| 00:01.49 | *** join/#bzflag anom4ly (~anom4ly@adsl-98-70-44-20.gnv.bellsouth.net) |
| 00:21.32 | bryjen | i'd thought using /flag give was confusing it, but even with a normal pickup, i'm getting end-shot kicked :/ |
| 00:29.33 | *** join/#bzflag spldart (~spldart2@c-98-198-227-61.hsd1.tx.comcast.net) |
| 00:29.39 | *** join/#bzflag spldart (~spldart2@bzflag/contributor/spldart) |
| 00:29.39 | *** mode/#bzflag [+v spldart] by ChanServ |
| 00:42.39 | *** join/#bzflag JeffM (~Jeff@unaffiliated/jeffm2501) |
| 00:42.39 | *** mode/#bzflag [+v JeffM] by ChanServ |
| 00:55.22 | bryjen | ah, fiddling with shotSpeed too. maybe that's the one |
| 01:44.59 | *** join/#bzflag noyb (~noyb@pool-173-60-180-80.lsanca.dsl-w.verizon.net) |
| 02:11.55 | *** join/#bzflag TimRiker (~TimRiker@bzflag/projectlead/TimRiker) |
| 02:11.56 | *** mode/#bzflag [+o TimRiker] by ChanServ |
| 02:32.38 | *** join/#bzflag ibot (~ibot@rikers.org) |
| 02:32.38 | *** topic/#bzflag is http://my.BZFlag.org || http://cia.vc/stats/project/BZFlag || http://my.BZFlag.org/w/Getting_Help || Channel Logs: http://ibot.rikers.org/%23bzflag/ || 2.4 beta now, test like the wind || The PLAN http://my.bzflag.org/w/Development_RoadMap |
| 02:32.38 | *** mode/#bzflag [+o ibot] by ChanServ |
| 02:37.03 | *** join/#bzflag ibot (~ibot@rikers.org) |
| 02:37.03 | *** topic/#bzflag is http://my.BZFlag.org || http://cia.vc/stats/project/BZFlag || http://my.BZFlag.org/w/Getting_Help || Channel Logs: http://ibot.rikers.org/%23bzflag/ || 2.4 beta now, test like the wind || The PLAN http://my.bzflag.org/w/Development_RoadMap |
| 02:37.03 | *** mode/#bzflag [+o ibot] by ChanServ |
| 03:24.17 | *** join/#bzflag bryjen (~bryjen@76.92.85.169) |
| 03:24.17 | *** mode/#bzflag [+v bryjen] by ChanServ |
| 03:41.47 | *** join/#bzflag Mopar_Madness (a81c8812@gateway/web/freenode/ip.168.28.136.18) |
| 04:01.31 | *** join/#bzflag DarkCalf (DC@173.231.40.98) |
| 04:36.40 | *** join/#bzflag Anxuiz (~Anxuiz@unaffiliated/anxuiz) |
| 05:46.07 | *** join/#bzflag noyb (~noyb@pool-173-60-180-80.lsanca.dsl-w.verizon.net) |
| 05:48.39 | *** join/#bzflag meeba (~lamer@c-71-196-238-53.hsd1.co.comcast.net) |
| 06:00.40 | *** join/#bzflag bryjen (~bryjen@76.92.85.169) |
| 06:00.41 | *** mode/#bzflag [+v bryjen] by ChanServ |
| 06:46.35 | *** join/#bzflag Marzipan- (~Marzipan@p5B2236F7.dip.t-dialin.net) |
| 06:47.52 | *** join/#bzflag Marzipan (~Marzipan@bzflag/player/Marzipan) |
| 08:09.05 | *** join/#bzflag bier|tp (~bier@p57A6759B.dip.t-dialin.net) |
| 08:20.23 | *** join/#bzflag Pimpi (~frank@gondolin.pimpi.org) |
| 08:21.36 | *** join/#bzflag Pimpinella (~frank@gondolin.pimpi.org) |
| 09:13.57 | *** join/#bzflag KTL (~KTL@213.219.155.152.adsl.dyn.edpnet.net) |
| 10:40.10 | *** join/#bzflag trepan (~trepan@204.237.8.232) |
| 10:40.10 | *** join/#bzflag trepan (~trepan@unaffiliated/trepan) |
| 10:40.10 | *** mode/#bzflag [+v trepan] by ChanServ |
| 10:51.20 | *** join/#bzflag jcp (alex@bzflag/contributor/javawizard2539) |
| 12:18.20 | *** join/#bzflag fahadsadah (fahad@unaffiliated/fahadsadah) |
| 12:56.43 | *** join/#bzflag mdskpr_ (~mdskpr@108.25.145.104) |
| 14:19.56 | *** join/#bzflag meeba (~lamer@c-71-196-238-53.hsd1.co.comcast.net) |
| 14:43.53 | *** join/#bzflag bier (~bier@p57A67D62.dip.t-dialin.net) |
| 15:04.30 | *** join/#bzflag spldart (~spldart2@c-98-201-137-215.hsd1.tx.comcast.net) |
| 15:04.40 | *** join/#bzflag spldart (~spldart2@bzflag/contributor/spldart) |
| 15:04.40 | *** mode/#bzflag [+v spldart] by ChanServ |
| 15:08.09 | *** join/#bzflag TimRiker (~TimRiker@bzflag/projectlead/TimRiker) |
| 15:08.09 | *** mode/#bzflag [+o TimRiker] by ChanServ |
| 15:23.32 | *** join/#bzflag Anxuiz_ (~Anxuiz@c-76-22-102-23.hsd1.wa.comcast.net) |
| 15:23.32 | *** join/#bzflag Anxuiz (~Anxuiz@unaffiliated/anxuiz) |
| 15:25.03 | *** join/#bzflag bryjen (~bryjen@76.92.85.169) |
| 15:25.04 | *** mode/#bzflag [+v bryjen] by ChanServ |
| 15:38.53 | *** join/#bzflag JeffM (~JeffM@69.36.85.141) |
| 15:38.54 | *** join/#bzflag JeffM (~JeffM@unaffiliated/jeffm2501) |
| 15:38.54 | *** mode/#bzflag [+v JeffM] by ChanServ |
| 15:50.33 | CIA-5 | BZFlag: 03JeffM2501 07http://my.bzflag.org * r7767 10/w/Development_RoadMap: updates for multiple versions based on discussions in IRC and forums |
| 15:51.06 | *** join/#bzflag temporalD (~a_temp_di@bzflag/serverop/TemporalDistraction) |
| 15:56.31 | JeffM | what is more popular open treads or covered treads? |
| 15:57.14 | Thumper_ | I currently use covered treads (had open treads for ages though) |
| 15:57.28 | JeffM | coverd treads seem the best looking to me :) |
