| *** tpb has joined #tp | 00:00 | |
| *** ChanServ sets mode: +o tpb | 00:00 | |
| jotham | nash: i take it you are a CG programmer? | 00:01 |
|---|---|---|
| nash | No... CG in this context? | 00:03 |
| nash | I program for embedded linux and other embedded OSes mostly | 00:03 |
| jotham | computer graphics | 00:03 |
| nash | We specialise in a graphics toolkit which works with evas | 00:03 |
| * nash wasn;'t sure if you meant Computer Games | 00:03 | |
| nash | I do a bit of graphical optimisatiosn however | 00:03 |
| jotham | heh i didn't know comptuer game programmer was a proper field, it's either graphics, ai, networking, etc | 00:04 |
| nash | You would be correct I suppose. | 00:05 |
| jotham | i am just trying to figure out the best way to draw a capsule shape, i.e. 2 half circles either end of a rectangle | 00:05 |
| nash | Problem is, I have the artistic talent of a one-legged duck | 00:05 |
| jotham | i am reasonably compitent at programming and asthetic things | 00:06 |
| jotham | depending who you ask ;-) | 00:06 |
| nash | Inkscape - draw a circle, add the appriopriate shadows and highlights, then add a cyclinder with the lists | 00:06 |
| nash | dup the circle in the appropriate places | 00:06 |
| jotham | i meant algorithmicly heh | 00:06 |
| jotham | i can draw one fine using half circles and a rectangle in whatever library | 00:07 |
| nash | Don't - prerender is much faster, unless you are using GL, and even then... | 00:07 |
| jotham | yeah i am prerendering, but i am prerendering in code | 00:07 |
| nash | Well... draw a circle, add the appropriate... ;-) | 00:07 |
| jotham | nevermind then...i was just curious as i am sure i can use pythons Numeric module to create a good filter for what i need | 00:08 |
| jotham | right now i am just using an object i made in photoshop saved as PNG | 00:08 |
| nash | Makes sense to me ;-) | 00:08 |
| jotham | but i don't know what the view is on dependance of library sprites in the file system | 00:08 |
| nash | Hence do it as something like SVG which can be sent across the network, and can be scaled on the target device... | 00:10 |
| jotham | yeah i do a lot of flash work in my job, and have been looking at svg a bit, but it hasn't been suitable for a project yet | 00:10 |
| jotham | but i think vector graphics are an excellent thing | 00:11 |
| nash | Indeed | 00:11 |
| jotham | have you seen amanith? | 00:11 |
| nash | No | 00:11 |
| jotham | google it, vector rasteriser that uses hardware | 00:12 |
| jotham | there are python bindings for it, i am considering using it for another project | 00:12 |
| *** JLP has joined #tp | 01:06 | |
| mithro | hey people | 05:11 |
| mithro | jotham / nash: ping? | 05:11 |
| jotham | sup | 05:11 |
| nash | pong | 05:11 |
| mithro | jotham: just reading scoll back now :) | 05:13 |
| mithro | jotham: feel free to require png files on the filesystem for the battleviewer if that is what you where concerned about | 05:13 |
| jotham | reason i haven't got any progress to show you is i haven't made a single laser beam i like | 05:14 |
| mithro | jotham: he he okay | 05:14 |
| nash | jotham: Make them x-ray lasers... so you don't need to see them ;-) | 05:15 |
| mithro | jotham: why not add a bunch of different ones and get other opionons? | 05:21 |
| jotham | well i think i'm gonna have to draw them dynamically and then use the Numeric module to apply filters to them (Gaussian ect) | 05:22 |
| jotham | i've experimented with rotating sprites but they look ugly as when you have to resize them to handle different distances between emitter and target | 05:22 |
| jotham | i dunno, i'll get there - just know i am working on it =) | 05:22 |
| jotham | i should have something tomorrow | 05:22 |
| jotham | and once i have the drawing method sorted we can play with variety | 05:22 |
| mithro | okay cool | 05:31 |
| mithro | nash: how goes your entry? :P | 05:46 |
| nash | okay. Haven't had much programming time recently - need to find a new place to live, internet issues... urr :-( But making progress. | 05:46 |
| jotham | that sucks, where do you live nash | 05:47 |
| nash | Concord (Sydney) | 05:47 |
| jotham | cool | 05:48 |
| mithro | nash: how are you talking to the server? are you planning on writing a libtpproto-c or just leaving it all in the client? | 05:59 |
| nash | It's in the client as such for now | 06:00 |
| *** tpb has joined #tp | 06:00 | |
| *** ChanServ sets mode: +o tpb | 06:00 | |
