Tuesday, 2010-04-06

*** tpb has joined #tp00:00
*** ChanServ sets mode: +o tpb00:00
*** ezod has quit IRC00:07
*** ezod has joined #tp00:08
*** jmtan has quit IRC00:09
* llnz works00:27
*** Erroneous has quit IRC01:02
*** verhoevenv has quit IRC01:14
*** mithro has joined #tp01:21
*** nash has joined #tp01:30
*** jnengland77 has joined #tp01:41
*** llnz has quit IRC03:35
*** llnz has joined #tp03:44
*** konr has joined #tp04:14
*** StupidIncarnate has joined #tp04:31
StupidIncarnatehey all04:31
llnzhi StupidIncarnate04:46
llnzmithro: see the request to contribute to a book on software architecture last week?04:49
*** cool_coder has joined #tp05:01
cool_coderhi05:02
cool_coderI'm a third year electrical engineering student at National University of Singapore looking forward to participate in Google summer of code 2010. I noticed that Thousand Parsec was accepted as a mentoring organization. After looking at the project ideas, I have decided to apply for the project on cloning an existing game in thousand parsec.05:04
cool_coderi would like to know what is expected from a student applicant05:05
cool_codercan some one help?05:05
cool_coderthanks05:05
cool_coder:)05:05
tansellllnz, yeah - I've been pondering it05:07
llnztansell: me too05:07
tansellI wouldn't call Thousand Parsec all that beautiful :)05:07
llnztansell: but the architecture is still the same05:08
llnzmore or less05:08
tansellwell, maybe the new protocol library05:08
llnzcool_coder: have you read this page: http://www.thousandparsec.net/wiki/Google_Summer_of_Code05:08
tpb<http://ln-s.net/1ffR> (at www.thousandparsec.net)05:08
tansellI'm not sure I have the time to write anything either05:08
cool_coderyups05:09
cool_coderi have read it05:09
cool_coderI have been mainly using Windows (Visual Studio 2010 with TFS) for gaming till now. I would like to know if you have any specific requirememnts.05:10
*** konr has quit IRC05:10
*** shenki has quit IRC05:10
llnzcool_coder: specific requirements? like what?05:11
llnzover here in the channel cool_coder...05:12
cool_coderok... :)05:12
llnzthings like your expirence you should put in your application05:12
llnzproposal that is05:12
cool_coderok ....05:12
cool_coderDo you have something like an application template05:13
cool_coderor do I decide what to put in?05:13
llnzhave a look at the "writing proposals" section of our GSoC page05:14
llnzwe will also give you feedback on your proposal directly, and you can make changes right up to the deadline05:14
cool_coderohh ok05:16
cool_coder:)05:16
*** shenki has joined #tp05:16
cool_coderi am afraid there are not much details in the writing proposals section...05:19
cool_coderor am i missing something?05:20
cool_coderanyways i will draft one asap05:20
cool_coderany feedback is appreciated05:20
cool_coderalso, do we need to put in the high level specs?05:21
cool_coderjust found the tips for writing proposals05:23
cool_coder:)05:23
StupidIncarnateFor the new things to tp04, it lists a research support. What exactly is that?05:27
StupidIncarnateI'm not sure if my vague idea is correct05:28
llnzStupidIncarnate: that is being able to research technologies, know what you know about the technologies, and know what technologies the other players have05:28
llnzso like the tech tree in Civilization, etc05:28
StupidIncarnateah ok05:29
StupidIncarnateis it still being defined?05:29
llnzsort of05:29
StupidIncarnateerrr implimented*05:29
llnzno one has started implementing it yet05:29
StupidIncarnateguess I won't deal with it then05:31
llnzthe design isn't quite finished, it's on the wiki05:31
llnzit wouldn't take much to get it finished and start implementing it05:32
llnzit's not too difficult05:32
llnztime to run05:32
* llnz wanders off05:32
llnzlater all05:32
StupidIncarnatesee yeah05:32
*** llnz has quit IRC05:32
*** StupidIncarnate has quit IRC05:37
*** StupidIncarnate has joined #tp06:23
*** shenki has quit IRC06:36
*** shenki has joined #tp06:38
*** konr has joined #tp07:01
*** jnengland77 has quit IRC07:38
StupidIncarnatehmmm, on the media support for tp04, is that just making sure to select the right images and whatnot for a specific game?08:06
tansellStupidIncarnate, it means that the server tells the client which media to use08:21
StupidIncarnatedoes it send the location of that media as well or is the client supposed to know where it is already?08:22
StupidIncarnateif we're talking about the web client08:23
StupidIncarnateactually it has to have it since there's multiple clients08:23
StupidIncarnatenevermind08:23
StupidIncarnateso just the name and whatnot then?08:24
*** null_000 has joined #tp08:32
JLPgood morning everyone08:35
null_000morning world08:35
StupidIncarnatemorning08:35
*** mmaruseacph2 has joined #tp08:36
Epyon_Anyone seen llnz?08:40