| 15:57.37 | Thumper_ | ditto |
| 15:57.50 | JeffM | is there any reason to turn off animated treads? |
| 15:58.07 | TD-Linux | I use open treads |
| 15:58.13 | BulletCatcher | I have always used open treads. |
| 15:58.16 | TD-Linux | isn't that the default? |
| 15:58.31 | JeffM | yeah, it was the only mode for animateds at one time IRRC |
| 15:58.36 | JeffM | so a lot of people pick it up |
| 15:59.06 | JeffM | ok, so can't remove it :) |
| 15:59.12 | JeffM | but could make it per tank |
| 15:59.24 | JeffM | so you could customize |
| 15:59.25 | BulletCatcher | Tanks with covered treads just look boring. :-) |
| 15:59.49 | Thumper_ | BulletCatcher: your tank or the other guy? |
| 16:00.13 | JeffM | have to make your view draw everyone the same tank tho when you get CB |
| 16:00.20 | KTL | stealth tanks are a bit more vulnerable if they leave a trail |
| 16:00.41 | JeffM | yeah I don't think treads should be an option for that reason |
| 16:00.49 | JeffM | treadmarks I mean |
| 16:00.52 | mdskpr_ | cloaking too |
| 16:01.02 | JeffM | it has impact like fog does |
| 16:01.07 | KTL | and cl in no-radar maps can be seen when they fall |
| 16:01.14 | JeffM | good |
| 16:01.14 | BulletCatcher | No, treadmarks are an important weakness of the ST and CL flags. |
| 16:01.20 | TD-Linux | I'm actually playing with a map that uses fog as darkness |
| 16:01.23 | JeffM | yeah so everyone should see them |
| 16:01.42 | JeffM | yeah that was a common trick in 1998 black fog |
| 16:01.53 | TD-Linux | still fresh for bzflag |
| 16:01.57 | JeffM | sadly |
| 16:02.02 | BulletCatcher | Thumper_: I don't quite understand your question. |
| 16:02.08 | KTL | a humming diesel-engine sound for nearby tanks :D |
| 16:02.20 | JeffM | for CL treads I could see a point in making them fade slightly faster but I think they should be there |
| 16:02.28 | Thumper_ | BulletCatcher: I like open tracks for MY tank (out the window) but covered tracks on other tanks since I think they look better |
| 16:02.34 | TD-Linux | client side scripting which can place opengl light sources is another option :> |
| 16:02.51 | JeffM | TD-Linux, our lighthing model isn't that complex |
| 16:02.57 | JeffM | we don't do volumetric shadows |
| 16:03.09 | TD-Linux | I know |
| 16:03.11 | JeffM | you'd want shaders to do the lighting |
| 16:03.15 | Thumper_ | BulletCatcher: the question is which tank looks boring? Yours or your opponents or both? |
| 16:04.20 | BulletCatcher | I can't see my own tank out the window at all, except in shadows. |
| 16:04.34 | JeffM | there is an option to show your own tank |
| 16:04.42 | JeffM | and with uncovered treads you see them rollin |
| 16:04.49 | JeffM | you see them haitin |
| 16:04.58 | BulletCatcher | looks again |
| 16:05.08 | KTL | (track marks and treads : different things) |
| 16:05.14 | JeffM | thinks there are too many options |
| 16:10.31 | BulletCatcher | I don't show my own tank because it blocks the view looking downward. |
| 16:10.33 | BulletCatcher | If I did show my own tank, then animated treads on it would be too distracting and I would want fenders. |
| 16:11.27 | BulletCatcher | So, we are completely opposite on this matter, Thumper_. :-) |
| 16:11.32 | Thumper_ | :) |
| 16:12.04 | Thumper_ | likes the visual cue when narrow/tiny is stolen while the rabbit |
| 16:12.15 | Thumper_ | (or cloaking) |
| 16:13.26 | JeffM | yeah I'm not sure why people would want to see there own tank |
| 16:13.47 | JeffM | well visual clues are good but there are better ways for us to do that |
| 16:13.52 | JeffM | our HUD needs to show things better |
| 16:14.29 | Thumper_ | but it does block the view which is definitely a drawback (and probalby explains my crappy score ;) |
| 16:14.29 | JeffM | but that's a different release :) |
| 16:18.59 | BulletCatcher | The reduced view isn't a problem on non-jumping maps, but those are rare these days. |
| 16:23.38 | *** join/#bzflag bier|tp (~bier@p57A67D62.dip.t-dialin.net) |
| 16:27.21 | trepan | on no-radar servers, CL track marks are all that you get |
| 16:27.39 | trepan | if you want to adjust the fade, i've already provided for it: _trackFade |
| 16:28.44 | trepan | (track marks are all that you get assuming hunting, identify, or hud markers to give the tank away) |
| 16:31.01 | *** join/#bzflag Anxuiz__ (~Anxuiz@c-76-22-102-23.hsd1.wa.comcast.net) |
| 16:38.17 | KTL | well, there is also the hunt-warning sound |
| 16:38.45 | KTL | very important on that mystic island map (or whatever-it-is-called) |
| 16:41.11 | trepan | ah, right; _forbidHunting didn't make it in, or _forbidIdentify |
