IRC log for #bzflag on 20130727

01:18.34*** join/#bzflag DTRemenak (~DTRemenak@about/essy/CrazyCoder/DTRemenak)
01:18.34*** mode/#bzflag [+v DTRemenak] by ChanServ
01:58.02*** join/#bzflag Gnurdux (~gnurdux@CMU-869651.WV.CC.CMU.EDU)
01:58.36*** join/#bzflag jnoob22 (~jnoob22@c-76-25-106-67.hsd1.co.comcast.net)
01:59.11jnoob22err uh.... just got banned on mofo... thought it was freedom of speech. did not threaten anyone or do anything illegal
02:00.20jnoob22was playing as 'alcatrazz'... come on... cut a brother some slack.
02:18.34*** join/#bzflag ahs3- (~ahs3-@adsl-065-005-193-158.sip.rdu.bellsouth.net)
02:25.39khonkhortisanmofo is a bad server - people swear and get banned there
02:25.47khonkhortisanseparately or together
02:34.26allejomofo is a great server, what are you talking about? :P
02:47.58*** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535)
02:53.44I_Died_Oncehttp://pastebin.com/mMPdnXqQ
02:55.25khonkhortisanI've gotten something like that - consistently when another game was running
02:56.10I_Died_Onceno other game is running
02:56.20I_Died_Oncemine is going to be a fairly jacked up ordeal
02:56.33I_Died_Oncebrought a bunch of parts off of new egg
02:56.47I_Died_Oncemade my own pc
02:57.06I_Died_Once01:00.0 VGA compatible controller: Advanced Micro Devices [AMD] nee ATI Pitcairn [Radeon HD 7800]
02:59.34I_Died_Onceper the ATI page....
02:59.36I_Died_Oncehttp://support.amd.com/us/kbarticles/Pages/amdcatalyst13-4linreleasenotes.aspx
02:59.46I_Died_OnceAMD Radeon™ HD 7800 Series
02:59.57I_Died_Oncemy card is supported by that driver
03:00.18I_Died_Oncerunning the proprietary drivers
03:00.43I_Died_Oncethis time, its the one downloaded from ati's site
03:00.59I_Died_Oncethough when I installed the ones via apt, I got the same error
03:01.16I_Died_Oncerunning a dual monitor, separate X sessions
03:04.20I_Died_Oncesweet
03:04.22I_Died_Oncehttp://support.amd.com/us/kbarticles/Pages/AMDCatalyst1211betadriver.aspx
03:04.38I_Died_OnceFEATURE HIGHLIGHTS OF THE AMD CATALYST 12.11 BETA DRIVER:
03:04.38I_Died_OnceThis driver introduces significant performance improvements for many games across ALL 28nm AMD Radeon™ HD 7000 Series products: HD 7700, HD7800 and HD7900.
03:04.50I_Died_OnceAMD Radeon™ HD 7800 Series
03:04.55I_Died_Oncethats me
03:37.44*** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535)
03:51.30*** join/#bzflag jeffm (~jeffm@cpe-76-167-236-199.socal.res.rr.com)
03:51.31*** join/#bzflag jeffm (~jeffm@unaffiliated/jeffm2501)
03:51.31*** mode/#bzflag [+v jeffm] by ChanServ
04:12.39*** join/#bzflag Zac_ (ca59b013@gateway/web/freenode/ip.202.89.176.19)
04:15.48Zac_:D
04:15.50Zac_blast
04:16.10Zac_<PROTECTED>
05:11.12*** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535)
05:34.15*** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535)
06:13.59*** join/#bzflag Marzipan (~Marzipan@sign-4db60d78.pool.mediaWays.net)
07:29.49*** join/#bzflag Gnurdux (~gnurdux@c-67-186-57-145.hsd1.pa.comcast.net)
09:09.43*** join/#bzflag contempt (contempt@unaffiliated/contempt)
09:29.12*** join/#bzflag Constitution (~josh@bzflag/developer/Constitution)
09:29.42*** join/#bzflag Pimpinella (~frank@gondolin.pimpi.org)
10:15.59blast007I_Died_Once: run it windowed first, and then pick a resolution
10:16.09blast007bzflag -window 640x480
10:38.14*** join/#bzflag alpha1-2 (~nb_j@host203.190-230-164.telecom.net.ar)
10:40.57alpha1-2Hi, where is located the msglog.txt in windows system? In Linux it is 1 level up from the config folder.
