IRC log for #wowace on 20110315

00:09.32*** join/#wowace kapipi (~kapipi@raichu.netlab.jp)
00:10.11Repo10big-wigs: 03Maat * r8293 Blackwing/Nefarian.lua: back to not warning for electrocute during the RP
00:28.33*** join/#wowace Next96 (Next96@121.129.140.181)
00:32.14Zhinjioso what mod is the one that gives boss strats when you type !boss?
00:32.22Zhinjiosomeone had that running yesterday, and it was awesome.
00:48.26Hjalterob boss mods
00:48.34HjalteZhinjio: ^
00:48.36Zhinjiothanks
00:51.25Zhinjiowow. that mod fails at localization
00:53.02*** join/#wowace mckenziemc (~Mark@dialup-4.246.251.240.Dial1.SanJose1.Level3.net)
00:53.46HjalteYeah, I don't think he's the greatest coder, or english speller, but it's a very handy addon.
01:10.19*** join/#wowace nebula169 (~nebula@c-75-73-235-102.hsd1.mn.comcast.net)
01:10.27Zhinjiototally
01:10.43Zhinjioespecially for people who haven't really done much cata dungeoning. (like me)
01:17.36Repo10startip: 03starlon * r966 Modules/Appearance/Appearance.lua:
01:17.37RepoBorders are no longer dealt with from this module since the Borders module was created.
01:18.26Repo10libscriptable-1-0 (experimental): 03starlon * r66 / (2 files in 2 directories): Licensing
01:19.41Repo10starvisuals-1-0: 03starlon 04v1.4.36 * r92 : Tagging as v1.4.36
01:19.42RepoOptimizations, particularly blending.
01:23.16*** join/#wowace Groktar (~gr@c-76-20-77-109.hsd1.ca.comcast.net)
01:26.15Groktari kinda got my thumb in the cheese grater
01:26.20Groktartook off some skin
01:26.23Groktarblood everywhere
01:26.34Groktaroffers you some cheese
01:26.51nebula169now you'll have a wicked scar to impress the ladies with
01:27.21Groktarhehe
01:36.40*** join/#wowace sylvanaar_ipad (~sylvanaar@64.134.178.186)
01:38.08mckenziemc~fiskerslap sylvanaar_ipad
01:38.08purlACTION slaps sylvanaar_ipad around a bit with Fisker.
01:38.43*** part/#wowace mckenziemc (~Mark@dialup-4.246.251.240.Dial1.SanJose1.Level3.net)
01:38.44sylvanaar_ipadWe need sylslap
01:52.25sylvanaaryes we do
02:00.30Ackishttp://i51.tinypic.com/zufshz.jpg
02:05.58sylvanaarhttp://taggingrobot.com/je1yx
02:07.26Repo10arl: 03pompachomp 07master * 2.2.2-224-g23b2151 Database/Mob.lua: [+1 commit] The Winterfall Totemic mob is in Winterspring not Felwood
02:10.52*** join/#wowace kandarz (~kandarz@c-67-183-240-167.hsd1.wa.comcast.net)
02:23.46WobinAckis: what are you trolling =P
02:24.35Ackisnothing
02:24.40AckisI did a pug ICC25 man last night
02:24.45Ackisthat's a result from it
02:24.58Ackiswhen I told two peopel to roll to do the quest for the legendary and their guild lost
02:25.31Wobinterrible
02:25.53Wobinsylvanaar: interesting robot... it needs more RESTful links though imo
02:27.03*** join/#wowace nebula169 (~nebula@c-75-73-235-102.hsd1.mn.comcast.net)
02:27.25Wobinhttp://www.youtube.com/watch?v=ablNcskmvL4&feature=player_embedded#at=15
02:27.37Ackisthe gm took all her members out of the raid over a level 80 quest :)
02:28.35*** join/#wowace wex_ (~wex@cm-84.208.106.46.getinternet.no)
02:28.39*** join/#wowace tem (~tardmrr@WoWUIDev/WoWI/Dongle/Tem)
02:28.39*** mode/#wowace [+o tem] by ChanServ
02:29.44Repo10startip: 03starlon * r967 Modules/UnitTooltip/UnitTooltip.lua: Bug fix
02:30.12Repo10startip: 03starlon * r968 Modules/Bars/Bars.lua: Remove whitespace in preset
02:31.01*** join/#wowace ocularis (~wex@cm-84.208.106.46.getinternet.no)
02:31.09Repo10libscriptable-1-0 (experimental): 03starlon * r67 / (5 files in 5 directories): Licensing and bug fixes
02:32.50*** join/#wowace ocularis (~wex@cm-84.208.106.46.getinternet.no)
02:33.47*** join/#wowace ocularis (~wex@cm-84.208.106.46.getinternet.no)
02:34.51Repo10startip: 03starlon 04v1.4.28-beta * r969 : Tagging as v1.4.28-beta
02:34.52RepoFixed a known bug where self:Stop() was causing errors.
02:43.20*** join/#wowace the-golem (~Darrin@99-187-124-218.lightspeed.nscrca.sbcglobal.net)
03:13.40*** join/#wowace nebula169 (~nebula@c-75-73-235-102.hsd1.mn.comcast.net)
03:15.13*** join/#wowace mckenziemc (~Mark@dialup-4.246.254.32.Dial1.SanJose1.Level3.net)
03:24.58*** join/#wowace Hjalte (~chatzilla@cpe.ge-0-2-0-835.arcnqu2.customer.tele.dk)
03:40.21Repo10startip: 03starlon * r970 StarTip.lua: Use LibCoreLite instead of the heavier LibCore.
03:43.30*** join/#wowace gix (~gix@p5B2B0CEA.dip0.t-ipconnect.de)
03:43.46*** join/#wowace Axodious (~0ki4e087@2002:6017:736a:0:9537:14dd:2f27:f7b0)
03:47.58*** join/#wowace Kivin (Kivin@dsl-69-171-138-113.acanac.net)
03:49.44KivinAnyone been able to cook up a fix to the missing micro menu in bartender4?
04:05.46*** join/#wowace tekkub (~tekkub@WoWUIDev/WoWI/Featured/Dongle/GitHub/Tekkub)
04:05.46*** mode/#wowace [+v tekkub] by ChanServ
04:09.26*** join/#wowace Computerpunk (~Computerp@79.115.140.238)
04:10.06*** join/#wowace Funkeh` (~funk@5ad4d5f9.bb.sky.com)
04:10.06*** join/#wowace Funkeh` (~funk@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh)
04:10.06*** mode/#wowace [+o Funkeh`] by ChanServ
04:12.13TinyboomWobin: that is a very disturbing video...
04:12.30WobinTinyboom: "SURPRISE!"