| nash | Basically when a message comes in, it pretty much gets thrown as an event into the application for the other bits to do what they like with it | 06:03 |
| jotham | hey mithro you said i should submit even my test scripts ? | 06:06 |
| jotham | if you want i can submit one i did when i was doing the numbers, it looks kinda cute | 06:06 |
| mithro | jotham: that would be good | 06:10 |
| CIA-11 | [email protected] * battleviewer/ (background.png background.png damage_test.py damage_test.py): damage animation test script | 06:14 |
| CIA-11 | [email protected] * web/ (5 files in 2 dirs): New Helpful Links section in the competition page. | 06:16 |
| mithro | nash: do you store a local copy of the universe? | 06:17 |
| nash | yep. | 06:19 |
| nash | Most of it | 06:19 |
| nash | Need to tweak how I update... | 06:19 |
| nash | at the moment each turn I'm just going "Give me everything" | 06:19 |
| nash | Is there a nice way of knowing when things have been destroyed BTW | 06:20 |
| * nash files it under "performance issue" | 06:22 | |
| mithro | nash: no | 06:22 |
| mithro | nash: you can use "get---ids" to find out all the ideas of existing objects | 06:23 |
| nash | okay. So basically I need to get list of all objects, and delete any which have been removed | 06:23 |
| nash | Yeah. Painful... | 06:23 |
| mithro | nash: TP04 will fix this problem | 06:23 |
| * nash notes the server needs to send out a new turn message with a list of "things changed" or something | 06:23 | |
| nash | Cool | 06:23 |
| nash | When will that be released? | 06:23 |
| nash | Is there a trial version I can use, and does the current darcs head cpp server support it? | 06:24 |
| mithro | * Difference support | 06:24 |
| mithro | The Get ID's frames will have a "from" field added. This is a SInt64 | 06:24 |
| mithro | which is the timestamp to get changes from. If it is -1 then it should | 06:24 |
| mithro | be all objects. | 06:24 |
| mithro | nothing supports it yet | 06:24 |
| mithro | we have been to busy trying to get everything in order for the comp :) | 06:24 |
| mithro | s/to/too | 06:24 |
| * nash updates FIXME to read: Will be fixed in tp04 | 06:26 | |
| mithro | I'm currently been redoing the starmap in tpclient-pywx-dev | 06:26 |
| nash | So how do you detect deleted/destroyed objects using that API? | 06:26 |
| mithro | you use get_xxx_ids to find out which ids have changed | 06:28 |
| mithro | then you get the objects with that ids that have changed (and notice which ones have been destroyed) | 06:28 |
| mithro | btw: most of the time turn generation will occur when clients are not connected | 06:28 |
| mithro | where as currently you have to get a list of ALL ids | 06:29 |
| mithro | (so if the universe has 10,000 objects you have to get 10,000 ids) | 06:29 |
| tpb | New news from http://www.thousandparsec.net/tp/rss.php: Another AI Competition Update. | 06:30 |
| nash | Okay | 06:31 |
| mithro | where as with TP04 you'll only need to get the objects which have changed | 06:32 |
| nash | Right | 06:32 |
| nash | So basically I'm doing the right thing | 06:33 |
| mithro | pretty muc | 06:33 |
| * nash hates it when ugly == right | 06:33 | |
| mithro | yes | 06:34 |
| mithro | there was a nicer solution, but it doesn't work with tpserver-cpp | 06:34 |
| nash | Ordered data? | 06:36 |
| mithro | yes - but that suffered from a problem with disappearing objects | 06:37 |
| nash | Ahh | 06:38 |
| nash | Anyway - I'm about to bail. | 06:41 |
| *** zipola has joined #tp | 06:43 | |
| *** nash has quit IRC | 06:47 | |
| *** llnz has joined #tp | 07:06 | |
| mithro | llnz: ping? | 07:14 |
| *** llnz has quit IRC | 07:27 | |
| *** zipola has quit IRC | 07:29 | |
| *** llnz has joined #tp | 08:34 | |
| jotham | a starfield is going to be pretty cpu intensive eh | 08:41 |
| jotham | it's a shame, my C++ one uses next to know CPU | 08:41 |
| jotham | hey mithro what's the deal with using other peoples code | 09:34 |
| jotham | or someone with better math knowledge could write a guassian blur using Numeric python | 09:38 |
| mithro | jotham: as long as it has a GPL compatible license and is properly documented | 09:41 |
| jotham | it's LGPL | 09:41 |
| mithro | thats fine then | 09:42 |
| jotham | problem is, it's for PIL so i'd have to hack it up to use it without pil, and still i'd have to do it with boost python or something | 09:42 |
| jotham | and i don't know anything about licensing stuff, maybe it's a waste of time, i'll use regular blur and see how it goes | 09:42 |