10:41.08alpha1-2systems*
10:41.44alpha1-2I already have this, but about config file:
10:41.46alpha1-2http://wiki.bzflag.org/Config_File
10:43.55blast007think it goes in the same relative location as the linux ones
10:44.26alpha1-2ok, I will try it then thx
10:44.53alpha1-2that way*
10:45.00blast007iirc, the windows binaries for ansifilter do not work
10:45.46blast007think they were missing the mingw DLL files
10:45.54alpha1-2when did you check it? recently?
10:46.02blast007like, a day or two ago
10:46.59alpha1-2the site says that it is "actively" maintaned by its owner... let me see...
10:48.20alpha1-2well last change at 04.01.2012
10:54.24alpha1-2<PROTECTED>
10:54.35blast007on a Windows XP system
10:57.58alpha1-2and did you do it with the 32 bit tool of Windows?
10:57.58blast007antifilter.exe complains about a missing libgcc_s_dw2-1.dll, and ansifilter-gui.exe complains about a missing mingwm10.dll
10:58.09blast007and both are likely missing many more than just those
10:58.32alpha1-2it is ANSIfilter
10:58.54blast007ansifilter.exe I mean :)
10:58.58alpha1-2he he
10:59.21blast007might work if I had MinGW installed here
10:59.48blast007but I'm not going to install that, and a normal user should not have to go through all that
10:59.57alpha1-2<PROTECTED>
11:00.16blast007I'm on 32-bit Windows, if that's what you mean
11:00.37alpha1-232 bit win xp? :P
11:00.46blast007yes
11:01.08alpha1-2there isn't any 64 bit too?
11:01.09blast007very few people run 64-bit Windows XP as it was not well supported
11:01.20alpha1-2oh ok
11:01.38alpha1-2have you tried comiling the source?
11:01.43alpha1-2compiling*
11:01.46blast007no, that would require MinGW ;)
11:01.54alpha1-2oh
11:02.05alpha1-2but it is command line to
11:02.10alpha1-2too*
11:02.26blast007"it" being what?
11:02.37alpha1-2I am interesting only in command line, but I don't know if it is possible without the GUI dependency
11:02.49alpha1-2ANSIFilter
11:03.02blast007the command line didn't work, like I said
11:03.30blast007it would need the mingw DLLs
11:03.45blast007or they'd need to statically link the mingw stuff
11:04.01alpha1-2but isn't mingw GUI?
11:04.10alpha1-2isnt it*
11:04.48blast007no
11:05.03blast007"MinGW (Minimalist GNU for Windows), formerly mingw32, is a distribution of the GNU Compiler Collection (GCC) and GNU Binutils for use in compiling native Microsoft Windows applications."
11:05.55alpha1-2ohh ok, I had heard about but never tryied it :P
11:06.11blast007my guess is that they never tested their release on a clean system
11:06.20blast007it would have probably worked fine on their development system
11:06.45alpha1-2tried*
11:07.12alpha1-2you mean ANSIFilter or mingw?
11:07.19blast007ansifilter
11:07.29alpha1-2oh
11:07.30blast007the developer used mingw to build it
11:07.37alpha1-2ok
11:07.45alpha1-2well, in linux it works well
11:07.46blast007but because they had mingw on their system, it already had the DLLs installed system-wide
11:07.52blast007oh I'm sure it does :)
11:08.47alpha1-2then you couldn't compile it with other C compiler?
11:08.56alpha1-2as you said?
11:09.03alpha1-2I mean
11:09.07blast007never said anything about another compiler
11:09.15alpha1-2it would be impossible cause of some static link?
11:09.54blast007if they had statically linked their binaries, then the libraries would have been built into the .exe files
11:10.08blast007in this case, they were dynaically linked, which means external files
11:10.21alpha1-2I am tired at this time , sorry :P
11:10.23blast007I mean, really all they had to do was bundle the DLL files
11:10.40alpha1-2my thoughts are a little missed he he
11:10.40blast007assuming the license of MinGW allows that
11:11.42blast007(dynamic is better in this case because two exe files could use the same DLL files and thus save disk space)
11:11.58alpha1-2I understand
11:12.07alpha1-2I see, I mean
11:12.32alpha1-2well, it seems that windows version will have to wait :)
11:13.03alpha1-2and why dont you want to install minwg?
11:13.58alpha1-2or compiling it with other compiler?
11:14.06blast007because I don't need it crapping up my system ;)
11:14.21alpha1-2compiling?
11:14.30alpha1-2without install notinh more?