| 16:42.37 | JeffM | trepan, yeah I'd like to make it per tank |
| 16:42.52 | JeffM | so it can be adjusted based on what flag you have |
| 16:42.55 | trepan | a lot of properties should be shifted to "per-tank" |
| 16:42.58 | JeffM | yeah |
| 16:43.02 | JeffM | one day |
| 16:43.49 | trepan | i used a per-client BZDB hack to setup per-tank speed/angvel on the PTF server, but there's a limited number of parameters that can use that trick |
| 16:44.35 | trepan | and it's not like it's going to be a memory hog -- only have a few tanks (as opposed to an RTS, where per-unit settings can start to add up) |
| 16:44.40 | JeffM | so how are we doing on bugs in beta2? |
| 17:01.19 | BulletCatcher | I found a minor one in /report. |
| 17:01.41 | BulletCatcher | And trepan just figured out that HAVE_GLEW isn't being used. |
| 17:01.52 | BulletCatcher | Both ought to be easy to fix. |
| 17:07.07 | BulletCatcher | HAVE_GLEW is used, but not defined when it should be. |
| 17:14.18 | BulletCatcher | The textured tank in zoomed radar changes the visual cues I need to do a pyramid jump to base on the HiX map. |
| 17:14.20 | BulletCatcher | I think I'd like the option to keep the white rectangle. |
| 17:15.10 | *** join/#bzflag trepan (~trepan@204.237.8.232) |
| 17:15.10 | *** join/#bzflag trepan (~trepan@unaffiliated/trepan) |
| 17:15.10 | *** mode/#bzflag [+v trepan] by ChanServ |
| 17:15.11 | BulletCatcher | Moving the textured tank to experimental quality would be good enough for me. |
| 17:18.30 | CIA-5 | BZFlag: 03bullet_catcher * r22007 10/trunk/bzflag/configure.ac: Define HAVE_GLEW when it is wanted and available. |
| 17:18.57 | JeffM | yeah just remove the tank |
| 17:19.07 | JeffM | then I can fix it later |
| 17:19.25 | JeffM | and windows is already set to use glew |
| 17:19.36 | JeffM | did that just before beta1 |
| 17:19.40 | BulletCatcher | Yeah, that was a fix for *nix only. |
| 17:20.36 | BulletCatcher | The textured tank on radar can be added whenever it is ready without a protocol change. |
| 17:20.58 | JeffM | yeah totaly |
| 17:21.07 | BulletCatcher | It isn't (and shouldn't be) a gameplay changer. |
| 17:21.58 | trepan | BulletCatcher: killed in "r19127 | brlcad | 2009-01-02 01:00:09 -0500 (Fri, 02 Jan 2009) | 1 line" -- been dead a while |
| 17:23.53 | JeffM | still sounds like we are getting pretty close |
| 17:24.19 | BulletCatcher | Yes, I think we are converging on a version we can release. |
| 17:24.22 | JeffM | maybe we can get an RC out this wekend or early in the week |
| 17:26.22 | BulletCatcher | brlcad's comment about "no longer need HAVE_GLEW" was probably true for 2.99. |
| 17:26.32 | JeffM | proibably |
| 17:26.42 | trepan | the 2.99.x code doesn't use HAVE_GLEW - is it required in 2.4? (if system glew doesn't work, we always build our own?) |
| 17:27.06 | JeffM | 2.4 built fine with no glew and with out the def |
| 17:27.10 | JeffM | so it must be condtional |
| 17:27.15 | JeffM | cus it didn't complain about headers or anything |
| 17:27.19 | BulletCatcher | 2.3 code has "#ifdef HAVE_GLUE" in several places. |
| 17:27.37 | JeffM | when I set the def i got missing headers untill I added them to the search path |
| 17:27.45 | JeffM | so yeah it's tied to the def |
| 17:27.56 | JeffM | it was optional for 2.0, required for 3 |
| 17:28.14 | trepan | and for 2.4? |
| 17:28.45 | BulletCatcher | Since we will now build GLEW if it isn't available, we could remove the #ifdef's and just expect it to be there. |
| 17:28.55 | BulletCatcher | I assume that is what happened in 2.99. |
| 17:29.00 | trepan | exactly |
| 17:29.24 | JeffM | yeah it's always built for widers |
| 17:29.30 | JeffM | in the chezzyest way:) |
| 17:29.44 | JeffM | I just included the single C file for it in the client code :) |
| 17:30.53 | CIA-5 | BZFlag: 03JeffM2501 * r22008 10/trunk/bzflag/src/bzflag/RadarRenderer.cxx: turn off the fancy tank for now, we'll fix it later. |
| 17:44.36 | JeffM | I'm curious, how many people run in low or medium quality? |
| 17:44.50 | Thumper_ | not me |
| 17:45.03 | JeffM | I'm wondering if we even need the tank LODs anymore |
| 17:45.58 | Thumper_ | running in lower quality gives me poorer framerates than high quality/experimental |
| 17:46.29 | JeffM | yeah it should |
| 17:46.41 | JeffM | cus it's made for software rendering and does things most cards suck at :) |
| 17:46.48 | Thumper_ | but it's counterintuitive (to me) :) |
| 17:46.51 | JeffM | it is |
| 17:47.06 | JeffM | just like many other things in bzflag :) |
| 17:47.10 | Thumper_ | heh |
| 17:47.21 | JeffM | really low and medium should just lower texture sizes and particile counts |