| mithro | it's C++? | 09:42 |
| jotham | yeah | 09:42 |
| jotham | but as a PIL plugin | 09:42 |
| mithro | basicly I would prefer a pure python solution for the basic stuff, you can add C++/PyRex/C stuff once we have a working solution | 09:43 |
| jotham | sure thing | 09:43 |
| mithro | infact you could rewrite it as a SDL C++ thing if you wanted - but first thing is to get something which works | 09:46 |
| jotham | yeah for sure | 09:46 |
| jotham | in my local copy of the darcs repo can i roll back a file to a previous state? | 09:50 |
| mithro | IE quick and dirty is definately okay for the moment | 09:50 |
| mithro | darcs revert | 09:50 |
| mithro | (if you havn't record the changes) | 09:50 |
| mithro | or darcs unrecord then darcs revert (if you havn't pushed) | 09:51 |
| *** llnz has quit IRC | 09:51 | |
| mithro | revert Revert to the recorded version | 09:52 |
| mithro | unrecord Remove recorded patches without changing the working copy. | 09:53 |
| mithro | rollback Record an inverse patch without changing the working directory. | 09:53 |
| mithro | lots of things you can do | 09:53 |
| *** llnz has joined #tp | 10:15 | |
| * llnz wanders off | 10:43 | |
| llnz | later all | 10:43 |
| *** llnz has quit IRC | 10:43 | |
| CIA-11 | [email protected] * battleviewer/ (laser_test.py laser_test.py): Laser test file. Work in progress. Nothing to see here. | 11:42 |
| jotham | oh well, least i got the line drawing code in and the blurring stuff sorted out, only need to blit the masks in and add some kind of perlin noise and we should have some ok looking laser beams | 11:43 |
| jotham | mithro: i'd like to get a look at those explosion animations as i am going to move the tinting pulse effect to when the lasers hit the ship | 11:44 |
| jotham | so i'll want to add in the explosions when death occures | 11:44 |
| jotham | even if you just give me the sequence i can recreate them as a spritesheet and animate them that way | 11:45 |
| *** mithro has quit IRC | 11:52 | |
| *** mithro has joined #tp | 11:54 | |
| jotham | ok, well i'm going to bed, later | 11:59 |
| jotham | shame about the guassian, it removes hard edges, something a normal blur doesn't do | 12:00 |
| *** jotham has left #tp | 12:01 | |
| *** JLP has quit IRC | 15:48 | |
| *** JLP has joined #tp | 15:51 | |
| *** adhoc has quit IRC | 17:01 | |
| *** zipola has joined #tp | 20:07 | |
| *** jotham has joined #tp | 20:31 | |
| jotham | morning | 20:31 |
| mithro | morning people | 21:39 |
| mithro | welp, off to uni | 21:44 |
| mithro | bblr | 21:44 |
| *** nash has joined #tp | 21:53 | |
| * nash waves | 21:56 | |
| *** mithro has quit IRC | 22:03 | |
| *** zipola has quit IRC | 22:08 | |
| *** adhoc has joined #tp | 22:22 | |
| *** mithro has joined #tp | 23:11 | |
| jotham | hey mithro you alive? | 23:11 |
| mithro | am now, but a bit lagged | 23:11 |
| jotham | where can i get those explosions | 23:12 |
| mithro | do you want the mng versions or the original PNG strips? | 23:12 |
| jotham | png strips, i'll (for now) just create a spritesheet out of them | 23:13 |
| mithro | okay, hold on | 23:14 |
| mithro | http://www.thousandparsec.net/svn/media/client/common-2d/foreign/vegastrike/explosion-small/ | 23:18 |
| tpb | <http://ln-s.net/InQ> (at www.thousandparsec.net) | 23:18 |
| mithro | http://www.thousandparsec.net/svn/media/client/common-2d/foreign/vegastrike/explosion-big/ | 23:18 |
| tpb | <http://ln-s.net/InR> (at www.thousandparsec.net) | 23:18 |
| mithro | you can use convert to get back png images | 23:18 |
| jotham | they look nice | 23:21 |
| jotham | by use convert, you mean use image magick? | 23:22 |
| mithro | convert abc.mng abc.png | 23:22 |
| jotham | yeah convert is part of the image magick suite | 23:22 |
| jotham | that's all i was asking | 23:22 |
| jotham | if you meant that program | 23:22 |
| mithro | i wasn't sure :P | 23:22 |
| jotham | before i started this i was writing a pygame tutorial | 23:24 |
| jotham | people ask the same damn questions every day in #pygame | 23:24 |
| jotham | where did you get all the graphics from | 23:27 |
| nash | It's a good question ;-) | 23:28 |
| mithro | i got these animations from a project called vegastrike - they are GPL so it's all fine | 23:29 |
| mithro | morning nas | 23:29 |
| mithro | s/nas/nash | 23:29 |
| jotham | doesn't look like convert works for going back to png | 23:30 |
| jotham | unless there is an option i am missing | 23:30 |
| jotham | turning an 806k MNG into a 182 byte PNG = no good | 23:30 |