11:14.39alpha1-2:)
11:16.36blast007they don't seem to have just a .zip set of binaries... looks like it's an installer that then pulls down dozens of other .lzma packages
11:18.02blast007the 'mingw-get' package *might* let it download stuff to an arbitrary directory, but meh.
11:18.42alpha1-2wait... are you installing mingw then?
11:19.14blast007no
11:19.26blast007just complaining about it ;)
11:20.28alpha1-2what do you use for compiling BZ in windows?
11:20.38blast007Microsoft Visual Studio 2010
11:20.51alpha1-2ok, just try to compile with it?
11:20.58blast007I don't have that on this system
11:21.11alpha1-2oh
11:21.32alpha1-2ok, no problem the.... anyways thanks for the info and I will see...
11:21.57alpha1-2maybe I will delete all windows references in the post
11:22.05alpha1-2for now
11:22.13alpha1-2then*
11:22.26blast007or just leave 'em for now
11:22.59alpha1-2this post is making me crazy ha ha
11:23.01blast007maybe just add a note that the ansifilter for windows will likely need an install of mingw to get the necessary DLLs
11:23.18alpha1-2but I like doing things well :)
11:23.42alpha1-2yes, I think so
14:35.59*** join/#bzflag ccb (627612cc@gateway/web/freenode/ip.98.118.18.204)
15:40.29*** join/#bzflag unclelightning (~chatzilla@70.16.74.53)
16:02.09*** join/#bzflag ashvala (~ashvala@unaffiliated/ashvala)
17:05.20*** join/#bzflag sirquine (~quine@c-76-120-71-113.hsd1.co.comcast.net)
17:26.47*** join/#bzflag Rob5a (~Rob5a@5.28.70.57)
17:53.37*** join/#bzflag JeffM (~JeffM@12.167.61.62)
17:53.37*** join/#bzflag JeffM (~JeffM@unaffiliated/jeffm2501)
17:53.37*** mode/#bzflag [+v JeffM] by ChanServ
18:04.32blast007anyone have a good method for getting IPv6/IPv4 addresses into a MySQL database?
18:04.48JeffMbinhex them!
18:04.49blast007so far the best storage method seems to be two BIGINT columns
18:05.08blast007but then the complicated part is getting data into those from PHP
18:05.23blast007oh, and unsigned bigints
18:05.49blast007PHP's int is signed, and is only 64-bit on 64-bit builds of PHP
18:07.02blast007so I was looking at some crazy methods of getting the data into the database, like sending it in as a hex string..
18:07.42blast007seems like a lot of crazyness though...
18:08.25blast007was looking at Postgresql again, but that seems to have fun issues of its own.. like how I can't seem to use PHP's PDO with postgres's inet column
18:08.59blast007then started in with the non-PDO pgsql_* functions, but then there's fun stuff like this: https://bugs.php.net/bug.php?id=36469
18:10.33blast007(and testing the script at the end, I still duplicate the issue with a recent PHP and a somewhat Postgre 9.1..)
18:10.48blast007somewhat recent Postgres*
18:12.25*** join/#bzflag ashvala (~ashvala@unaffiliated/ashvala)
18:16.13BulletCatcherIf you just need to store and retrieve the IP addresses, a varchar type will get the job done with only moderate inefficiency.
18:16.34*** join/#bzflag Not-002 (~notifico@198.199.82.216)
18:16.34Not-002[02bzflag] 07JeffM2501 pushed 031 commit to 03master [+0/-0/±1] 13http://git.io/hO60EQ
18:16.35Not-002[02bzflag] 07JeffM2501 0304ea75b - Make developer agreement change
18:16.56blast007it'd be nice to be able to search IP ranges
18:16.59JeffMBulletCatcher, you can probalby sign the form now ;)
18:17.02JeffMhttps://docs.google.com/document/d/1XQ0CMURf5D991LbQOrUMvgAyZaDw7ugu_rAijPYF-5Q/edit?usp=sharing
18:18.16BulletCatcherIt might be nice to enable signing in a way that does not require a Google account.
18:18.40JeffMif I can figure out how to do that, sure
18:19.03BulletCatcherCan we just commit an agreement message to the (or a) repo?
18:19.24JeffMI was thinking that we make each agreement person submit an issue tracker
18:19.28blast007where does it say it requires a google account?
18:19.30JeffMan issue to the tracker
18:20.01blast007it's a google form - I don't think it needs a google account to fill it out
18:20.38blast007unless it's gonna prompt for a login after filling it out.. I'm viewing it right now, and I'm not signed into an account
18:20.49BulletCatcherI'd rather not add the complexity of involving Google in a legal agreement if we can avoid it.