| 17:47.25 | JeffM | and turn off shadows etc... |
| 17:47.32 | JeffM | but still draw the same way |
| 17:47.48 | JeffM | then they would do what they say they do |
| 17:48.08 | JeffM | we allready assume everyone has SOME kind of hardware acceleration |
| 17:48.21 | JeffM | and we can even assume hardware T&L |
| 17:48.37 | JeffM | I don't think there are any fillrate only cards out there. |
| 17:49.30 | JeffM | we are even close to where we could require bassic shaders |
| 17:50.36 | trepan | easy enough to support a "crap" setting, no textures, no shaders, no nothing (and no extra options for that mode) |
| 17:50.49 | JeffM | yeah |
| 17:51.07 | trepan | and you jumpd straight from crap to minimal shaders |
| 17:52.28 | JeffM | I mean we JUST desuported riva 128's |
| 17:52.31 | JeffM | how sad is that |
| 17:52.42 | JeffM | and the Voodoo2 |
| 17:53.38 | BulletCatcher | One of BZFlag's unique features has always been that it runs on old hardware. I see no reason to rush to change that. |
| 17:53.55 | JeffM | yeah but how old? |
| 17:54.10 | JeffM | the riva was made in 1997 |
| 17:54.15 | trepan | fwiw, supporting a crap mode isn't hard, it's supporting 12 different options that are basically just for crap h/w that's a pita |
| 17:54.19 | JeffM | and how much do you let that slow down progress? |
| 17:54.24 | JeffM | yeah |
| 17:54.35 | JeffM | I'd like to move all the options into high med low |
| 17:54.40 | JeffM | simplify it |
| 17:54.44 | BulletCatcher | Good questions, JeffM. I don't have a clear answer. |
| 17:54.55 | JeffM | we don't need 8 different texture modes when only 2 are ever used |
| 17:55.18 | trepan | actually ... i'd support all the texture modes, but let materials define them |
| 17:55.28 | JeffM | ok I'll buy that |
| 17:55.31 | trepan | (but there would be no global texture mode) |
| 17:55.50 | JeffM | because the interpolation affects how it looks |
| 17:55.50 | trepan | well, nothing user visible; there'd be a default, obviously |
| 17:55.54 | JeffM | yeah |
| 17:56.03 | JeffM | you may want to keep it nearest to make it look 8 bit :) |
| 17:56.24 | trepan | there are cases for requiring NEAREST texture modes |
| 17:56.34 | JeffM | yeah when you don't want it fuzzy |
| 17:56.47 | JeffM | so yeah I buy that |
| 17:56.59 | JeffM | but we don't need to do it in the options nor globaly |
| 17:58.00 | JeffM | BulletCatcher, I'm ok with not requiring top of the line hardware but I think we need to have a line we draw, like say "5 year old hardware" |
| 17:58.06 | JeffM | so we can know what to expect |
| 17:59.44 | trepan | and older cards are supported, they just draw plain colors (no textures, no lighting, no attempts to fake lighting, etc...) |
| 17:59.51 | JeffM | yeah |
| 18:00.01 | JeffM | it's not like it won't work, it'll just not be as pretty |
| 18:00.28 | trepan | tbh, in my setup, even crap mode requires textures :) |
| 18:00.38 | JeffM | yeah I think we can use textures |
| 18:00.45 | JeffM | I can't think of anything that can't draw em |
| 18:00.59 | JeffM | we could just limit the size |
| 18:01.20 | trepan | it's not just texture size, it's screen size that's the big one |
| 18:01.28 | JeffM | yeah |
| 18:01.30 | trepan | (fillrate, as you say) |
| 18:01.32 | JeffM | that eats vram |
| 18:01.38 | BulletCatcher | The trick is to find the right balance between developer effort and support for players with old hardware. |
| 18:01.45 | JeffM | BulletCatcher, yes |
| 18:02.00 | JeffM | I don't think we should require GL4 for example |
| 18:02.02 | trepan | but for BZ, no textures in a-la-crap mode, support all the way back to VIC20 |
| 18:02.21 | JeffM | well you have to have some textures, fonts use textures |
| 18:02.42 | trepan | heh, ya; that's why I did it too ... totally forgot |
| 18:02.48 | JeffM | thats why I think the texture off mode now is stupid |
| 18:02.52 | BulletCatcher | I don't think we necessarily need to remove support old hardware unless it causes real trouble for devs. |
| 18:02.52 | JeffM | cus we turn it on for text |
| 18:02.56 | trepan | could go with bitmaps, but that's even more work |
| 18:03.02 | trepan | BulletCatcher: it does |
| 18:03.08 | JeffM | BulletCatcher, no just when we clean up the code we can make some assumptions |
| 18:03.15 | JeffM | and not have to worry about special cases |
| 18:03.27 | trepan | and I've done more gfx work for this game then anyone else, so I am an authority :) |
| 18:03.33 | BulletCatcher | On the other hand, it may be time to stop pretending that we support old operating systems like IRIX and BeOS. |
