was a massive change, resulting taking something like 75% less damage than a tower shield, currently, if actually surrounded by a lot. (And '11 attacks in one turn' isn't that contrived, either. That can happen with just 2 wolf spiders and 2 redbacks.) And I feel that trying to nerf SH power to make that less absurd would result in them being far too 05:26:41 poor against single enemies. So I think the block limit does actually fairly significant work. But I could see simplifying it so that it was always something like 3 (regardless of gear) instead of caring about your base shield type (which has always been just a little weird with how many non-equipment ways to get SH that still leave one with 1/turn blocks) 05:26:42 <04d​racoomega> However, the 'turn vs. action' bit is still a little weird. Like how a fast weapon effectively results in a lot more blocks per aut than a slow weapon, even using the same shield. And I'm not really sure that is intuitive (or mechanically positive for the game). Wondered if it should just be 'blocks per 10 aut' instead and recharge probabilistically? (But then I wonder what number we choose for 'blocks/10 aut'. Realistically, 0.7 05:26:42 is the most common attack speed by the time a character has trained enough shield skill for this to matter (or marginally less, I guess, because of shield penalties?). So a kite shield at 0.7 delay is slightly over 4 blocks (and a buckler with a rapier is about 4). Of course, you block less rapidly when doing anything else, or earlier in the game, but I could see just standardizing at '4 blocks / 10 aut' if you think that could be a good idea. 05:26:56 <04d​racoomega> (Incidentally, while working on this, I realized that fsim has been bugged regarding shields since probably forever, and will often overestimate their power against fast monsters. Since it basically always alternates 'player does an attack' and 'monster does an attack' and recharges blocks between these. So any monster that is capable of getting 2 actions before the player gets a one again will have fewer blocks to deal with in 05:26:56 reality than fsim thinks. For some monsters, this can be actually pretty meaningful (ie: wolf spider can sometimes do 4 attacks per player attack, which is twice as many as a buckler could block, but fsim thinks it could block all of them). It's accurate against most normal speed things in most situations, at least. ) 05:48:20 03DracoOmega02 07* 0.34-a0-458-g2effaad7b1: Fix Sacrifice Words giving the wrong mutation 10(2 minutes ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/2effaad7b1f0 05:48:20 03DracoOmega02 07* 0.34-a0-459-g872aa19456: Remove an out-of-date part of talisman descriptions (mysticsailboat) 10(2 minutes ago, 1 file, 0+ 4-) 13https://github.com/crawl/crawl/commit/872aa194569a 05:48:20 03DracoOmega02 07* 0.34-a0-460-gac1049ab15: Fix !f inscription not working for weapons on (a) (code2828) 10(58 seconds ago, 1 file, 1+ 1-) 13https://github.com/crawl/crawl/commit/ac1049ab1539 06:26:00 <02M​onkooky> I think in general the fewer 'per action' effects the better 06:26:46 03DracoOmega02 07* 0.34-a0-461-gf09bc62600: Prevent casting Surprising Crocodile while a tree (Acrobat) 10(67 seconds ago, 1 file, 3+ 1-) 13https://github.com/crawl/crawl/commit/f09bc6260025 12:21:19 New branch created: pull/4683 (1 commit) 13https://github.com/crawl/crawl/pull/4683 12:21:20 03radio-gra02 07https://github.com/crawl/crawl/pull/4683 * 0.34-a0-462-g2a2f2a244c: make Aphotic Marionette fail on illusions 10(9 minutes ago, 2 files, 15+ 1-) 13https://github.com/crawl/crawl/commit/2a2f2a244c59 15:41:09 Unstable branch on underhound.eu updated to: 0.34-a0-461-gf09bc62600 (34) 21:08:48 <04d​racoomega> This spreadsheet finagling has taken so long that I might actually take a break from it tomorrow to tidy up item ego branch and do a few other small things. (This was supposed to be a small thing! And then it was not >.>) 22:35:19 Unstable branch on crawl.develz.org updated to: 0.34-a0-461-gf09bc62600 (34) 22:45:52 <09h​ellmonk> There are only long tasks and longer tasks 22:57:44 Windows builds of master branch on crawl.develz.org updated to: 0.34-a0-461-gf09bc62600 23:14:12 Unstable branch on cbro.berotato.org updated to: 0.34-a0-461-gf09bc62600 (34) 23:55:14 Monster database of master branch on crawl.develz.org updated to: 0.34-a0-461-gf09bc62600