18:20.52JeffMI liked the idea of using github somehow then if a name changed, it'd just update automaticly
18:21.03JeffMI can put the document in the github wiki
18:21.05JeffMthat's easy
18:21.08blast007BulletCatcher: so who would you rather involve?
18:21.21JeffMthe github wiki is a repo
18:21.46BulletCatcherIt seems to me that we ought to be able to do it with just Github.
18:21.47JeffMand we can lock the wiki that tied to the code repo down to devs only
18:22.16blast007well, whatever, doesn't affect me anyway..
18:22.23JeffMyeah you arn't signing
18:25.09blast007joevano: did you have a chance to look into fixing this?  https://github.com/BZFlag-Dev/bzflag/tree/experimental
18:25.17blast007dunno what can be done about that
18:25.49blast007(all of those should be a branch, not a subdirectory in the experimental branch)
18:26.13JeffMBulletCatcher, I can throw it here, https://github.com/BZFlag-Dev/bzflag/wiki/Developer-Agreement
18:26.19JeffMshorter URL too
18:26.21blast007same for this: https://github.com/BZFlag-Dev/bzflag/tree/release_maint
18:26.24BulletCatcherIf we put the agreement into the repo, then each developer could branch an "I agree to this" commit.
18:26.59JeffMthats a lot of branch trash to weed through, I'd rather have a single branch and each of them commit to it with there name and "I agree" as the commit message
18:27.29JeffMor have them create or comment on an issue tracker ticket, but I don't know if the ticket can be exported, the repo in a branch thing can be
18:27.41blast007and this..  https://github.com/BZFlag-Dev/bzflag/tree/summer_of_code   and this isn't even part of BZFlag: https://github.com/BZFlag-Dev/bzflag/tree/gamestats_live
18:29.03blast007and why are we trusting a github repo over a google form?  is it really that much better?
18:29.21BulletCatcherIt's just less complicated.
18:29.38blast007couldn't someone force push a blank repo?
18:29.52JeffMI don't see that making a brach is less complicated then commiting to a file
18:30.31BulletCatcherWe implicitly trust github accounts, but it is harder to map them to stuff stored at Google.
18:31.14BulletCatcherI'm not sure whether an "I agree to be bound by commit abcdef" is sufficient or if it needs to be "I agree ..." followed by the entire agreement.
18:31.20JeffMI have no issue moving to Github that was my inital idea
18:31.55JeffMBulletCatcher, the agreement is just book-keeping the new dev agreement is a "by submmiting" so legaly we don't need anything
18:32.39JeffMI juist thought it'd be nice to ensure that everyone who gets push access has read it
18:33.00JeffMso everyone knows what they are doing
18:33.11BulletCatcherTo me it just seems less complicated to do it all in github than to require a side trip to Google.
18:33.35BulletCatcherBesides, I try to minimize my interactions with Google as much as possible.
18:34.25blast007do you like Adobe better?
18:34.43BulletCatcherNope.
18:34.52blast007https://www.echosign.adobe.com/en/pricing.html  5 e-signatures a month for free ;)
18:35.21JeffMTim wanted just automatic "on submit", so we can remove the agreement
18:35.25JeffMit means no minors
18:35.40JeffMbut thats what we were at before
18:36.13blast007why even bother with any of it then?
18:36.21blast007let Tim change the wording if he wants
18:36.36JeffMhe's agreeed to the doc I have now
18:36.42JeffMall I need to do is link to it and it's done
18:37.07JeffMhe wanted to change from CLA to CA
18:37.16blast007apparently not good enough for everyone though ;)  needs to be a github repo now
18:37.22JeffMnope
18:37.27JeffMit just need to be hosted somewhere
18:37.33JeffMI'll use the git wiki
18:37.36JeffMno form
18:37.39JeffMno acceptance
18:37.48JeffMjust like it used to be
18:37.57JeffMsubmit a work, it's bound by the agreemment
18:37.58BulletCatcherIt makes sense to keep it with the repo to which commits are made.
18:38.16JeffMto be nice we should let all new submiters know that the rules have changed
18:38.37blast007BulletCatcher: not really
18:39.24JeffMthe old one was only kept with the code
18:39.42blast007most of the form info was private information - so how do you propose storing that in a git repo?  or would you just not take that information and only get a name and a "yes I agree!"