| 18:03.55 | JeffM | heh |
| 18:07.03 | BulletCatcher | Maybe after he returns from vacation, blast007 will provide a histogram of client operating systems seen at Missile Wars. |
| 18:08.05 | BulletCatcher | BTW, the real-time /report message to admins looks like this: |
| 18:08.10 | BulletCatcher | **"Bullet Catcher" reports: nice map |
| 18:08.13 | BulletCatcher | I thought the ** was a remnant of the 2.0 /me handling, but both 2.0 and 2.3 work the same way, so I'll leave it alone. |
| 18:11.07 | trepan | MsgQueryGL backport? :) |
| 18:11.07 | trepan | (that's what's really needed, client strings will only get you so much) |
| 18:15.37 | BulletCatcher | Getting more details about client configurations is worthwhile, but we need to design it carefully to balance our interests vs. player privacy. |
| 18:17.55 | *** join/#bzflag trepan (~trepan@204.237.8.232) |
| 18:17.55 | *** join/#bzflag trepan (~trepan@unaffiliated/trepan) |
| 18:17.55 | *** mode/#bzflag [+v trepan] by ChanServ |
| 18:19.53 | BulletCatcher | said after trepan lost connectivity: Getting more details about client configurations is worthwhile, but we need to design it carefully to balance our interests vs. player privacy. |
| 18:33.19 | trepan | BulletCatcher: easy to setup a BZDB var to block queries |
| 18:34.40 | trepan | rather than have MsgQueryGL, i'd rather the info be sent to the list server; a more limited audience, and one more likely to use the info |
| 18:35.00 | JeffM | yeah the server dosn't care |
| 18:35.07 | JeffM | the game server I mean |
| 18:35.28 | BulletCatcher | As long the default setting avoids "spyware!" accusations, it should be fine. |
| 18:36.01 | BulletCatcher | I agree that the list server is a better place to collect those stats. |
| 18:36.24 | BulletCatcher | Tack it on to token queries. |
| 18:37.31 | trepan | that's part of the problem, MsgQueryGL is nice in that it can query a number of different GL parameters |
| 18:38.02 | trepan | but that would involve client->listserver (hello), listserver->client (which params), client->listserver (results) |
| 18:38.25 | JeffM | BulletCatcher, yes I'd tie to an option for "anon useage stats" that they can opt out of |
| 18:38.39 | BulletCatcher | These complications are why I say it needs careful design. :-) |
| 18:38.45 | JeffM | yep |
| 18:39.06 | JeffM | there are other apps that do similar so we just have to look around and see what's acceptable in this day |
| 18:39.45 | JeffM | we can even make it a condtion of global registration |
| 18:39.57 | BulletCatcher | Collecting web cam photos is right out. :-) |
| 18:39.59 | JeffM | but yes it shoul have a spec |
| 18:40.01 | trepan | in 6pt font? :) |
| 18:40.20 | JeffM | trepan, that'd be bigger then some I've seen |
| 18:40.27 | JeffM | I mean we are keeping NON anon stats now with list server logs |
| 18:40.53 | JeffM | and I think we get platform from the curl hit |
| 18:41.02 | JeffM | so that should be in weblogs |
| 18:42.22 | *** join/#bzflag I_Died_Once (~I_Died_On@unaffiliated/idiedonce/x-1828535) |
| 18:49.14 | BulletCatcher | http://pastebin.com/mwBA21vu has a histogram of over 40000 version strings seen at BRL for roughly the last year. |
| 18:49.17 | BulletCatcher | Windows dominates, Linux and Mac are about even, with also-rans by Solaris, FreeBSD, and a lone OpenBSD client. |
| 18:49.19 | BulletCatcher | I made no attempt to remove duplicate entries by the same player. |
| 18:50.01 | JeffM | wow that's a big number for windows |
| 18:50.18 | BulletCatcher | Yup. More than all others combined. |
| 18:51.15 | JeffM | heh 3 bsd people |
| 18:51.31 | JeffM | what version is darwin 8? |
| 18:51.37 | JeffM | 10.5 I hope? |
| 18:52.06 | JeffM | ahh 15 bsd, not 3 |
| 18:52.49 | Constitution | assuming that reflects the execution OS, not the build OS, 10.6 gives darwin8 |
| 18:53.42 | JeffM | it could be build OS |
| 18:53.53 | JeffM | someone would have to look into curl to see what it does |
| 18:53.56 | BulletCatcher | I get darwin8.11.1 when I run the official 2.0.16 app on Mac OS 1.4. |
| 18:54.07 | JeffM | or try the same bin on 2 OSs |
| 18:54.30 | BulletCatcher | So I think darwin8 reflects the build host. |
| 18:54.55 | JeffM | that would make sense seeing what is shown for windows too |
| 18:55.03 | Constitution | the official builds have been built on 10.4 for intel, and 10.3 for ppc, for the last several revisions |
| 18:55.06 | JeffM | since there are 0 64 bit windows reproted |
| 18:55.17 | *** join/#bzflag sirquine (~quine@216.7.151.125) |
| 18:55.19 | JeffM | ok so that's not useful for determining specific Os then |