04:12.35Wobinonly in thailand =)
04:15.29*** join/#wowace pompy (~Mike@c-76-116-64-179.hsd1.nj.comcast.net)
04:15.46Tinyboomidd
04:20.15*** join/#wowace bluspacecow (~bluspacec@60-234-165-167.bitstream.orcon.net.nz)
04:39.11*** part/#wowace Seerah (~Ryan@adsl-226-64-216.mem.bellsouth.net)
04:43.10*** join/#wowace Thrae (~ircuser@generalmx-1-pt.tunnel.tserv13.ash1.ipv6.he.net)
04:53.11*** join/#wowace DarkAudit (~Brian@64-181-5-194.unassigned.ntelos.net)
05:05.10*** join/#wowace tigin88 (~tigin88@99.170.180.116)
05:25.15*** join/#wowace Next96 (Next96@121.129.140.181)
05:37.20*** join/#wowace Sapu (~Sapu94@24-205-239-250.dhcp.snlo.ca.charter.com)
05:39.24*** join/#wowace Zhinjio (~Geo@home.trippy.org)
06:10.25*** join/#wowace Dotted^ (~Dotted@94.191.197.195.bredband.3.dk)
06:13.50*** join/#wowace Cyrez (~Cyrez@c-67-180-232-95.hsd1.ca.comcast.net)
06:16.25*** join/#wowace Baraius (~bhuddlest@cpe-72-177-68-136.austin.res.rr.com)
06:43.02*** join/#wowace faCe| (~face@p5B3E7605.dip.t-dialin.net)
06:44.46*** join/#wowace profalbert (~profalber@80-121-65-202.adsl.highway.telekom.at)
06:47.44*** join/#wowace Skizelli (skizelli@c-67-169-162-154.hsd1.ca.comcast.net)
07:14.54*** join/#wowace vince` (vince@adsl-99-169-81-158.dsl.pltn13.sbcglobal.net)
07:15.57*** join/#wowace rikai (~rikai@cpe-72-224-115-102.maine.res.rr.com)
07:15.57*** join/#wowace rikai (~rikai@unaffiliated/rikai)
07:16.54*** join/#wowace dubf (~quassel@84.53.31.14)
07:17.18*** join/#wowace plexiglass (~plx@93-33-141-116.ip44.fastwebnet.it)
07:17.45*** join/#wowace Antiarc|Laptop (~Chris@ip70-190-192-186.ph.ph.cox.net)
07:19.30*** join/#wowace ajmitch (~ajmitch@vps.ajmitch.net.nz)
07:19.30*** join/#wowace ajmitch (~ajmitch@ubuntu/member/ajmitch)
07:20.52*** join/#wowace Aens|Superiority (~a@69-196-151-1.dsl.teksavvy.com)
07:35.51*** join/#wowace Elkano (~elkano@wpa-uds168-30.funklan.uni-saarland.de)
07:35.51*** join/#wowace Elkano (~elkano@WoWUIDev/WoWAce/Elkano)
07:35.51*** mode/#wowace [+v Elkano] by ChanServ
07:58.28*** join/#wowace Monolit (~postid08@89.112.4.92.pppoe.eltel.net)
08:07.04*** join/#wowace cncfanatics1 (~cncfanati@91.215.157.107)
08:09.37*** join/#wowace profalbert (~profalber@80-121-65-202.adsl.highway.telekom.at)
08:10.15Repo10average-combat-mana (experimental): 03myrroddin * r9 / (4 files in 2 directories): - completely changed frame code location and details
08:10.16Repo- font effect code options rewritten
08:10.17Repo- typos, bugs, yadda yadda. bar displays now, but still errors to deal with
08:10.23*** join/#wowace selckin (~selckin@unaffiliated/selckin)
08:13.01*** join/#wowace Caleb| (~caleb@fibhost-66-83-191.fibernet.hu)
08:16.47*** join/#wowace Fisker (~na@62.61.142.209.generic-hostname.arrownet.dk)
08:25.05*** join/#wowace Kalroth (~kalroth@mail1.retailplanit.com)
08:42.06*** join/#wowace harl (harl@dslb-088-076-055-229.pools.arcor-ip.net)
08:43.18*** join/#wowace Computerpunk_ (~Computerp@79.117.225.164)
08:44.30*** join/#wowace faCe| (~face@tiliacordata.informatik.uni-bremen.de)
08:45.38*** join/#wowace profalbert (~profalber@80-121-65-202.adsl.highway.telekom.at)
08:46.52*** part/#wowace profalbert (~profalber@80-121-65-202.adsl.highway.telekom.at)
08:48.11harlwuff wuff
09:01.59*** join/#wowace the-golem (~Darrin@99-187-124-218.lightspeed.nscrca.sbcglobal.net)
09:05.02*** join/#wowace SqueeG_ (~SqueeG@cpe-67-250-123-187.hvc.res.rr.com)
09:09.47*** join/#wowace Dotted (~Dotted@79.138.175.229.bredband.oister.dk)
09:12.32*** join/#wowace TNSe (~evil@198.62-97-238.bkkb.no)
09:14.41*** join/#wowace airtonix (~T_T@unaffiliated/airtonix)
09:15.03*** join/#wowace isman (~drag@mna75-2-82-67-196-34.fbx.proxad.net)
09:15.38*** join/#wowace mckenziemc1 (~Mark@dialup-4.246.248.145.Dial1.SanJose1.Level3.net)
09:18.12*** join/#wowace Mikk (~Mikk@78-73-98-74-no162.tbcn.telia.com)
09:18.12*** join/#wowace Mikk (~Mikk@WoWUIDev/mikk)
09:18.12*** mode/#wowace [+v Mikk] by ChanServ
09:19.46*** join/#wowace Megalon (~Megalon@d86-32-171-161.cust.tele2.at)
09:24.37*** join/#wowace Adirelle|work (~Adirelle@tok69-5-82-235-150-60.fbx.proxad.net)
09:24.49*** join/#wowace Nickenyfiken (~Miranda@192.121.174.5)
09:50.26*** join/#wowace Thrae (~ircuser@generalmx-1-pt.tunnel.tserv13.ash1.ipv6.he.net)
09:51.46*** join/#wowace profalbert (~profalber@80-121-65-202.adsl.highway.telekom.at)
09:52.13Archarodimwhat is this scorecardresearch.com/beacon.js on wowace.com??
09:52.29*** join/#wowace pompy (~Mike@c-76-116-64-179.hsd1.nj.comcast.net)
09:57.21TorhalArcharodim: More than likely, ads.
09:57.54Archarodimit displayed a popup to probe me about my search habits...
09:57.57TorhalArcharodim: Yup. http://www.scorecardresearch.com/home.aspx?action=cookieTest&trackid=-1&sitegroup=20&siteid=10300&languageid=1
09:59.00TorhalKind of a fucky thing to have on a developer forum/site...but whatever.
10:00.58Repo10decursive: 03Archarodim 07master * 2.6.1-11-g2ed0733 / (4 files in 2 directories): [+1 commit] - Spell link and spellID can be used as input in custom spell UI.
10:05.29*** join/#wowace airtonix (~T_T@unaffiliated/airtonix)
10:09.38*** join/#wowace airtonix (~T_T@unaffiliated/airtonix)
10:15.29*** join/#wowace plx (~plx@93-33-141-116.ip44.fastwebnet.it)
10:16.56*** join/#wowace profalbert (~profalber@80-121-45-114.adsl.highway.telekom.at)
10:22.25*** join/#wowace airtonix (~T_T@unaffiliated/airtonix)
10:26.27*** join/#wowace Vilkku (Vilkku@dsl-tkubrasgw1-fe0bfa00-122.dhcp.inet.fi)
10:34.24*** join/#wowace airtonix (~T_T@unaffiliated/airtonix)
10:46.18Repo10inline-aura: 03Adirelle 07master * v1.6-beta-9-4-g159d920 / (3 files in 1 directory): [+3 commits]
10:46.19Repo159d920: Fixed health and power thresholds.
10:46.20Repoffa1841: Recheck tokens and GUIDs on UNIT_FACTION and UNIT_TARGETABLE_CHANGED, since I suspect funny things could happen.
10:46.21Repofa577e9: Embed latest BugGrabber *tag*.
10:49.35*** join/#wowace daev` (~bleh@pool-71-188-132-215.aubnin.fios.verizon.net)
10:56.27RepoNew addon: http://www.wowace.com/addons/libcombatlogevent-1-0/. LibCombatLogEvent-1.0. Adirelle (Manager/Author). Experimental. Approved by Torhal.
11:22.09*** join/#wowace Thrae (~ircuser@generalmx-1-pt.tunnel.tserv13.ash1.ipv6.he.net)
11:27.20profalbertwhy didn't anyone think of that before?
11:32.55*** join/#wowace cncfanatics1 (~cncfanati@91.215.157.107)
11:34.11*** join/#wowace arkanes (~arkanes@python/site-packages/arkanes)
11:35.06*** join/#wowace mojosdojo- (~mojosdojo@p5491F753.dip.t-dialin.net)
11:40.37kagaroAdirelle:  there is a UNIT_DISSAPATES combat event as well thats isnt on the wiki page
11:46.18*** join/#wowace Sliker (~ponies@127.121.125.91.rb5.adsl.brightview.com)
11:49.36*** join/#wowace Lysithea (~o@h172n7-far-a12.ias.bredband.telia.com)
11:53.11vhaarrAdirelle: does that tag stuff work when I only mark releases manually through the files interface on wowace.com?
12:13.17Adirelle|workvhaarr, I hope so
12:14.04Adirelle|workvhaarr, you mean you don't "tag repository" but update the file thing ?
12:15.06Adirelle|workkagaro, update the wiki page then !
12:23.46*** join/#wowace cncfanatics1 (~cncfanati@91.215.157.107)
12:26.53*** join/#wowace tem (~tardmrr@WoWUIDev/WoWI/Dongle/Tem)
12:26.53*** mode/#wowace [+o tem] by ChanServ
12:31.42nebula169the file management web ui doesn't repackage the file or change anything scm-wise, so i don't think packager would care about it's changes
12:32.42nebula169who knows what kind of python voodo-magic it uses, though
12:33.06Repo10professionsvault: 03oscarucb 044.2 * r207 : Tagging as 4.2
12:35.09*** join/#wowace stolenlegacy (stolenlega@chello213047049072.5.graz.surfer.at)
12:42.56*** join/#wowace sylvanaar_work (~sylvanaar@tut.qvssoftware.com)
12:43.03nebula169Adirelle|work: any reason you returning a table instead of just passing the args along?
12:43.52Adirelle|worknebula169, yep
12:44.02Adirelle|workand the packager voodoo magic is open source actually
12:44.37sylvanaar_workscrew flash - seriously
12:45.16nebula169i know :p i was just being facetious
12:50.51Xinhuanare you convinced having a library handle CLEU dispatches is faster?
12:50.55Xinhuanhave you done any testing
12:50.57Xinhuanon performance
12:51.16Xinhuani thought we discussed this 2 years ago
12:51.23Xinhuanwhich was why it wasn't implemented
12:51.59Xinhuan(by faster, i mean whether its faster than having AceEvent just dispatch CLEU events)
12:52.39Adirelle|workXinhuan, I'm not sure it is really faster, however, I'm using this kind of stuff in several addons, plus, it better handles the argument
12:52.59*** join/#wowace Funkeh`` (~funk@5ad4d5d5.bb.sky.com)
12:53.01Xinhuanbetter? how
12:53.13Xinhuanby passing a table instead of a list of args is better?