18:40.37JeffMthe form is not required
18:40.44JeffMnor is any acceptance
18:40.51JeffMthat was my idea
18:41.03blast007I'm not referring to that now - just wondering how BulletCatcher was expecting the git repo idea to work
18:42.36JeffMwell I expected us to have a branch with a single file, that had the document in it, a list of github usernames that had agreeed, and instructions that said that each user add there name to the document and commit with "I agree to the terms of the copyright assignment"
18:42.45BulletCatcherAFAIK, we haven't kept any kind of non-public documentation up to now.
18:42.47BulletCatcherI was not assuming that we would start doing it.
18:42.47JeffMthen we get a list of commits that is the list of agreeed users
18:43.00blast007BulletCatcher: uh, yeah, we have :)
18:43.03blast007COPPA letters?
18:43.26BulletCatcherOkay, but that is outside the scope of developer agreements.
18:43.54JeffMwe never had any form of acceptance ether, not that we stored
18:43.58blast007not outside the scope of "non-public documentation" though
18:45.00BulletCatcherblast007, do you think there is any info related to developer agreements that we should keep in a non-public place?
18:45.30blast007well the form did include the home address of the developer
18:45.36blast007so, at least that, no?
18:45.52blast007and for minors, that's a bit more touchy
18:45.59JeffMthe form isn't important anymore since it's assignment not license
18:46.11JeffMexcept for minors, yes
18:46.11BulletCatcherIf we were to keep that, then I would agree with keeping it non-public.
18:46.20BulletCatcher(that == home address)
18:46.45blast007so, do we start removing contributions from those under the age of 18?
18:47.02JeffMthat's up to Tim
18:47.05blast007(or whatever their local age of majority is)
18:47.07JeffMhe's the one liable
18:47.17BulletCatcherI am not ready to make that jump at this time.
18:47.18JeffMif we don't have the form I say we don't accept any new ones
18:47.50BulletCatcherYup.  We should not accept commits from minors in the future.
18:48.12JeffMI would like some form of acceptance so we can, but it's not a deal breaker
18:48.36BulletCatcherWe don't need anyone's permission to refuse commits.
18:48.42blast007maybe we'll do that for the start of a new codebase
18:48.47JeffMat this point I am sick of it, and totaly regret bringing it up, but Tim's made a choice to change and I feel I should implement his choice
18:52.53JeffMfor now I say we remove the form, keep the doc in the github wiki and go from there
18:53.06JeffMif we want a form then we make our own form on our own site at a later date
18:53.13JeffMno google, no github
18:53.43JeffMand yeah we ask about age for all new submiters
19:21.01JeffMok, so I have local changes to bzflag, how do I make a private branch? and is that branch just local or can I push it to github so I can access it on other computers?
19:29.48blast007private branch, as in, not part of the BZFlag-Dev/bzflag repo?
19:30.47JeffMwell it's going to be temporary just while I figure something out, so I didn't think it should have to go in bzflag, but maybe thats where that kinda thing goes
19:31.02blast007fork the main repo and push the branch there
19:31.11blast007push the branch to your fork, I mean
19:31.45JeffMk
19:31.52JeffMthen just kill the fork when I am done, right?
19:35.46blast007yup
19:36.26blast007you could even have BitBucket clone your github fork and then it could be a private repo ;)
19:36.33blast007er, clone the github repo
19:36.34JeffMheh
19:36.44JeffMI don't need it to be hidden
19:36.50JeffMjust not listed with bzflag
19:37.30blast007top secret battle tank simulator for the government!
19:37.49JeffMno, just the abstract classes needed make windows not need CURL
19:44.15blast007maybe I should just use MongoDB
19:44.19blast007I hear it's web scale
20:22.16Not-002[02bzflag] 07JeffM2501 pushed 031 commit to 03master [+5/-0/±2] 13http://git.io/k8herQ
20:22.17Not-002[02bzflag] 07JeffM2501 03b2fa822 - ignore IPCH dir
20:30.42JeffMhow bad would it be if bz required .net 2 on windows?
20:40.36*** join/#bzflag alpha1-2 (~nb_j@host203.190-230-164.telecom.net.ar)
21:22.50*** join/#bzflag leemar (~leemar@24.218.60.25)
21:33.21*** join/#bzflag leemar (~leemar@c-24-218-60-25.hsd1.ct.comcast.net)
21:33.46*** part/#bzflag leemar (~leemar@c-24-218-60-25.hsd1.ct.comcast.net)
22:36.30*** join/#bzflag sirquine (~quine@76.120.71.113)
22:37.43*** join/#bzflag JeffM (~JeffM@cpe-76-167-236-199.socal.res.rr.com)
22:37.44*** join/#bzflag JeffM (~JeffM@unaffiliated/jeffm2501)
22:37.44*** mode/#bzflag [+v JeffM] by ChanServ
22:41.46*** join/#bzflag ahs3- (~ahs3-@adsl-065-005-193-158.sip.rdu.bellsouth.net)
22:44.49*** join/#bzflag alpha1-2 (~nb_j@host203.190-230-164.telecom.net.ar)
22:50.02alpha1-2O.o why the big difference between thw windows log and the Linux one, talking about cache size for msglog?? (windows gave me 55 PDF pages, and Linux just 8 (in 1 test)!!)