| 18:55.25 | JeffM | but still good for platform |
| 18:55.48 | BulletCatcher | I get the same darwin8.11.1 for the 2.3 client I built on 10.4. |
| 18:56.12 | JeffM | it would be nice at some point to filter that by just people who authenticated then make them uniques |
| 18:56.34 | JeffM | that would give us the platforms of our core long term playerbase |
| 18:56.49 | BulletCatcher | I'll look into that. It will require a more complicated script. |
| 18:57.06 | JeffM | no rush |
| 18:57.12 | JeffM | just would be nice to have at some point |
| 18:57.33 | BulletCatcher | The pastebin data is from a long 1-line unix command. |
| 18:57.34 | JeffM | that at least would give us some kind of expectation for downloads for a new version |
| 18:57.40 | JeffM | heh |
| 18:59.13 | BulletCatcher | Perl FTW, and I'll commit it to the misc/ directory. |
| 18:59.53 | JeffM | probably dosn't need /misc |
| 18:59.58 | JeffM | put it in the web repo |
| 19:00.11 | JeffM | a end user would not need it but someone running a list server would |
| 19:00.36 | BulletCatcher | Well, this version will get its input from game server logs. |
| 19:00.41 | JeffM | ahh ok |
| 19:00.45 | JeffM | then yeah misc |
| 19:01.19 | Constitution | BulletCatcher: I theorize, then, that windows clients crash four times more often than mac or linux, thereby resulting in four times as many total joins ;) |
| 19:01.52 | JeffM | riiiiggghhht... that's why brlcad had to ship the mac build in debug to get better logs for a decade :) |
| 19:02.28 | trepan | Constitution: s/clients// ? |
| 19:03.08 | Constitution | precisely, trepan |
| 19:03.36 | BulletCatcher | Another triumph for open review and the scientific method! |
| 19:05.04 | trepan | Constitution: i hear that winders is much better these days, don't be hatin' :) |
| 19:07.36 | BulletCatcher | With r22008 the Anisotropic menu item is functional on Linux. |
| 19:07.38 | BulletCatcher | I don't know where to look for its effect, though. |
| 19:08.08 | trepan | it works, just look at the textures in the distance |
| 19:08.56 | Pimpi | many *NIX users have programming skills, but only very few windows users. -> Bugs on *NIX get fixed, those on windows not |
| 19:09.14 | trepan | (the one's with a plane normal that is just slighty off-perpendicular to your viewing direction, ex: the ground) |
| 19:09.46 | JeffM | it should look smoother |
| 19:09.52 | trepan | less smooth, actually |
| 19:09.59 | JeffM | less blurry |
| 19:10.06 | trepan | exactly |
| 19:10.14 | BulletCatcher | Okay, I can see that higher Anisotropic settings increase the detail in distant grass. |
| 19:10.17 | JeffM | righter I should hav esaid :) |
| 19:10.39 | JeffM | pimpi mostly cus many projects make it hard to build on windows |
| 19:11.42 | Pimpi | JeffM, wouldn't mingw builds be more, well, native? |
| 19:12.17 | JeffM | typical linux developer :) |
| 19:12.27 | JeffM | god no, it is iso NOT native that it huts |
| 19:12.29 | JeffM | hurts |
| 19:12.56 | JeffM | dosnt use direct X, dosn't use the MS compiler |
| 19:13.07 | JeffM | ends up building a linux ish app |
| 19:13.25 | JeffM | there IS a free compiler on windows |
| 19:13.28 | JeffM | has been for a while |
| 19:13.35 | JeffM | it's called visual studio |
| 19:13.38 | trepan | the SpringRTS have to use mingw, but have odd constraints |
| 19:13.44 | trepan | *but they have |
| 19:13.52 | JeffM | yeah it can be made to work but it's a bit of a pain to setup |
| 19:14.03 | JeffM | that drives some windows devs away |
| 19:14.11 | JeffM | they have to change there build style to match your unixisms |
| 19:14.31 | JeffM | thats why I try to keep the vc projects very easy to not drive away windows devs |
| 19:14.44 | trepan | they're a cmake crew now, makes everyone happy |
| 19:15.00 | JeffM | yeah cmake can build the projects so it isn't too bad if it's well documented |
| 19:15.08 | trepan | and projects with only VC projects tend to keep *nix devs away, swings both ways :) |
| 19:15.10 | JeffM | premake should be able to do the same thing |
| 19:15.15 | JeffM | yeah I agree |
| 19:15.29 | JeffM | unless it's .net stuff, then the vc projects build on anything :) |
| 19:15.40 | trepan | ya, wouldn't take me long to finish the premake stuff off if I had a winders box to play |
| 19:16.01 | JeffM | but I would not say that there are "less" developers on windows |
| 19:16.10 | JeffM | percentage wise of users maybe |
| 19:16.14 | JeffM | but not in number |
| 19:16.15 | trepan | professionally, there may well be more |
| 19:16.20 | JeffM | yeah |