| nash | mithro: Tab complete doesn't work? ;-) | 23:30 |
| * nash is still confused over "GPL" with images ;-) In fact GPL and anything but code, and marked up documentation ;-) | 23:31 | |
| jotham | hum wonder if there is a -split | 23:31 |
| mithro | jotham: it was working for me here | 23:31 |
| jotham | just now? | 23:31 |
| mithro | jotham: which one are you trying to convert? | 23:32 |
| jotham | Version: ImageMagick 6.2.4 11/24/06 Q16 http://www.imagemagick.org | 23:32 |
| tpb | Title: ImageMagick: Convert, Edit, and Compose Images (at www.imagemagick.org) | 23:32 |
| mithro | Version: ImageMagick 6.2.4 11/01/06 Q16 http://www.imagemagick.org | 23:32 |
| tpb | Title: ImageMagick: Convert, Edit, and Compose Images (at www.imagemagick.org) | 23:32 |
| jotham | bandcamp:14631$ convert orange.mng orange.png; ls -l orange.* | 23:32 |
| jotham | -rw-r--r-- 1 jotham jotham 824321 2007-02-13 12:21 orange.mng | 23:32 |
| jotham | -rw-r--r-- 1 jotham jotham 182 2007-02-13 12:27 orange.png | 23:32 |
| mithro | convert should automatically split it when you go from a animation to single image format | 23:33 |
| jotham | works with blue.mng | 23:33 |
| mithro | weird | 23:33 |
| mithro | dunno what is up with the orange one | 23:34 |
| jotham | damn it was the only one i wanted to use for now =) | 23:34 |
| jotham | so if my friend makes us a quick explosion for us to keep that's cool? | 23:35 |
| jotham | he's a 3D animator | 23:35 |
| jotham | otherwise i can just do it, a small fireball is all we need for now | 23:35 |
| mithro | yes, as long as he is willing to release it under the GPL | 23:37 |
| mithro | nash: the advice we have been given about images and GPL is to just say "lalalala can't hear you" :P | 23:38 |
| jotham | well like i said, i know nothing about licenses, it would belong to this project for you to license as you please | 23:38 |
| nash | ;-) | 23:38 |
| nash | mithro: Probably best to have the save file from the original program if you can - eg xcf or svg file | 23:39 |
| nash | We also do that commercially just in case someone claims we stole their art | 23:39 |
| mithro | jotham: that would be fine, as nash said the source would be cool too | 23:39 |
| jotham | well it'd be using a particle effect plugin that belongs to a comercial software package, but sure i could give you the project file | 23:40 |
| mithro | jotham: that would be okay | 23:41 |
| mithro | jotham: licensing is really simple (if you don't really care :) - if you created, you can put it under any license you want | 23:44 |
| mithro | GPL - I want my work to be free and to remain free | 23:44 |
| mithro | BSD - do what the hell you want with it | 23:44 |
| mithro | all the other licenses are not all that useful (arguably) | 23:45 |
| nash | mithro: LGPL is very useful, and may be relevant for artwork... | 23:47 |
| nash | All depends on what linking means in this context. | 23:47 |
| nash | Probably best to make sure you are clear on what you intend to allow people to do with the artwork. | 23:48 |
| mithro | nash: linking and stuff has problems when you talk about python programs too | 23:48 |
| * nash doesn't touch python, so it hasn't bothered him yet... | 23:49 | |
| mithro | libtpproto-py is LGPL with the following clarification | 23:49 |
| mithro | It is the author's understanding that Python code which calls this | 23:49 |
| mithro | library is "linked" to it and not a derivitive work. | 23:49 |
| nash | Cool. Seems reasonable to me. | 23:50 |
| mithro | interrupts are making my brain hurt | 23:51 |
| nash | As in int 21h, or trap? | 23:51 |
| mithro | nash: as in on a PIC, you would think it would make sense to have 0 == interrupt disabled, 1 == interrupt enabled | 23:52 |
| mithro | but some interrupts have 0 == enabled, others have 1 == enabled | 23:53 |
| nash | Ahh... yep.., | 23:54 |
| * nash notes this is why nash became essentially pure software... I used to do a lot of OS and microkernel stuff | 23:55 | |
| mithro | plus i've got a bit of heart burn going on which is making it hard to concentrate | 23:55 |
| nash | Urgh.. and stupid licence conversations on IRC tooo ;-) Do you have some ant-acid for the heartburn? | 23:56 |
| jotham | heh | 23:59 |
| jotham | i am an artist, i just make stuff and i don't care who uses it or steals it or wahtever | 23:59 |
| mithro | jotham: then use GPL or BSD license pretty much | 23:59 |
Generated by irclog2html.py 2.5 by Marius Gedminas - find it at mg.pov.lt!