22:50.10*** join/#bzflag ashvala (~ashvala@unaffiliated/ashvala)
22:52.17alpha1-2linux txt: 29.3 KiB; windows txt: 197.8 KiB; both reaching the memory limit I think
22:52.43*** join/#bzflag randomparticle (~randompar@about/essy/snick/randomparticle)
22:52.49blast007it's probably based on the number of lines, not the actual memory size
22:53.37alpha1-2what do you mean?
22:54.11blast007I mean it probably only stores a certain number of messages
22:54.45alpha1-2but both logs are supposed to be similars in content
22:55.20alpha1-2(not the same players but the same behaviour I mean, like having the log in full size)
22:55.34alpha1-2the difference is huge!
22:56.07blast007same resolution?  same chat box size?
22:56.12alpha1-2maybe some code difference, maybe some library behaviour difference?
22:56.45blast007seems like the amount of text that fits inside the chatbox may affect the scrollback limit
22:56.59alpha1-2well I have 1024x786, it isn't too small
22:57.04alpha1-2768*
22:57.19blast007I mean, are you using the exact same resolution and chat box size on both Windows and Linux?
22:57.33alpha1-2ah chat box size too...
22:57.39alpha1-2let me think
22:58.19blast007look at maxLines and maxScrollPages in src/bzflag/ControlPanel.cxx, and look at the addMessage method there
22:58.45alpha1-2ok
22:58.48blast007maxLines is set to 30 at first, but then set somewhere else based on the number of lines of text that can fit within the box
22:59.09alpha1-2ohh that can be!
22:59.37alpha1-2I remember the first chat resolution and it was with small letters
23:00.01alpha1-2and the other player maybe never changed it
23:00.07alpha1-2me yes
23:00.12alpha1-2let me see
23:00.29blast007ah yeah, font size as well :)
23:01.16alpha1-2yeah, I think that font size is auto resized by box changes
23:01.30alpha1-2oh no, i m wrong
23:01.40alpha1-2its independent
23:01.56alpha1-2I remember changing it too :P
23:02.09alpha1-2many years ago (joke)
23:03.49alpha1-2maxLines = 30
23:04.20blast007maxLines = int(messageAreaPixels[3] / lineHeight);
23:05.02alpha1-2I am reading the   /* draw messages part right now....
23:06.04alpha1-2<PROTECTED>
23:06.06alpha1-2<PROTECTED>
23:06.07alpha1-2<PROTECTED>
23:07.35alpha1-2well,  not value for this variable
23:07.41alpha1-2stored
23:08.18blast007that just stores the number of pages
23:08.22alpha1-2am i right is thsi the final cache size?
23:08.27blast007defaults to 20
23:08.28alpha1-2this*
23:08.51alpha1-2(maxScrollPages)
23:09.01blast007so it is 20 * the maxScrollLines, which is calculated based on the number of lines (at whatever font size you have) can fit within the chat box
23:09.02alpha1-2for me 9?
23:09.04alpha1-28*
23:09.38alpha1-2ohh let me see in my config file
23:09.41blast007the scrollPages BZDB var defaults to 20
23:10.14blast007_maxScrollPages gets set to the value of the scrollPages BZDB variable
23:10.35alpha1-2set cpanelfontsize 2
23:10.56alpha1-2set radarsize 14
23:11.19alpha1-2that is all
23:13.02alpha1-2I have a screenshot right now
23:13.09alpha1-2an*
23:13.25blast007you were right the first time ;)
23:13.31alpha1-227 lines i cpunt
23:13.33alpha1-2count*
23:14.28alpha1-2so the formule is?
23:14.47alpha1-2maxScrollPages?
23:15.12blast007that's one piece..
23:17.06alpha1-2ok I have this:
23:17.15alpha1-2maxLines = 30
23:17.17alpha1-2maxScrollPages = 8 lets say
23:17.18alpha1-2ok 8* 30 is what I had more or less (I had 8,2 * 27 lines)
23:23.44alpha1-28 * 30 lines = 240 lines
23:23.53alpha1-2real value: 8.14 x 26.7 lines = 217 lines
23:24.12alpha1-2more or less similar? or I am missing something?