| 19:16.41 | JeffM | profesional nix developers are generaly relegated to some specific industries |
| 19:16.46 | trepan | but hackers, i'd say they gravitate towards posix |
| 19:16.48 | JeffM | mostly server level stuff |
| 19:16.49 | Pimpi | goes to the basement and tries to look unsuspicious |
| 19:17.05 | JeffM | pimpi as long as it's not your mom's basement |
| 19:17.13 | JeffM | and you arn't sporting an neckbeard |
| 19:17.21 | Pimpi | ;) |
| 19:19.03 | *** join/#bzflag Erroneous (~DTRemenak@69.36.85.141) |
| 19:19.58 | moiree | BulletCatcher: when configuring --without-sdl this is needed: <http://pastebin.com/sHFnY9NQ> |
| 19:21.26 | BulletCatcher | Thanks, moiree. I'll confirm and commit. |
| 19:21.33 | JeffM | why would you build with out sdl? |
| 19:21.43 | BulletCatcher | Because we can. |
| 19:21.47 | JeffM | why? |
| 19:21.51 | trepan | JeffM: why would you? :) |
| 19:22.01 | BulletCatcher | I have never bothered to install SDL on my Mac. |
| 19:22.06 | JeffM | trepan, cus I get more features out of native windows |
| 19:22.07 | moiree | Why is it a configure option? |
| 19:22.09 | trepan | once SDL1.3 is out, the force feedback issue should go away |
| 19:22.12 | trepan | what else is there? |
| 19:22.19 | JeffM | more buttons IIRC |
| 19:22.30 | trepan | again, SDL1.3 (soonish?) |
| 19:22.37 | JeffM | then yeah I could use it |
| 19:23.07 | JeffM | tho I may get better resolution and montitor detection but I have not checked to see if they have enhanced that in 1.3 |
| 19:25.27 | BulletCatcher | At BRL there were 7 different players with *BSD clients in 2010. |
| 19:25.31 | BulletCatcher | None so far in 2011. |
| 19:25.34 | BulletCatcher | A closer look reveals that only 1 Solaris client was not me. |
| 19:26.39 | BulletCatcher | I think it is quite safe to maintain our emphasis on Windows, Mac, and Linux support. |
| 19:30.00 | trepan | and iphone/android :) |
| 19:34.47 | BulletCatcher | Yes. We should not change our emphasis on those platforms. |
| 19:36.37 | trepan | you semantic trickster, you |
| 19:36.58 | BulletCatcher | :-) |
| 19:56.27 | Erroneous | wow, there was a solaris client other than you? :) |
| 19:58.41 | BulletCatcher | Yup. SF patch 3152896 is for Solaris, even. |
| 19:58.43 | BulletCatcher | That may make 3 of us! |
| 19:58.57 | Erroneous | trepan: there was a significant (~10%) performance difference between SDL and native on windows last time I checked (perhaps due to audio). that was a couple years ago; we would want to retest that before using SDL on windows. |
| 19:59.20 | trepan | audio could go to OpenAL is well |
| 19:59.29 | spldart | Hrm... I should make it three.. I wonder if it would be fun to play with |
| 19:59.38 | Erroneous | yup |
| 19:59.41 | spldart | errh.. 4 |
| 20:04.37 | BulletCatcher | The fun has gone out of Solaris since Oracle bought Sun. |
| 20:04.39 | BulletCatcher | I can't really recommend getting started with Solaris at this point in history unless you are getting paid for it. |
| 20:10.05 | Pimpinella | has anyone tried to build bz on OpenBSD recently? |
| 20:10.40 | Pimpinella | bzfs alone would do |
| 20:14.05 | Pimpinella | though OpenBSD might be able to run the binux binary |
| 20:14.19 | Pimpinella | s/binux/linux/ |
| 20:16.16 | BulletCatcher | moiree: The --without-SDL problems are much worse when SDL is not actually present on the system. |
| 20:16.17 | BulletCatcher | I'm working on it. |
| 20:46.39 | *** join/#bzflag allejo (~allejo@cpe-76-95-144-121.socal.res.rr.com) |
| 21:07.42 | CIA-5 | BZFlag: 03bullet_catcher * r22009 10/trunk/bzflag/src/platform/XWindow.cxx: Include OpenGLGState.h first to ensure that glew.h is included before gl.h and glext.h. |
| 21:08.01 | BulletCatcher | That ought to fix your problem, moiree. |
| 21:17.47 | allejo | does anyone know what version of xcode the bzflag xcode project was made in? |
| 21:21.47 | BulletCatcher | The r16034 log message from 2007 starts with "update the project to XCode 2.4.1 .xcodeproj project file format". |
| 21:21.53 | BulletCatcher | It hasn't changed since then. |
| 21:22.34 | BulletCatcher | s/r16034/r16024/ |
| 21:23.12 | allejo | thanks BulletCatcher |
| 21:44.58 | CIA-5 | BZFlag: 03bullet_catcher * r22010 10/trunk/bzflag/src/ (7 files in 7 dirs): Add $CONF_CPPFLAGS to AM_CPPFLAGS as needed to use the bundled GLEW on Mac OS X in the absence of SDL. |
| 21:46.02 | allejo | ahhhh BulletCatcher, i think you may have just solved my compiling problem... *me tries an svn update* |
| 21:46.44 | BulletCatcher | MacWindow.cxx still needs to be fixed, but that will help. |