12:53.26Xinhuanhow's that better
12:53.40Adirelle|workthey are named, and I'll fix all for once for the change of 4.1
12:53.59Xinhuantable lookups are way less efficient than args on the stack when passed directly
12:54.04Xinhuanby a magnitude
12:54.11Xinhuanits worse
12:55.50Adirelle|workI never said it was faster, at best it is more convenient
12:55.59Xinhuanalso having to make copies of arguments to change is not necessarily a good thing
12:56.15Xinhuansome addons like to remap things like
12:56.19XinhuansourceName
12:56.24Xinhuanif its nil, map to UNKNOWN
12:56.25Xinhuanfor eg
12:58.40Xinhuani'm suggesting that some research/timings be done because i have reasons to believe its less efficient, plus this wasn't baked right into AceEvent-3.0 directly
12:58.54Xinhuanabout 2-3 years ago
12:59.14sylvanaar_workhow efficient does it need to be?
12:59.53Xinhuanwell, why do you need a library to dispatch CLEU events then? with securecall overheads
13:00.54sylvanaar_worksounds like it makes the code cleaner
13:01.14Xinhuanyou can implement this in your own addon in about 5 lines of code
13:01.49Xinhuanif self[eventName] then self[eventName](...) end
13:02.00Xinhuanunder the CLEU handler
13:02.38nebula169it seems the thing adi is using it for is the named args as a convenience thing, which is kind of neat
13:03.33Xinhuani suppose i can agree with that
13:03.52arkanesupdating in a single spot for the 4.1 change sounds worth it all on its own to me, honestly
13:03.56nebula169kind of restrictive in place of creating your own maps, though
13:03.58Xinhuanand sylvanaar, considering CLEU events account for the majority of events in any wow playthrough
13:04.09Xinhuani'll say efficiency is damn important
13:04.12arkanesthere's going to be a lot of fallout from that change
13:04.13Xinhuanespecially in combat
13:04.47arkanesCLEU still happens less often than OnUpdate
13:04.57arkanesby a subtantial margin
13:04.58XinhuanOnUpdate isn't an event
13:05.13Xinhuanits a script handler
13:05.14arkanesI'm aware
13:05.38arkanesbut of doing something on OnUpdate isn't a performance issue, neither is doing it in CLEU
13:06.01arkanesa table lookup isn't going to kill you
13:06.14Xinhuana single one no
13:06.19Xinhuanbut if you need to use about 6 of those args
13:06.21MikkI think a combatlog library can be a good idea.. but not implemented like that
13:06.22Xinhuanits 6 lookups
13:06.31*** join/#wowace Nickenyfiken (~Miranda@192.121.174.5)
13:06.32arkanesit's a constant time slowdown
13:06.44Xinhuanits not really constant time
13:06.47Xinhuancan be proven
13:06.59arkaneshows it not constant time?
13:07.13MikkWhat could be a performance gain is adding other filters - not just on events
13:07.18Xinhuanmake a huge table and fill it up
13:07.20MikkAnd absolutely not mapping args to descriptive names
13:07.22Xinhuanand check for access times
13:07.27arkaneswe're not talking about a huge table
13:07.31Xinhuanlarger tables have a longer access time
13:07.37arkanesand thats still constant time
13:07.52arkanesbecause the tables not changing as a result of the number of CLEU updates that happen
13:07.58Xinhuanits not constant since it increases with table size
13:08.12arkanesit's constant because it doesn't scale with the facter we're increasing, which is the number of CLEU events
13:08.24arkanestable size isn't a variable in this equation
13:08.36*** join/#wowace alcaras (~chatzilla@c-24-61-214-249.hsd1.ma.comcast.net)
13:08.38Xinhuanhave you considered the fact that because the same table is being passsed every dispatch
13:08.43Xinhuanevery CLEU has to clear the table
13:08.47Xinhuanwhich is another overhead
13:08.57Mikkmmm it should at a minimum keep one table per event type
13:08.59arkanesbut again, a constant one
13:09.02*** join/#wowace faCe| (~face@tiliacordata.informatik.uni-bremen.de)
13:09.05Mikkbut still mapping to descriptive names is unnecessary overhead imnsho
13:09.13Xinhuanyes, but its significant
13:09.14Mikkyes yes arkanes constant means its good
13:09.19arkanesdoesn't mean it's good
13:09.22Mikkeven if it's 100ms * 10000 per second?
13:09.30arkanesit's not 10k per second, though
13:09.33arkanesso why make stuff up?
13:09.46Mikkillustrating the point that "it's constant" isn't an argument
13:09.49arkanesit's an easily predictable quantity
13:10.10arkanesit's constant is establishing a baseline
13:10.35arkanesit means that it's easy to know what the cost is and to measure how much it affects you
13:10.50Mikkyes
13:11.17arkanesand seriously, if doing 1, or even 10 table lookups or even entire table creations is what pushes your addon over the limit from playable to not, something else is wrong
13:11.37Xinhuanumm, if 10 addons start to use the library
13:11.45Xinhuanthen that's the same inefficiency multiplied 10x
13:11.48arkanesunmapping arguments in to a table, even a CLEU event argument, is not some horrible breach of performant programmign ethos
13:11.49Mikkyou do realize that you get a ridiculous amount of CLEU events per second in raid combat yes
13:11.54Xinhuanplus its a library - library code are meant to be efficient
13:11.58Xinhuanotherwise why use it
13:11.59arkanesMikk: it's not really that many over all
13:12.14arkanesa couple hundred tops
13:12.48Adirelle|workback
13:12.51Xinhuanyes, a couple 100 muiltiplied by say 10 addons
13:12.53arkanesXinhuan: the same reason why anyone uses any library, for convenience
13:13.09Mikkthat's not the only reason to use a library
13:13.09Adirelle|workthe table isn't the most important thing of the library though
13:13.14arkanesyou guys realize you're just making an argument against using CLEU at all for anything, right?
13:13.33Xinhuanwell, i'm mainly pointing out 2 facts - A) i believe passing a table with named arguments is far less efficient with reasons B) CLEU dispatches were discussed a few years back for AceEvent but not implemented, hence i question its efficiency
13:13.34Adirelle|workI could pass the full argument list, or remove the library for wowace and use it in my addons anyway
13:13.35arkanesif you're going to freak out because theres an extra table created you may as well simply not use the combat log
13:13.59arkanesthere is no question that it is far less efficient in absolute terms
13:14.26arkanesbut it's also far more efficient then, say, anythign that any scrolling combat text addon does
13:14.32arkanesor any death log, or recount
13:14.34Adirelle|workhowover, the full argument things is going to be sensible to 4.1 breakage more than the table
13:14.41Mikkwhy are you arguing FOR making something less efficient than it can be?
13:14.55arkanesMikk: when there's corresponding utility
13:15.12Xinhuanthere hasn't been any utility arguments apart from named arguments in a table
13:15.19arkanesand the 4.1 change
13:15.42Xinhuanwhich essentially requires a 1 line change
13:15.46nebula169i don't see why people are fretting so much
13:15.48Xinhuanjust as the library also does
13:15.50nebula169yar
13:15.50arkanesthe additional overhead from using named arguments will be a tiny fration of the actual work done in an a handler by any non trivial addon
13:15.57Xinhuanand the owners of addons still need to repackage their addon
13:16.19arkanesso arguing against it on efficiency grounds is tenuous at best
13:16.32Adirelle|workby the way, it ignores unregistered events, so there is overhead only for those ones
13:16.34Mikki'll gladly argue against anything inefficient
13:16.43Mikkbecause inefficiency has bitten us in the ass enough
13:16.48arkanesthe combat log itself generates significantly more overhead than this
13:17.08nebula169that doesn't change the fact people are just suggesting you not add more to it
13:17.08Xinhuanyes, pity nobody has written a replacement for the combat log - because it isn't replaceble
13:17.34Xinhuanloads of shit requires constants defined in the combat log
13:17.40XinhuanBlizzard_CombatLog that is
13:17.51Xinhuanso it can't really be not-loaded and replaced
13:18.25arkanesI'd be interested if someone actually ran the metrics and came up with a reasonable number for the amount of work you could expect to do per combat event without the user seeing an effect from it
13:18.35arkanesI suspect that it's far larger than you might think
13:18.50Xinhuanthat's exactly what i proposed: conduct some tests and measurements
13:19.00arkanesan absolute cost of various dispatch techniques doesn't really answer the question of whether or not they're worth a tradeoff
13:19.34Xinhuanfor the record, Recount already does its own internal CLEU dispatch
13:19.42arkanesmost stuff does
13:19.58arkanesif you use CLEU you usually need more than 1 type, and that means some kind of dispatch
13:20.08Adirelle|workrecount also handles almost *all* CLEU events
13:20.17Xinhuannope, if you only need about 3 or 4 types, an if-else is sufficient
13:20.23Xinhuanfunction dispatches are still slower
13:20.23arkanes...