23:25.02alpha1-28.14 * 30 lines = 244 lines
23:25.34blast007dunno.. use your debugger :)
23:26.16alpha1-2where does enter "cpanelfontsize" in all this?
23:27.15alpha1-2fontSize = useBigFont ? 12.0f : 8.0f;
23:27.33alpha1-2<PROTECTED>
23:27.35alpha1-2<PROTECTED>
23:27.36alpha1-2<PROTECTED>
23:27.38alpha1-2<PROTECTED>
23:27.39alpha1-2<PROTECTED>
23:27.41alpha1-2<PROTECTED>
23:27.42alpha1-2<PROTECTED>
23:27.44alpha1-2<PROTECTED>
23:27.45alpha1-2<PROTECTED>
23:27.47alpha1-2<PROTECTED>
23:27.48alpha1-2<PROTECTED>
23:27.50alpha1-2<PROTECTED>
23:29.37blast007dunno
23:29.39JeffMI don't think that code is used for the control panel
23:29.48JeffMset a breakpoint and step in
23:30.24JeffMwere the input text files the same line counts before you converted to PDF?
23:31.11JeffMand IIRC the "log" on linux is stdout, on windows it's code I wrote that makes a file
23:31.11alpha1-2I find cpanelfontsize here:
23:31.17alpha1-2http://pastebin.com/KpjQtwR0
23:31.32JeffMso on linux more things may go to stderr instead of stdout, and on windows I think ANYTHING goes to the log
23:31.48JeffMif you are talking about the client
23:31.50alpha1-2but I have it set to 2
23:33.05blast007JeffM: he's using /savemsgs
23:33.07blast007afaik
23:33.20JeffMoh then I have no idea what that does :)
23:33.42JeffMI just remember having to hook up a special file to save chat on windows for donna
23:33.43blast007just dumps what's in the All tab to a file in the config directory, optionally with ANSI color codes
23:33.50JeffMahh
23:34.00JeffMthen both clients should store similar stuff
23:34.06blast007yep
23:34.06JeffMand rendering will be the same
23:34.11JeffMthe font systems are identical
23:34.28JeffMhome grown
23:34.39JeffMand horrible
23:35.01alpha1-2:P
23:35.02JeffMdid you and BulletCatcher come to an agreement to where to store the developer doc?
23:35.05alpha1-2it is nice BTW
23:35.10JeffMno it isn't
23:35.13JeffMI wrote it
23:35.20JeffMit could be so much better
23:35.22alpha1-2very useful, for all players I think
23:35.33blast007alpha1-2: he's talking about the font system, not the command
23:35.35JeffMthe FTGL one is better
23:35.53JeffMwell would be if we didn't flush state so much
23:39.18alpha1-2finally, it seems that it is something related to panel/font sizes, but I still think that it is a huge diference: 7 / 1 :P
23:40.08alpha1-2maybe some libraries difference related between Linux and Windows version?
23:40.19alpha1-2about cache size?
23:40.33JeffMthe chat panel code is all in bzflag
23:40.34alpha1-2just a ocurrence, I dont know the code
23:40.40JeffMso I don't know what 'library' you are thinking of
23:41.39alpha1-2the whole message windows (All/Chat/Server/Misc) are only inside BZFlag client?
23:41.49JeffMyes
23:41.55alpha1-2ok
23:41.57JeffMwhere else would they be?
23:42.10JeffMthere is no "libTankGame" on linux or windows
23:43.06alpha1-2well, I said I didnt know the code, and if we talk about "cache", though I know that it doen's use any file for store it, maybe some O. S. library used (for access memory) ... but just imgagining it :)
23:43.27JeffMwut?
23:43.39alpha1-2eh! dont kill me!
23:43.45JeffMthe standard C++ lib handles memory access
23:43.50alpha1-2let me imagine things!
23:43.52alpha1-2:)
23:44.13JeffMif it did not return the same thing on both OSs when they asked for the same ammount then the OSs would be broken
23:44.27JeffMa "cache" is just ram
23:44.42JeffMor whereever you choose to store it
23:44.52JeffMmy guess is you have some setting different between the 2
23:44.55alpha1-2ok, if I understand well, first: BZ handle the data; second: it sends it to system, right?