StupidIncarnatebit ago, went out08:40
Epyon_dang08:40
StupidIncarnateerr hours ago*08:40
null_000you can always check the logs for details08:40
Epyon_nash, tansell, ping?08:43
tansellEpyon_, pong?08:43
Epyon_Ah great08:43
Epyon_tansell, you can probably answer my question if you have a moment?08:43
* nash waves08:43
Epyon_:)08:44
tansellpossibly08:44
tansellI'm a bit in at out atm08:44
Epyon_I've spent a couple of hours thinking about tpserver concurrency08:44
tansellsure, what about it?08:45
Epyon_And the solutions I've come up with are quite surprising -- seems that with a proper reorganization, there's no need for any mutexes/concurrency at all08:45
tansellEpyon_, why is that surprising?08:45
Epyon_if I understand correctly, each player only needs to "communicate" with his own data, right?08:46
Epyon_And with the universe/game data only in read-only mode08:46
Epyon_Then when the turn end comes, you just update the state of the world by using the player supplied orders08:46
Epyon_Am I correct, or is there something I overlooked?08:46
tansellEpyon_, well, you need to lock the players out from making changes while the turn generation occurs08:47
tansellbut otherwise your pretty much right08:47
Epyon_Hence, if we divided the persistence properly, there'd be no write conflics... and read conflicts could be prevented by making sure we load all needed data from persistence to memore08:47
tansellhow could you have read conflicts if the data never changed?08:48
Epyon_tansell, yes, but I'd need to do that anyway :)08:48
Epyon_tansell, read on modification...08:48
Epyon_tansell, also the question if a given persistence model has concurrent read support08:49
tansellI'm not sure what you mean by concurrent read support?08:49
Epyon_for example if it would be state-full then we might have clashes on a concurrent read08:49
Epyon_but that's not much of an issue08:49
tansellEpyon_, I still don't understand the problem you are currently describing08:50
Epyon_tansell, thats okay, I still have problems with comunicating it but it seems clear now :)08:51
Epyon_proper design = no concurrent write, and concurrent read is harmless if we make sure that the persistence doesn't depend on any state (e.g. doesn't modify anything)08:52
Epyon_tansell, so now the main question -- I already prepared a network side proposal of the project -- with boost.asio at the center. Should I expand it with the asynchronous data flow project, or should I write it up as a separate project?08:54
tansellEpyon_, would have to ask llnz about that I'm afraid08:54
tansellEpyon_, I would like to see tests in tpserver-cpp personally08:55
Epyon_Both of them are tightly bound together -- for example the boost::asio asynchronous model is a lot better than the plain boost::thread method08:55
Epyon_Unit tests of blackbox tests?08:56
tansellboth08:56
Epyon_Do have something that would work as a dummy client?08:57
tansellEpyon_, chat with null_000 about that08:58
tansellhe has been thinking about using the AI to test the server08:58
null_000I was thinking of focusing on an AI one could play against (and have some fun)08:59
null_000not on a server-testing-AI (as a primary goal)08:59
null_000but a bunch of AI playing can always find some bugs08:59
null_000in a pure black box style (but there should be some code to test if everything went as planed)09:01
Epyon_tests are at best when they can realiably repeat a failure... would the AI client be determinictic?09:08
null_000yes they would or they could be made deterministic09:09
*** marcel_ has joined #tp09:09
* null_000 has to go now09:11
marcel_hi all and bye null09:11
Epyon_tansell, also, I've got that tpservercpp Lua-ruleset builder proposal floating around -- shall I finish it up?09:11
Epyon_null_000, is your ai-client going to be called null-ai? :D09:13
null_000I will only extend daneel-ai so the client will still be called daneel-ai... but I might put null-ai somewhere ^_^09:26
*** null_000 has quit IRC09:26
StupidIncarnateWhat's an example of a possible aged property that needs to be displayed by the client?09:33
*** marcel_ has quit IRC09:34
StupidIncarnateAny aged property I can think up are for instance upgrades to ships, and in that case the property would just be replaced with the newer one09:34
*** marcel_ has joined #tp09:35
*** nash has quit IRC09:45
StupidIncarnatewell, that's what I would assume anyway09:46
*** tansell-laptop has quit IRC10:11
*** mithro has quit IRC10:13
*** tansell-laptop has joined #tp10:29
*** Epyon_ is now known as epyon10:48
*** tradedaemon has joined #tp11:05
*** StupidIncarnate has quit IRC11:06
*** verhoevenv has joined #tp11:48
*** verhoevenv_ has joined #tp11:58
*** verhoevenv has quit IRC11:58
*** mithro has joined #tp12:07
*** cool_coder has quit IRC12:25
*** cool_coder has joined #tp12:27
*** mithro has quit IRC12:35
*** cool_coder has joined #tp12:53
cool_coderhi12:54