13:20.30arkanesif/else is dispatch
13:20.42Xinhuandispatch implies a function call
13:20.45Xinhuana message being sent
13:20.47arkaneswhatever
13:20.52Xinhuanif-else doesn't do that
13:21.07Mikk25man raid easily reaches 500 events/sec btw
13:21.11Mikkwas digging through combatlogs
13:21.31Xinhuanyou should see the # of events on fights with a lot of adds
13:21.38arkanesI know how many it is
13:21.43Mikkyou said less than 100
13:21.47arkanesno, I said a few hundred
13:22.00arkanes"couple"
13:22.07arkanesalthough I did have "few" in mind
13:22.32vhaarrI'll just say one thing; this is exactly how DBM passes the CLEU arguments around; in a map - if that's not enough to scare anybody away from this method I don't know what is
13:22.41Xinhuanif a single dispatch is 0.1ms slower, 500 events makes that 50ms slower... that's 20 FPS
13:22.48arkanesvhaarr: why would that scare anyone away from it?
13:22.53arkanesmillions of people run DBM
13:23.02arkanesthe technique obviously works without melting peoples computers
13:23.04vhaarryeah that doesn't mean the authors know what they're doing
13:23.12arkanesno, it doesn't
13:23.18arkanesbut it also doesn't mean that they don't
13:23.20vhaarrI don't know how long you've been around
13:23.27vhaarrbut Mikk and Xinhuan have certainly seen their share of DBM shit
13:23.35arkanesI've seen plenty of shit
13:23.51arkanesbut you've got your cause and effect backwards
13:23.58arkanesif doing this is bad, then DBM is bad for doing it
13:24.11arkanesbut if it's not bad, it's not bad regardless of whether or not DBM is bad for other reasons
13:24.21Xinhuanyou got to realize, just because millions of people use an addon doesn't mean its coded well
13:24.23XinhuanGearscore
13:24.26arkanesso DBMs badness doesn't figure into it
13:24.31XinhuanKTM
13:24.36arkanesXinhuan: nobody is making the claim that it does
13:24.52arkanesXinhuan: simply that it self evidently works without crippling overhead
13:25.00arkaneswhich should be obvious, really
13:25.21Xinhuanyes but if you run 5 copies of DBM, you might feel the difference
13:25.38Xinhuan(5 addons that uses the same technique)
13:25.40arkanesrecount or skada or whatever do far more table churn with the contents of the CLEU than would be generated by this technique
13:25.58*** join/#wowace Axodious^ (~0ki4e087@2002:6017:736a:0:a131:a6d2:e632:5e3b)
13:26.06Xinhuansigh, i don't think anyone ever mentioned about table churn
13:26.14arkanesso as relative overhead, rather than absolute, it's small
13:26.41arkanespassing arguments in a table instead of on the stack is slower because of the need to create and/or clear tables
13:27.03Xinhuanits not just that, accessing the arguments is also a magnitude slower
13:27.06arkanesand unpack them on the other side, of course
13:27.08arkanesyes
13:27.22Mikkthat's "maybe" though
13:27.27Xinhuanif you need to unpack them to be faster, why even pack them in a table
13:27.35Mikkconsider the case of several addons only needing a small subset of the CLEUs
13:28.12Mikkthe alternative is for WoW to create stack variables for all addons times all CLEUs
13:28.31arkanesthere's probably a tipping point somewhere in various use cases
13:28.41arkanesI don't think I agree that most addons only use small subsets, though
13:28.52Mikkcombatlog parsers obviously use shitloads
13:28.58arkanesmost stuff at least does swings and spells which is most of them
13:29.06Mikkbut there's also quite a few addons subscribing to CLEU that only need one type of events.. or one source
13:29.08arkaneswell, perhaps 30%
13:29.12arkanesauras are the other big chunk
13:29.30Mikkor one destination
13:29.55Mikkmaybe i should say class of source/destination too
13:30.24Xinhuanyou might be overestimating CLEU use
13:30.37Mikki may
13:30.39Mikkor not
13:30.50Mikklike this nice useful addon for holy priests that my gf found
13:30.50Xinhuanwhat CLEU events do you think Omen might be interested in? make a guess
13:30.57Mikkplops up a /say when she casts a lightwell
13:31.00Mikkand when someone uses it badly
13:31.06Mikkit does this by combatlog event parsing
13:31.15arkanesthere's probably a lot of unneccesary CLEU use
13:31.32MikkOmen? umm not many
13:31.35Mikkif at all
13:31.46arkanesOmen relies pretty much entirely on threat events I'd think
13:31.51Xinhuanthat's correct, it didn't use CLEU until patch 4.0
13:32.01Mikkit used to be all but that's another story
13:32.11arkanesXinhuan: that sound backwards to me
13:32.20Mikkand now you only use a few CLEU events right?
13:32.44Mikk(curiosity: which and why?)
13:32.44Xinhuanafter patch 4.0, it now monitors for Misdirect, tricks, mirror image, hand of salv, fade (aura gain/losses)
13:32.50Xinhuanwhen any of these are detected
13:32.59*** join/#wowace Mihau (~asmith@69.73.16.202)
13:33.03Xinhuanit starts to record damage done from Src to Dst and add estimated threat to their targets
13:33.16Xinhuanwhen the aura fades, it removes the temporary threat
13:33.26arkanesyou could do that without CLEU
13:33.34Xinhuanso the temporary portion of threat gained is shown in grey
13:33.49nebula169you need the cleu for the damage for predictions
13:33.52Xinhuanfor example, a tank may have 500k threat, but 100k of that is temorary threat
13:34.02Xinhuanso 400k of that bar is brown (for warrior) and 100k is grey
13:34.13Xinhuanand that 100k will disappear after 30 seconds
13:34.16arkanesthats a lot of work
13:34.37arkanesfor what I'd call a not very important feature, but clearly there are people with different opinions about the relative tradeoff of efficency vs features ;)
13:34.40sylvanaar_workhttp://www.youtube.com/watch?v=s_HUYi9aVvI&feature=player_embedded
13:34.41Xinhuanit is, but its quite simple to track damage done, only 4 damage events
13:35.04Xinhuanthere's no threat modifiers on any dps class anymore
13:35.20*** join/#wowace Mihau_ (~asmith@69.73.16.202)
13:35.44Xinhuanits mainly an estimate, since it doesn't account for threat done by non-damage
13:36.25nebula169why not use the threat data and mirror it for the duration of the buff? in the case of misdirect/tricks
13:36.47arkanesnebula169: the threat is generated on the target of the tricks and its rolled up into their own threat
13:36.54Xinhuanthe threat data supplied gives the total value, you dont' know how much of it is temporary
13:36.58arkanesyou dont know how much of it came from someone else
13:37.02nebula169i'm aware
13:37.08nebula169take the hunter's threat
13:37.16nebula169apply it to his misdirect target
13:37.23arkanesthe hunter doesn't have any
13:37.26Xinhuanthe hunter doesn't gain any
13:37.32RepoNew addon: http://www.wowace.com/addons/average-combat-mana/. Average Combat Mana. myrroddin (Manager/Author). Experimental. Approved by Ackis.
13:38.01nebula169ah, i see the flaw in my thought process, lol
13:38.06Xinhuanalso the other reason is a Fade/Mirror Image change
13:38.22Xinhuanit used to deduct your threat value by an absolute amount (4.1 mil threat)
13:38.25Xinhuanso it went negative
13:38.32Xinhuannow it doesn't do that in 4.0
13:38.41Xinhuanit sets your threat to exactly 0
13:39.01Xinhuanso previously, omen could just add that 4.1 mil to show you your "real amount of threat"
13:39.09Xinhuanthat you have if Mirror image isn't active
13:39.19Xinhuanso mages know exactly where they will be at in threat when it expires
13:39.23*** join/#wowace kadrahil (~kad@unaffiliated/kadrahil)
13:39.25Xinhuannow you can't, it just gives you 0
13:39.44Xinhuanso you need to track threat-you-have-when-MI-went-up + damage done during MI
13:39.58Xinhuanto show the same value
13:41.42RepoNew addon: http://www.wowace.com/addons/interrupthelper/. InterruptHelper. Morsker (Manager/Author). Experimental. Approved by Ackis.
13:41.47RepoNew addon: http://www.wowace.com/addons/snitch/. Snitch. Febris_EU (Manager/Author). Experimental. Approved by Ackis.
13:42.41RepoNew addon: http://www.wowace.com/addons/ezinterrupt/. ezInterrupt. Daylesan (Manager/Author). Approved by Ackis.
13:45.01*** join/#wowace cncfanatics1 (~cncfanati@91.215.157.107)
13:49.46Repo10inline-aura: 03Adirelle 07master * v1.6-beta-9-5-gf244920 StateModules.lua: [+1 commit] I like how UnitPower and UnitPowerMax sometimes return inconsistent values.
13:50.44arkanesXinhuan: you mentioned before that table access time grew as the table size did, is that something you've actually observed and measured in wow?
13:50.55arkanesXinhuan: or were you just talking about the cost of collision chaining in general?