23:44.57JeffMbut use a debuger, step in and find out
23:45.01blast007hehehe... libTankGame..  ;D
23:45.01JeffMno
23:45.14JeffMwhy would it send it to the system?
23:45.19blast007we should totally make a libTankGame
23:45.23JeffMit's a list of strings
23:45.23JeffMwe display it
23:45.26JeffMheh
23:45.36JeffMwell TankGame.dll on windows
23:45.41alpha1-2:)
23:46.05alpha1-2ohhh ok, I am tired tbh :P.... my brain cant think clearely
23:46.12alpha1-2yeah , RAM
23:46.25JeffMhow much basic C++ do you know?
23:46.38alpha1-2but what I meant it is that before save it to ram BZ habndle WHAT to save :)
23:46.56JeffMyou don't "save" things to ram
23:47.03alpha1-2and that is the code that we , with blast , were discussed
23:47.07JeffMIIRC we just keep a list
23:47.09alpha1-2discussing*
23:47.38JeffMdid you verify that the input was identical? and all the settings are identical?
23:47.43blast007alpha1-2: you could probably just add a mode to output HTML logs if you wanted ;)
23:47.46alpha1-2yeah, a list in dynmic memory right (I know about :)
23:47.56blast007then you could do '/savemsgs -h'
23:48.10JeffMas oposed to 'static' memory?
23:48.14alpha1-2yes
23:48.33JeffMwhat is 'static' memory?
23:48.47alpha1-2when you use pointers, all is stored in dynamic memory
23:49.08blast007https://en.wikipedia.org/wiki/Static_random_access_memory  ;D
23:49.15blast007we store all our chat in the BIOS chip
23:49.24JeffMI think you are trying to use the terms "Stack" and "heap"
23:49.47JeffMboth are 'dynamic'
23:49.51JeffMas in they can change
23:50.42alpha1-2yeah "Stack" and "heap"
23:50.55alpha1-2not if u are using C
23:51.29JeffMyes even if you are using C ;)
23:51.36JeffMc and c++ both use them
23:51.52JeffMthe chat messages are stored in a standard std::deque, it is iterated and saved to a file
23:52.02alpha1-2when you save a simple variable it is stored as stack
23:52.03JeffMso if you get a different number of lines, then the splitting code has to be different
23:52.15JeffMsimple == local
23:52.19alpha1-2when u use dynamic structures (with pointers) it is stored as heap
23:52.24JeffMit can be a complex type
23:53.00alpha1-2like lists
23:53.01JeffMand globals are different
23:53.21JeffMso were the screen sizes and the chat box sizes identical between the 2 OSs?
23:53.22alpha1-2I mean: int a= 1;
23:53.38alpha1-2and in the other hand: lists, arrays, etc
23:53.44JeffMsure, but CMyClass a(); is also on the stack :)
23:53.54JeffMand it's a complex type
23:53.59JeffMdon't worry about it
23:54.07JeffMit's not what is at fault here
23:54.19JeffMthe deque is in the heap
23:54.21JeffMit's global
23:54.25alpha1-2I am not talking of global variables, what does it have to do woth this
23:54.27alpha1-2?
23:54.30blast007alpha1-2: use your debugger.
23:54.48alpha1-2much more simple, just C language and a simple integer variable
23:54.49JeffMalpha1-2, because the message list is a global ;)
23:54.57blast007we cannot remotely debug the code running on your computer
23:55.23blast007(well, there's probably a *way* to do that.. but...)  ;)
23:55.31JeffMthe ControlPanel class is the same between windows and linux
23:55.39JeffMI see no windows specific saving code
23:55.40alpha1-2ok ok
23:55.49JeffMand all the standard C++ it uses has to work the same
23:55.57JeffMthere is no magic
23:56.03JeffMthe input to the save code has to be different
23:56.26JeffMif they received the same chat lines, then they were split differently
23:57.51JeffMmy guess is your font size or screen size was different between the 2 with windows being set to have shorter lines, so more chat lines were broken into 2 lines rather then one
23:57.55alpha1-2ok ok... I was just wondering cause of the big difference, but thanks anyways it was an aproximation
23:58.46alpha1-2ohhhhh
23:58.55alpha1-2yeah yeah
23:59.00alpha1-2broken lines
23:59.12alpha1-2if u have 7 /1
23:59.32alpha1-2if u have 1/1 sorry
23:59.43JeffMbut seriously, fire up the debuger on windows and just step into it
23:59.44alpha1-2and by size u can have 1/ 2 lets say
23:59.58alpha1-2by broken lines u can have 1/ 4 lets say

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