cool_coderi have made the first draft of my proposal12:55
cool_coderfor GSoc12:55
cool_coderany feedback is appreciated12:56
cool_coderi have not included any high level specs though12:56
cool_coderhttp://www.thousandparsec.net/wiki/Http://www.example.net/wiki/Summer_of_code_cool_coder_tp#About_Me12:56
tpb<http://ln-s.net/5_+-> (at www.thousandparsec.net)12:56
ezodcool_coder: i recommend putting your proposal in the gsoc app12:59
cool_coderezod: i have submitted alr :)12:59
mmaruseacph2hi13:00
mmaruseacph2i've submitted my app both on the wiki page and on the gsoc site13:00
mmaruseacph2if you have feedback or questions please contact me and i'll try to solve everything in at most 24 hours13:00
cool_coderi would like some feedback so that i can improve upon it13:00
ezodi'll be looking at everyone's proposals shortly13:01
cool_coderthanks a lot13:03
mmaruseacph2thanks ezod :)13:04
*** cool_coder has quit IRC13:12
*** jmtan has joined #tp13:31
*** mmaruseacph2 has quit IRC13:40
*** Dreamsorcerer has joined #tp13:42
*** Dreamsorcerer has left #tp13:45
*** Dreamsorcerer has joined #tp13:46
*** mithro has joined #tp14:04
*** tradedaemon has quit IRC14:28
*** jmtan has quit IRC14:42
*** jmtan has joined #tp14:48
epyonwoah, somebody stole my last years GSoC proposal for Boost :-/15:09
*** jmtan has quit IRC15:18
*** null_000 has joined #tp15:25
marcel_Tell him its not allowed15:29
epyonmarcel_, too late, it's already submitted to formal review (it was last years proposal)15:35
*** Demitar has quit IRC15:37
marcel_You cant remove proposals after submittion?15:38
*** Greywhind has quit IRC15:41
*** Greywhind has joined #tp15:42
*** null_000 has quit IRC15:48
*** null_000 has joined #tp15:58
epyonmarcel_, someone took my last years proposal, wrote a library and submitted it to boost16:08
epyonwithout any acknowledgement16:09
null_000so someone implemented your idea?16:10
epyonyeah16:15
null_000bummer...16:18
*** StupidIncarnate has joined #tp16:24
StupidIncarnateis Greywhind around?16:25
*** Demitar has joined #tp16:35
*** Erroneous has joined #tp16:50
*** mmaruseacph2 has joined #tp16:52
*** StupidIncarnate has quit IRC16:52
*** verhoevenv_ has quit IRC17:02
*** Noughmad has joined #tp17:39
*** peres has joined #tp18:23
*** tanoku has joined #tp18:29
*** isra17 has joined #tp19:01
isra17hey19:01
ezodisra17: hi19:11
*** Dreamsorcerer has left #tp19:18
isra17I'm trying to run tpserver-cpp but it kept stuck at adding a file descriptor19:19
isra17and I can't connect with any client19:21
*** mmaruseacph2 has quit IRC19:42
isra17nvm, the configuration file was not found...19:52
*** isra17 has quit IRC20:00
*** peres has quit IRC20:05
CIA-70llnz libtpproto-cpp * r1a18069f4771 / (10 files in 2 dirs):20:12
CIA-70Update to m4 macros for Boost, and extra headers.20:12
CIA-70Thanks to Marcel Hauf and JLP20:12
*** llnz has joined #tp20:30
*** konr has quit IRC20:32
llnzmorning all20:40
null_000evening llnz20:40
llnzhi null_00020:41
null_000(it's 22:40 PM here... thats the reason for evening)20:42
llnz8:44am here20:44
null_000that... new zealand right?20:46
kerbe23:50 here (if anyone is interested. :P ) almost tomorrow already. :)20:50
llnznull_000: yeah20:50
llnzits also wednesday, so you are almost at today kerbe20:51
null_000llnz: I've been there for 4 months... so I recognise the time difference20:52
llnznull_000: cool20:52
llnz:-)20:52
*** JLP has quit IRC20:52
*** JLP has joined #tp20:53
*** JLP has joined #tp20:53
null_000llnz: beautiful country... and a bit like my country actually ^_^20:53
*** ChanServ sets mode: +o llnz20:56
*** llnz changes topic to "Thousand Parsec - www.thousandparsec.net || Why not help out? - www.thousandparsec.net/wiki/Getting_started_with_development || Logs at www.thousandparsec.net/~irc/ || http://www.thousandparsec.net/planet || TP accepted to GSoC 2010! || Students: put your proposals on http://socghop.appspot.com/ now!"20:56
*** llnz sets mode: -o llnz20:56
llnzi will have to visit there one day20:57
null_000sure ^_^20:57
*** konr has joined #tp21:02
epyonllnz, have you read my email?21:05
llnzepyon: yes, will reply shortly21:05
marcel_Is it possible to download different images for ships depending on the ruleset from the server?21:19
llnzmarcel_: yes21:27
marcel_Is that included in the protocol 3, 4 or a seperate protocol?21:30
*** Noughmad has quit IRC21:35
*** tanoku has quit IRC21:35
llnz4, media parameter for objects21:51
*** null_000 has quit IRC21:51
*** isra17 has joined #tp22:34
isra17hello22:34
*** marcel_ has quit IRC22:46
*** epyon has quit IRC22:46
*** mithro has quit IRC23:48

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!