13:51.26Xinhuani think it was measured loooong ago when "global _G pollution" avoidance was all the rage
13:52.04Xinhuantables in Lua double in size whenever you reach half the capacity filled
13:53.05Xinhuanerr wait
13:53.07Xinhuanthat's wrong
13:53.08Xinhuanheh
13:53.20Xinhuanthe array part doubles in size whenever it reaches full
13:53.24Xinhuanthe hash part, i have no idea
13:54.23arkanesyeah normally hash table access is linear (sometimes better) in the number of collisions
13:54.36arkanesincidentally that double affects insertion, not access :P
13:56.10arkanesfor access to be measurable non-linear with the relatively poor tools we can use to profile, the lua table implementation would have to be really suboptimal or you'd have to have extremely contrived data
13:56.48*** join/#wowace prencher (~na@unaffiliated/prencher)
13:59.46Xinhuanfor all intents and purposes, looking up a hash is still significantly costly compared to variable access on the stack
14:00.27arkanesright
14:00.29Xinhuanmy TSP algorithm ran 30% faster when i converted 2 dimensionable tables a[x][y] of size N*N to a[x*N-b]
14:00.44Xinhuana[x*N-y] i mean
14:00.50arkanesthats interesting
14:01.03Xinhuanits just a table of distances between node x to y
14:01.08Xinhuanweight tables
14:01.24arkanesit means that table access was 60% of the running time in your TSP
14:01.35Xinhuanpretty much
14:01.40arkaneswhich is surprising
14:01.57Xinhuanit has to access the table data fairly frequently
14:02.24Xinhuanthe algorithm picks a random target location to go to next, based on
14:02.39arkanesand those were integer indices?
14:02.42Xinhuanhow far away the next location is away from the current one, and a pheromone trail
14:02.44Xinhuanyes integer
14:03.02arkaneswell thats even more surprising
14:03.02winkXinhuan: that's pretty common, with multidim arrays, in every language
14:03.14Xinhuannot at all wink
14:03.31XinhuanC and C-like languages implement multidim arrays as base-memory access + offset pointers
14:03.44arkanesthe most common way to implement multidimensional arrays is exactly as Xinhuan did manually
14:03.49winkXinhuan: we once sped up an AI in java by like 30% by just using a single-dim vector with offset for the playing field :P
14:04.03Xinhuanso a[x][y] is really converted to a[(x-1)*N+y] by the compiler
14:04.15winkwell, c* may be an exception then, good to know
14:04.19arkanesof course without data structure support in the runtime for that you have to do it manually
14:04.44XinhuanC and C-like languages use row major
14:04.51Xinhuanand all arrays are contiguous
14:05.28Repo10ezinterrupt: 03Daylesan * r22 core.lua:
14:05.29Repo- Fixed bug where cast alerts would not trigger on some mobs because they were considered neutral.
14:05.31Repo- Added some more text to the config menu.
14:05.38arkanesyou could use a metatable, but without a tuple-like structure to hold the coordinates it'd be messy
14:06.07Xinhuanthe locations in this case is really that each "node" is represented by an integer
14:06.15Xinhuanfrom 1-N
14:06.39arkanesXinhuan: anyway, integer access to a lua table should be very very fast and frankly I'm surprised that the pointer-chasing done to track down your value was that much more expensive than the floating point math to calculate the distances
14:07.05Xinhuanthat's because table lookups are *really expensive* in Lua, people don't get this
14:07.17arkanesthey aren't, though
14:07.21Xinhuanbut they are
14:07.22arkanesespecially not integer ones
14:07.31arkanesthey really aren't
14:07.33Xinhuani sped up the algoritm another 10% by using LOADS of temporaries
14:07.48Xinhuanlocal a = t[x]
14:07.58Xinhuanlocal c = a[y] + a[z]
14:07.59arkanesyeah thats not surprsing
14:08.05arkanesand it's also not consistent with your double access
14:08.06Xinhuanthat avoids t[x] lookups twice
14:08.06arkaneswhich is odd
14:08.42arkanes10% from optimizing into locals is consistent with my past experiences
14:09.06Xinhuanthe result of using a flat 1D table though, means that i can only have a maximum of 729 nodes
14:09.16Xinhuananymore it will exceed the 4 mb table limit imposed by Lua
14:10.51*** join/#wowace groktar (~gr@209-234-197-181.static.twtelecom.net)
14:11.18winkwtf lua..
14:12.06Xinhuansuddenly learning a lot of things about lua? ;D
14:12.53arkanesXinhuan: you shouldn't have gotten any more benefit from normalizing t[x][y] into a flat table than you did from optimizing into a local
14:13.15arkanespossibly even more savings from the local, since you're not calculating the offset
14:13.46winkXinhuan: not suddenly, gradually :)
14:13.48Xinhuannot at all, because t[x][y] is a different table
14:14.16Xinhuani mean, it was used in a loop differently in a way
14:14.28Xinhuanthat couldn't really local t[x] for any benefit
14:14.45arkanesno, I mean that in both cases you're saving a single table lookup
14:14.51arkanesyou were doing 2, and now you're doing 1
14:15.30arkaneswere you doing anything in that inner loop *except* the table lookup?
14:16.01Adirelle|workdoes anyone know what is the character limit of in-game mail bodies ?
14:16.20Xinhuani think you might need to look at the code
14:16.21Xinhuanlol
14:16.23Adirelle|work(and if SendMail is protected or requires an hardware event)
14:16.35Adirelle|workI've done it, but there is only a 500-limit on the EditBox
14:16.52Adirelle|workI'm not sure if it is a softcap or an hardcap
14:17.00arkanesare you trying to write a distributed computation system based on instant in-game mail via the g-mail talent?!
14:17.20Adirelle|workexactly, to parse the CLEU events
14:17.43arkanesthus saving the table lookup!
14:17.46arkanesexcellent idea
14:17.56Adirelle|workactually I was thinking about storing/sharing configuration the in-game mails
14:18.00Xinhuanarkanes, http://paste.wowace.com/3379/
14:18.04XinhuanSolveTSP() is the function
14:18.08*** join/#wowace tardmrr (~tardmrr@WoWUIDev/WoWI/Dongle/Tem)
14:18.08*** mode/#wowace [+o tardmrr] by ChanServ
14:18.24Xinhuanu and v are the variables used for the [x*N-y] calculation
14:18.52Xinhuanso you will often see stuff like local u = i*numNodes-j
14:18.54*** join/#wowace vrak_ (vrak@putsch.kolbu.ws)
14:19.10*** join/#wowace Brunera (lasseb@colargol.tihlde.org)
14:19.24Xinhuanits notable that i don't just have a weight table
14:19.53Xinhuani also have a pheromone table and a probability table for each edge
14:21.10MikkSendMail is somewhat special
14:21.23MikkIt seems to require a hardware event for the first send to a recipient
14:21.27MikkBut not consecutive ones
14:22.08Xinhuanits probably to prevent automated bots spamming gold mails
14:22.17Xinhuanso you still have to click the first one?
14:23.40Mikkyes
14:23.52Mikki haven't tried with different recipients i should add
14:23.58Xinhuanarkanes, the bulk of the loop is lines 372-521, about 99% of the time is spent there, but about half the time is spent in the TwoOpt() function call for each tour inside that loop
14:24.14MikkXinhuan: nuke the Foreground button already, Background only =)
14:24.19Mikkwith a processing priority slider!
14:24.33Xinhuanheh
14:24.47Xinhuanforeground's still good for anything less than 150 nodes
14:25.06Mikkyeah but i also see people going "uhh my wow hung, *alt-f4*"
14:25.07Xinhuani haven't included code a guy sent in to move nodes
14:25.19Mikk"routes suck" -> deinstall
14:25.27Xinhuanyeah except when you click on the "Foreground" button
14:25.30Xinhuana tooltip appears
14:25.35Xinhuansaying it may hang for a while
14:25.38Xinhuanyou can't miss that
14:25.44Mikkas of when? oO
14:25.53Xinhuanas of version 1.0
14:26.05Xinhuanthe tooltip is OnEnter of the button
14:26.09Mikkoh those
14:26.11Mikknoone reads those
14:26.19Xinhuanthey will when it hangs LOLOL
14:26.27Megalonthe tooltip warning was there since ever on routes i think
14:26.44Megalonmost awesome addon ever ;x
14:26.50Mikkwell i'm an avid routes user and i didn't notice the tooltip ever :P
14:27.14Xinhuanprobably because you never used foreground?
14:27.15Xinhuan:D
14:27.17Mikkoh i did
14:27.18Torhal|workMikk: I did. WHen you hover over the button
14:27.28Mikkand background
14:27.32Xinhuansee, you're just weird
14:27.34Mikkand noticed the huge difference in time for me
14:27.37Mikkbecause i use vsync
14:27.40Mikkhence my changes
14:28.05Xinhuanah, i don't uses vsync, foreground usually takes 3x-5x faster for me
14:28.19Xinhuanbut that's really CPU dependant
14:28.49Mikk3x - 5x even with my changes?
14:28.50Xinhuanya
14:28.50Mikkthen again it's tuned to be semicomfortable now
14:28.53Mikkhence why a prio slider would be Good
14:28.56Xinhuanthe changes affect both foreground/background equally
14:29.00Xinhuanso the ratio stays the same
14:29.15Xinhuan(apart from the FPS limiting one)
14:29.44Xinhuanits smoother now with the changes
14:30.02Xinhuanbefore it would be like 10 FPS during the TwoOpt phases and 20 FPS during the main TSP loop
14:30.14Xinhuanwhile its now consistent no matter what
14:30.29Xinhuanat the expense of the FPS check call every loop
14:30.52Mikkyeah but GetTime is very very cheap
14:31.33Xinhuana function call is still expensive
14:31.41Xinhuani in-lined the intersection test code for that reason
14:32.18Mikkmmmmm yes and no remember that it's an API call and not something that needs to push lots of crap on stack
14:32.25Mikkfeel free to time it
14:32.43Mikki did but i forgot the numbers now =(
14:32.53Xinhuanan API call still pushes stuff on the stack >.>
14:33.34Mikkeither way the end result is that it's actually faster
14:33.55Mikkeven though it's less punishing on FPS
14:34.06Mikk02763346
14:34.14Xinhuani haven't actually included the code someone sent in
14:34.17Xinhuanto move nodes
14:34.23Mikkgrats i just gave you a bnet authenticator number, guess my login & password hurry =)
14:34.47Mikktoo late!
14:34.50Xinhuanif there are 3 nodes that are concave, it tries to move the middle node inwards along the perpendicular bisector of the 2 outer nodes
14:35.10Xinhuanuntil it reaches the "cluster radius" limit
14:35.37Xinhuanif it is possible to move the middle node onto the actual line between the 2 outer nodes, the code even removes the middle node
14:37.47Mikkokay, tested it. on my oldish CPU it takes 1 second to execute 1 mill GetTime()
14:39.08Xinhuanthat's pretty darn fast
14:39.20Mikkincluding "if >endTime" check it's 1.07 uS
14:40.36Mikkhaving said that, calling "local function dummyFunc() return 1234 end" is actually more than 10 times faster
14:40.49Mikklua really is fast for a scripting language
14:41.36Xinhuanis that even a valid comparison
14:41.46XinhuanGetTime does involve a system call
14:41.51sylvanaar_workyou use winauth or something Mikk
14:42.00Mikkcellphone bnet auth
14:42.24sylvanaar_workah
14:43.04Xinhuanyeah cellphone auths are 8-digit
14:43.13Xinhuanphysical ones are 6-digit
14:44.51sylvanaar_workauthenticators are annoying, less so if you use winauth or some software authenticator
14:45.10sylvanaar_workhaving to key 8 digits though would be annoying
14:45.17Axodiousthey should make an auth type based on specific hardware plugged into the machine imo ;o
14:45.28Xinhuanwhy is software ones less annoying
14:45.33mckenziemc1stop trying to hack people's accounts, then you won't find them annoying :)
14:45.35sylvanaar_workwinauth lets you hotkey the entry
14:45.48Xinhuanso...?
14:45.49sylvanaar_workit feeds the digits and the enter key
14:46.06sylvanaar_workyou dont have to look at ti
14:46.07sylvanaar_workit
14:46.16Xinhuani see
14:46.23Xinhuanhow do you backup your auth?
14:46.32sylvanaar_workto a file
14:46.57sylvanaar_workits very handy
14:47.20Xinhuanwhat if you want to login somewhere else
14:47.26sylvanaar_workyou take the file with you
14:47.34Mikkon your cellphone! :D
14:47.36Xinhuanwouldn't you need to install winauth then?
14:47.42sylvanaar_workyeah
14:47.46Xinhuanisn't that more annoying?
14:47.49sylvanaar_workno
14:47.57sylvanaar_workhow many places do you play
14:47.58Mikkdepends if you play a lot somewhere else i guess
14:48.05Axodiousill just stick to my hardware auth and the extra ~10 seconds to type the code (until/if blizz makes their own software)
14:48.10Xinhuansome places like LAN shops dont' allow you to install anything
14:48.15Xinhuanhow do you deal with that
14:48.18sylvanaar_worki dont play there
14:48.22MikkAxodious: blizzard making their own software isn't going to happen - it would be fairly pointless
14:48.30Mikkbecause then the hackers would know exactly what to target
14:48.32Mikkand read keys off of
14:48.43Xinhuanthat's the point of a (physical) auth for me though - i can play anywhere
14:48.45Axodiouswell, 3rd party software still falls under anything that blizzard can/wants to do to shutdown an account
14:48.52Xinhuaneven compromised computers
14:49.14Xinhuanand i already carry a bank phsyical auth with me anyway, one more doesn't hurt my keychain!
14:49.15sylvanaar_workactually Xinhuan bliz will detect that you are at a new location and cut you off anyways
14:49.33sylvanaar_workthen you have to log into the website too
14:49.36Xinhuanwell yeah... true
14:49.54MikkAxodious: eh yeah if you want to look at it that way you shouldn't install windows either because it's 3rd party software
14:50.01Xinhuanwell they didn't used to do that until recently
14:50.24Axodioussomething like winauth in my opinion is something blizz won't want around
14:50.31Mikkwhy ever would they not want it around
14:50.39Axodiousbecause it's blizz?
14:50.51Mikkparanoid much?
14:50.58Axodiousnot really
14:51.23Axodiouswe're talking about security measures they've implemented, wouldn't they want to keep that as closed as possible/
14:51.46Ackisclosed security is bad security
14:51.47sylvanaar_workit is a deriviative of their android app
14:53.34Mikkit's a security measure meant to protect the user.
14:53.42Mikknot a security measure meant to protect their servers or whatever
14:54.10sylvanaar_workit protects their $$, because restoring an account costs X dollars
14:54.11Xinhuanhas blizz made any indication whether it supports the use of non-blizzard auths?
14:54.21*** join/#wowace mono|strigoy (~postid08@89.112.4.92.pppoe.eltel.net)
14:54.21Mikkand the workings of number generators like this is well understood
14:54.32Mikktake a unique key, repeatedly hash it on a timer
14:54.39Mikkgive the last few digits to the user
14:55.19sylvanaar_workit is sort of like one time pad
14:55.33*** join/#wowace Sapu (~Sapu94@24-205-239-250.dhcp.snlo.ca.charter.com)
14:55.49Mikkquite so
14:56.02Mikknot quite as secure as a one time pad
14:56.05Mikkbut secure enough
14:57.20jnwhitehGoogle's 2-factor auth has one-time pads and backups to the token hash
14:57.29jnwhitehand the application-specific passwords are a really interesting look at things
15:01.53*** join/#wowace kadrahil (~kad@unaffiliated/kadrahil)
15:03.16Adirelle|workfunny thing when route (or gathermate) hangs up : move the cursor out of the button, the tooltip remains when the cilent unfreezes and stuck even if you close the windows using escape
15:04.05Fiskerok jnwhiteh
15:04.09Fiskerhow does one 2-factor auth?
15:07.49*** join/#wowace Slayman (~dejhap08@asaru.net)
15:12.45mckenziemc1how does one Fisker?
15:14.34Slaymanone apparently kills ppl and talk about it ad nauseum
15:14.54mckenziemc1i once killed you, Slayman
15:14.57mckenziemc1like that?
15:15.01Slaymanyupp
15:15.36*** join/#wowace arkanes_ (4aca4ddf@python/site-packages/arkanes)
15:24.29Elkano~emulate Fisker
15:24.29purlsomebody said emulate fisker was "<reply> OH MY GOD I AM GOING TO CUM"
15:24.38arkanes_...
15:24.44Ackislol
15:25.02*** join/#wowace stew-a (~Stewart@unafilliated/stewa/x-008753)
15:34.10*** join/#wowace mckenziemc (~Mark@dialup-4.246.250.242.Dial1.SanJose1.Level3.net)
15:44.02*** join/#wowace Funkeh` (~funk@5ad8521c.bb.sky.com)
15:44.06*** join/#wowace Funkeh` (~funk@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh)
15:44.06*** mode/#wowace [+o Funkeh`] by ChanServ
15:45.16*** join/#wowace EthanCentaurai (~ethan@92.24.184.46)
15:45.25*** join/#wowace Sapu94 (~Sapu94@24-205-239-250.dhcp.snlo.ca.charter.com)
15:49.33*** join/#wowace Sliker (~ponies@127.121.125.91.rb5.adsl.brightview.com)
15:57.18sylvanaar_work~factinfo emulate Fisket
15:57.18purlthere's no such factoid as emulate fisket, sylvanaar_work
15:57.20sylvanaar_work~factinfo emulate Fisker
15:57.20purlemulate fisker -- created by Mikma <i=mikma@reaktio.net> at Mon Dec  4 23:01:20 2006 (1561 days); last modified at Sun Oct 28 09:55:21 2007 by Kemayo!n=kemayo@pool-71-103-215-86.lsanca.dsl-w.verizon.net; it has been requested 29 times, last by Elkano, 32m 51s ago.
15:58.28sylvanaar_workKemayo: you so nasty!
15:59.30sylvanaar_work2007 seems like just yesterday
15:59.56nevcairielthe good old days
16:00.03sylvanaar_workindeed
16:00.04nevcairielman we've been at this for so long already
16:00.43sylvanaar_workyeah it has been a while hasnt it
16:01.17*** join/#wowace Motig (~Motig@dhcp-077-249-165-208.chello.nl)
16:02.18*** join/#wowace Groktar (~gr@c-76-20-77-109.hsd1.ca.comcast.net)
16:20.19*** join/#wowace Odlaw (~ozzy@2001:470:1f05:1f4:30fe:ee94:8b25:874f)
16:30.25*** join/#wowace syeren (~syeren@cpc2-salf5-2-0-cust876.10-2.cable.virginmedia.com)
16:32.04syerenDoes anyone use Foobar 2000 here?
16:32.16syerenI've installed a new soundcard, and now Foobar won't play any music -_-
16:33.08selckinput the speakers in the right hole
16:33.14selckinunmute it
16:33.21syerenEvery other media player works.
16:33.22syeren;p
16:34.11selckini've had random application pick the digital output instead of the normal one
16:34.16selckinso disable that
16:36.44*** join/#wowace ghli (~Ghli@netblock-68-183-68-27.dslextreme.com)
16:40.17*** join/#wowace wolftankk (~wolftankk@222.64.180.18)
16:40.46syerenI found out the problem, selckin.
16:40.49syerenThanks for your help ;p
16:41.45selckinwhat was it?
16:42.18syerenhttp://www.hydrogenaudio.org/forums/index.php?showtopic=76543&hl=ASUS+Xonar
16:43.43*** join/#wowace Silowyi (~silowyi_d@99-26-200-249.lightspeed.snantx.sbcglobal.net)
16:44.51*** part/#wowace bluspacecow (~bluspacec@60-234-165-167.bitstream.orcon.net.nz)
16:46.48*** join/#wowace cncfanatics1 (~cncfanati@91.181.150.83)
16:48.07*** join/#wowace Ingela (~Ingela@90-230-170-118-no35.tbcn.telia.com)
17:10.29*** join/#wowace Kalroth (~Kalroth@3e6be5cc.rev.stofanet.dk)
17:10.55*** part/#wowace profalbert (~profalber@80-121-45-114.adsl.highway.telekom.at)
17:12.25*** join/#wowace tardmrr__ (~tardmrr@74.80.58.143)
17:29.23Repo10big-wigs: 03funkydude * r8294 Plugins (10 files in 2 directories):
17:29.24RepoAdd patch by Aftermathh for beautycase skin, he says it works so if it doesn't we've got Fisker on standby.
17:43.22mojosdojo-http://www.steampowered.com/nvidia/
17:45.24*** join/#wowace Yoshimo (~Miranda@p54998F5D.dip0.t-ipconnect.de)
17:45.34*** join/#wowace TradeMark (~trademark@c-24-22-209-99.hsd1.wa.comcast.net)
18:04.30*** join/#wowace Slayman (~dejhap08@port-1613.pppoe.wtnet.de)
18:04.40*** join/#wowace Nickenyfiken (~Miranda@c83-255-103-187.bredband.comhem.se)
18:11.14*** join/#wowace Thaoky (Thaoky@81.245.18.177)
18:13.41*** join/#wowace profalbert1 (~profalber@80-121-31-254.adsl.highway.telekom.at)
18:14.39*** join/#wowace ghli (~Ghli@netblock-68-183-68-27.dslextreme.com)
18:16.24*** join/#wowace Lysithea (~o@h172n7-far-a12.ias.bredband.telia.com)
18:17.00*** join/#wowace Groktard (~gr@209-234-197-181.static.twtelecom.net)
18:19.06Sapu94question for somebody who knows a lot about the hidden addon channel...from what I can tell it's not possible for an addon on player A to send data to the same addon on player B unless it knows the name of player B or they are in the same raid / party / guild...is that correct?
18:20.04nevcairielyes
18:20.09Sapu94k thx
18:20.10nevcairielthere is no "global" addon channel
18:20.19Sapu94right that's what I thought
18:20.49Sapu94was trying to see if I could share AH scan data between multiple players...but guess it's not possible
18:21.57Sapu94actually...couldn't the addon create a chat channel and print in there? like /5 or something and just remove it from the user's chat window?
18:23.50nevcairielthats how addons usually do it
18:24.02nevcairielbut there is no really clean way to hide it from players
18:24.09nevcairielit'll usually mess with channel ordering and whatnot
18:24.21Sapu94yea right
18:24.25Sapu94mk
18:33.34Mikkwhisper comms is good
18:33.57Mikkuse AceComm btw
18:34.16Mikkand AceSerializer
18:34.20Zyncustom channels were the old way for addons to communicate, and if you enabled them in your chat ... holy mother of god, chinese walls of text
18:34.30Mikkthey deal with all the datapacking and rate limiting for you
18:34.45Mikkplus of course it was bloody annoying to have half your channels taken by addons
18:35.30*** join/#wowace cralor (~cralor@c-98-217-6-30.hsd1.ma.comcast.net)
18:36.18*** join/#wowace ghli (~Ghli@netblock-68-183-68-27.dslextreme.com)
18:37.13*** join/#wowace terra1 (~top@c-71-201-3-22.hsd1.il.comcast.net)
18:40.52quiescenso.o
18:42.32*** join/#wowace bien| (~bien@p57B66012.dip.t-dialin.net)
18:45.00*** join/#wowace Ghli_ (~Ghli@netblock-68-183-68-27.dslextreme.com)
18:46.40*** join/#wowace Funkeh` (~funk@5ad8521c.bb.sky.com)
18:46.40*** join/#wowace Funkeh` (~funk@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh)
18:46.40*** mode/#wowace [+o Funkeh`] by ChanServ
18:49.57Sapu94AceComm only deals with whisper / raid / party / guild...i'm talking about broadcasting to everybody on the server with the addon...
18:50.16Sapu94and the data is already serialized :)
18:50.30*** join/#wowace Ghli_ (~Ghli@netblock-68-183-68-27.dslextreme.com)
18:52.15*** join/#wowace Funkeh` (~funk@5ad71179.bb.sky.com)
18:52.16*** join/#wowace Funkeh` (~funk@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh)
18:52.16*** mode/#wowace [+o Funkeh`] by ChanServ
19:08.01*** join/#wowace Ketho (~Ketho@ip565063bd.direct-adsl.nl)
19:15.41*** join/#wowace nebula169` (~nebula@c-75-73-235-102.hsd1.mn.comcast.net)
19:20.28*** join/#wowace faCe| (~face@p5489BE76.dip.t-dialin.net)
19:34.10*** join/#wowace ghli (~Ghli@netblock-68-183-68-27.dslextreme.com)
19:35.33*** join/#wowace stolenlegacy_ (stolenlega@chello213047049072.5.graz.surfer.at)
19:46.50*** join/#wowace Seerah (~Ryan@adsl-226-125-136.mem.bellsouth.net)
19:47.55*** join/#wowace tekkub (~tekkub@WoWUIDev/WoWI/Featured/Dongle/GitHub/Tekkub)
19:47.55*** mode/#wowace [+v tekkub] by ChanServ
19:58.36Repo10inline-aura: 03Adirelle 07master * v1.6-beta-9-6-g6a39726 .pkgmeta: [+1 commit] Revert "Embed latest BugGrabber *tag*." as the packager does not properly handle files tagged as release using the web UI (instead of SCM tags).
20:14.56*** join/#wowace ShadniX_ (shadow1st@b045.apm.etc.tu-bs.de)
20:15.31Gnarfoz<Sapu94> was trying to see if I could share AH scan data between multiple players...but guess it's not possible <-- holy fuck would that be a lot of spam, why would you even want to create such an abombination?
20:16.11Repo10big-wigs: 03funkydude * r8295 Blackwing/Nefarian.lua: Blackwing/Nefarian: bugfix shadowblaze warning
20:17.36mckenziemcbecause Gnarfoz
20:23.38*** join/#wowace mojosdojo| (~mojosdojo@p4FE61CAE.dip.t-dialin.net)
20:23.45Repo10big-wigs: 03funkydude * r8296 Blackwing/Nefarian.lua: Blackwing/Nefarian: tweak message
20:29.07*** join/#wowace profalbert (~profalber@91-113-13-146.adsl.highway.telekom.at)
20:42.13*** join/#wowace mitchnull (~opera@catv3EC944A1.pool.t-online.hu)
20:43.54nevcairielvhaarr, Funkeh`: range checker issue, i was standing right next to someone, and he didnt show up in the checker (custom one, not boss toggled)
20:46.49nevcairielthis was at nefarian hc, in case the mind control maybe confuses it, or something? dunno how its implemented :)
20:47.50quiescensmagic!
20:49.11Funkeh`nevcairiel, was he in a group outwith the raid limit? e.g. group 3 for 10man
20:49.22nevcairielno
20:49.23mckenziemc~brand quiescens boot
20:49.23purlACTION grabs a branding iron & stamps a big boot on quiescens's forehead
20:49.31nevcairielgroup 5 in 25man
20:49.39quiescensdies.
20:50.32Funkeh`i assume it was only for this specific person nevcairiel
20:50.44nevcairieldidnt check everyone, biut yes
20:52.56nevcairielnoticed it in the middle of a fight, not much time to test stuff :D
20:53.37Funkeh`nevcairiel, you're a resto druid right?
20:53.42nevcairielyea
20:53.57quiescensdoes nefarian do stuff like back in bwl?
20:54.10nevcairielwell he does summon some silly skeletal adds
20:54.19nevcairieland he has a shadowflame breath
20:54.21nevcairielother then that
20:54.23nevcairielnot really
20:54.26quiescensthe last time i fought nefarian he turned me into a cat against my will
20:55.01quiescensalso everyone else spamming heals on the tank because the priests were all mindless and stacked dots on the tank
20:55.07Pneumatusold nef > new nef :<
20:55.15nevcairielclassic > new content
20:55.36Megalonback in the days everything was better
20:55.39Megalonespecially the future
20:56.05quiescenswe ran in and got a couple of attacks in on nef
20:56.13*** join/#wowace Elkano (~elkano@asapool1329.vpn.uni-saarland.de)
20:56.13*** join/#wowace Elkano (~elkano@WoWUIDev/WoWAce/Elkano)
20:56.13*** mode/#wowace [+v Elkano] by ChanServ
20:56.29PneumatusEbonroc > Ivoroc, etc
20:56.29quiescensa couple hundred damage at least!
20:56.56Megalonif only nef only had a couple of hundred hp!
20:57.12Pneumatusi suppose it could be worse
20:57.23Pneumatuswe could have to wear onyxia scale cloaks
20:57.34Megalonor hide inside the holes in the wall
20:57.37Pneumatusthey don't even work on the trash drakes shadowflame either :<
20:59.21nevcairieli still have mine
20:59.34Pneumatusi think my hunter still has one as well
20:59.45Pneumatusive still got nature res gear from doing huhuran also
20:59.50nevcairielhehe
20:59.56nevcairielthe good ol' times
21:00.01quiescensthat was so stupid
21:00.03quiescenso.o
21:00.13Pneumatusgearing for huhuran in maraudon was a bit fail
21:00.54nevcairieli dunno, at least you got to go to a bit more then the endgame content
21:01.17Repo10big-wigs: 03funkydude * r8297 Blackwing/Nefarian.lua:
21:01.18RepoBlackwing/Nefarian: reduce personal shadowblaze check to 1 second.
21:04.20Megalon:c
21:05.00Pneumatusanyone know with google apps if its possible to have multiple domains all filter into a single mailbox?
21:05.06Pneumatus*for email
21:05.36quiescensprobably
21:06.34Pneumatusi guessed it would probably be possible but i'll be damned if i can find anything in google's docs that confirm
21:06.48quiescensyou should be able to add aliases for your domain
21:07.03Yoshimo~seen Torhal
21:07.14purltorhal is currently on #wowprogramming (1d 15h 4m 23s) #wowuidev (1d 15h 4m 23s) #arl (1d 15h 4m 23s) #wowace (1d 15h 4m 23s) #curseforge (1d 15h 4m 23s). Has said a total of 45 messages. Is idling for 11h 1m 9s, last said: 'BWAH!'.
21:07.19quiescenssearch for domain alias or something like that
21:15.53Adirellevhaarr, so I've got the answer : the packager does not handle not-SCM-tagged releases
21:16.05Pneumatushmm, seems its only available in the business/educaton editions of apps
21:16.06Pneumatusthat sucks
21:26.23Torhal|workYoshimo: Hey there.
21:30.28Ackis~seen torhal
21:30.36purltorhal is currently on #wowprogramming (1d 15h 27m 45s) #wowuidev (1d 15h 27m 45s) #arl (1d 15h 27m 45s) #wowace (1d 15h 27m 45s) #curseforge (1d 15h 27m 45s). Has said a total of 45 messages. Is idling for 11h 24m 31s, last said: 'BWAH!'.
21:30.37Ackisseen spot
21:30.42Ackisrun torhal! run spot!
21:30.54quiescensthe google apps stuff seems a bit steep really, $50 per user or whatever
21:31.50Torhal|workbodyslams Ackis into a bonfire.
21:31.56Torhal|workcackles.
21:31.59Pneumatusquiescens: you only have to pay for it if you want more than 50 acconts, or any of the business features
21:32.01Pneumatusin theory
21:32.09quiescenseven so, per user
21:32.30quiescensis probably not the usual pricing system
21:32.39quiescensand 50 is a fair bit for a per user thingy
21:33.01Pneumatusidk, $50/year/user is probably cheaper than managing the equivalent infrastructure
21:33.44*** join/#wowace profalbert (~profalber@91-113-13-146.adsl.highway.telekom.at)
21:34.02Pneumatusit seems I could make a google apps account for each of my domains then suck all the mails into a single consolidated mailbox
21:34.46quiescenscouldn't you just use it for the one domain, and then set up a forward at the others in any number of ways
21:35.12Pneumatusi want to ditch my current mail server, which is the whole point
21:35.56Pneumatusi've got 5 domains, currently each with their own pop3 account, but ideally I want the whole lot filtering into a single account that i can hook up to outlook/my phone/etc
21:36.40winkgmail makes me cry
21:37.08winkI'm happy for everyone who loves it, but I just can't work with it
21:37.22Pneumatusi hate the web interface, but for an imap service its fine :P
21:37.39winkprefers dovecot
21:38.25Pneumatuswell, ive got dovecot running at the moment with roundcube for webmail but even so, i want rid of my own setup
21:38.28Pneumatusim fed up of managing it
21:38.39winkhehe
21:38.44winkthat's part of the fun
21:43.22*** join/#wowace daev` (~bleh@pool-71-188-132-215.aubnin.fios.verizon.net)
21:58.39*** join/#wowace daev` (~bleh@pool-71-188-132-215.aubnin.fios.verizon.net)
22:00.40*** join/#wowace Ingela (~Ingela@90-230-170-118-no35.tbcn.telia.com)
22:15.42*** join/#wowace Funkeh` (~funk@5ad71179.bb.sky.com)
22:15.43*** join/#wowace Funkeh` (~funk@WoWUIDev/WoWAce/Ace3/BigWigs/funkeh)
22:15.43*** mode/#wowace [+o Funkeh`] by ChanServ
22:19.52Sapu94@gnarfoz..it wouldn't send the data over the channel...would use the hidden addon channel for that once somebody requested it
22:20.06Gnarfozthat's... not really any better
22:20.17Gnarfozwell, better, but still wtf amounts of data
22:20.20Sapu94the "hey I have some data"...and "hey X player wants that data" would be over the channel
22:20.37*** join/#wowace Motig (~Motig@dhcp-077-249-165-208.chello.nl)
22:20.39Sapu94and it would be hidden from the user unless the user specifically tried to see it
22:20.52Gnarfozhooray for 4.1 server side addon communication prefix registering ;P
22:21.08Repo10raidbuffstatus: 03danielbarron 04v4.12 * r454 : RaidBuffStatus:
22:21.09Repo- Tagged for release.
22:21.24Sapu94still doesn't help my situation though heh
22:25.20Sapu94but....just realized it would take a minute and a half to send all that data...so probably not a good idea :p
22:38.56Repo10big-wigs: 03StingerSoft * r8298 / (2 files in 2 directories): ruRU update
22:41.48Repo10shunkit: 03LaoTseu * r43 ShuckIt.lua:
22:41.49RepoWhen there is a quest starting item, listen to QUEST_LOG_UPDATE to catch when the quest is accepted or abandoned.
22:43.14*** join/#wowace Hjalte (~chatzilla@cpe.ge-0-2-0-835.arcnqu2.customer.tele.dk)
22:44.59*** join/#wowace profalbert1 (~profalber@188-23-44-142.adsl.highway.telekom.at)
22:46.07*** join/#wowace mckenziemc (~Mark@dialup-4.246.248.58.Dial1.SanJose1.Level3.net)
22:48.17*** join/#wowace daev (~bleh@pool-71-188-132-215.aubnin.fios.verizon.net)
23:04.39*** join/#wowace Grokse (~gr@209-234-197-181.static.twtelecom.net)
23:07.37*** join/#wowace Grokse (~gr@209-234-197-181.static.twtelecom.net)
23:10.43*** join/#wowace cncfanatics2 (~cncfanati@91.181.150.83)
23:20.50Caleb|@wowalert
23:20.56Caleb|!wowalert
23:21.03quiescens~
23:21.07Caleb|~wowalert
23:21.07purlhttp://launcher.worldofwarcraft.com/alert (US); http://status.wow-europe.com/en/alert (EU) [NB: URL goes to 404 unless there's an active alert]
23:21.18Caleb|thx quiescens 3/1 not that bad ratio
23:21.26quiescensyou would have gotten there eventually
23:21.32quiescensthere are only so many punctuation marks
23:21.40Caleb|idd
23:23.48mckenziemcdirects quiescens to Unicode
23:25.38mckenziemc†kill quiescens
23:26.16mckenziemc→dosomething
23:26.25Repo10cowtip_luatext: 03profalbert 07master * v0.9.1-beta-1-g31759cc / (2 files in 1 directory): [+2 commits]
23:26.27Repo31759cc: add some more templates
23:26.28Repo4e31d39: update changelog
23:29.48mckenziemcstabs Torhal with an interrobang
23:30.29*** join/#wowace Grovyr (~Jay@cpe-075-181-060-209.carolina.res.rr.com)
23:36.09*** join/#wowace Brandon__ (~rbarreiro@87.196.30.97)
23:44.38Repo10inline-aura: 03Adirelle 07master * v1.6-beta-9-7-g9db0d8e StateModules.lua: [+1 commit] Hopefully fixed health/power state modules.

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.