| 21:48.39 | allejo | hmmm do you know how you would compile a Mac binary? My binaries aren't executable on other computers because of ZeroLinks which was in Xcode 2.x, iirc. |
| 21:49.37 | BulletCatcher | I follow the unix command line instructions: autogen.sh && ./configure && make . |
| 21:50.02 | allejo | Yea I do that for myself, but I mean in the sense of making a .app for others to use. |
| 21:51.05 | BulletCatcher | Making a .app seems to be a deep mystery that is not well understood in this channel. |
| 21:51.49 | allejo | I know the reason why a .app can't be made for distribution i just don't know how to fix it. |
| 21:52.51 | allejo | It has to do with ZeroLinks. I'm not to familiar with that is but if I remember my Xcode 2.x history, it had to static libraries or something like that. ZeroLinks got deprecated with Xcode 3 |
| 21:53.22 | BulletCatcher | hasn't a clue |
| 21:53.27 | allejo | But since the .xcodeproj file is 2.x, it still involves ZeroLinks. The project file would have to be updated to Xcode 3 |
| 21:53.54 | allejo | or Xcode 4, but Xcode 3 is the last free Xcode so that should be ideal. |
| 21:54.27 | allejo | If the Xcode project file is updated for Xcode 3, an .app should be able to be built |
| 21:54.38 | BulletCatcher | Would it be easy to do that? |
| 21:54.52 | McSpider | allejo, this might help you: http://meandmarkpublishing.blogspot.com/2007/01/zerolink.html |
| 21:55.11 | allejo | That's where I'm at a loss. I'm not even sure how the current .xcodeproj project file was made |
| 21:56.36 | allejo | Anyone have Xcode 2? |
| 21:57.35 | BulletCatcher | I have Xcode 2.4.1 on my 10.4.11 MacBook. |
| 21:58.11 | allejo | Can you try building an .app there and make sure to disable ZeroLinks? |
| 21:58.27 | allejo | With the link that McSpider provided... |
| 21:58.54 | BulletCatcher | I use the command line, so I'm not sure what the .app procedure is. |
| 21:59.33 | allejo | can you open the .xcodeproj in Xcode 2.4.1? |
| 22:00.32 | BulletCatcher | Yes. I get a pretty window. |
| 22:01.02 | allejo | In the build menu, is there an option to "Allow ZeroLink" or something like that? |
| 22:02.06 | BulletCatcher | Yes, but it is checked and not clickable. |
| 22:02.29 | BulletCatcher | "grayed out" is the term, I think. |
| 22:02.58 | JeffM | "disabled" would be the proper term IIRC |
| 22:03.17 | allejo | Let's try this, in the pretty window that opened, toward the top left hand corner, there is a drop down menu called "Active Build Config" |
| 22:03.36 | allejo | Is there a "Release" or "Deployment" in the drop down menu? |
| 22:04.00 | BulletCatcher | It is set for Development, and has Deployment and Default too. |
| 22:04.13 | allejo | Select Deployment |
| 22:04.39 | allejo | Then check back in the Build menu to see if you can disable the "Allow ZeroLink" thing |
| 22:04.51 | BulletCatcher | Done. Allow ZeroLink is still gray. |
| 22:06.12 | JeffM | BulletCatcher, if you are fighting glew, you can just do what I did on windows, include the single C file in the compilation of the bzflag executable ;) |
| 22:06.33 | JeffM | it's just a thin wraper arround extensions |
| 22:06.48 | JeffM | it is even |
| 22:07.03 | allejo | BulletCatcher: on a side note, Glew is still an issue for building on Macs |
| 22:07.34 | BulletCatcher | GLEW is done, but non-SDL on the Mac needs a bit more love. I just need to backport from 2.99. |
| 22:07.55 | allejo | BulletCatcher: Let me continue looking into ZeroLink in Xcode 2, but that should be the reason for the troubles of building .apps on a Mac |
| 22:09.15 | BulletCatcher | k |
| 22:10.16 | JeffM | don't break the build for it |
| 22:10.22 | JeffM | the bin we ship will be SDL |
| 22:10.31 | JeffM | you can fix it in .2 if you want |
| 22:12.25 | JeffM | wants customized tank models that can look like this, http://3.bp.blogspot.com/_6jUmhUYDLTY/TL0VSMxT8NI/AAAAAAAACeI/yrV1w-GmQck/s1600/MetalSlugTank.jpg |
| 22:12.26 | BulletCatcher | The only problem I have right now is in MacWindow.cxx, which should be used only on Macs and only when there is no SDL. |
| 22:12.57 | BulletCatcher | So the release should not be affected. |
| 22:13.01 | JeffM | k |
| 22:13.06 | JeffM | well don't kill yourself over it :) |
| 22:13.34 | allejo | Will bzflag still be able to be built on a Mac though, I get this error -> http://pastebin.com/L8eqBn11 |
| 22:13.35 | BulletCatcher | Yeah, but right now I can't build on *my* Mac! |
| 22:14.35 | BulletCatcher | I can fix that, allejo. |
| 22:15.00 | allejo | oh good :) |
| 22:15.19 | BulletCatcher | Please do a "make -k" to see if there are similar problems in other directories. The fix is per directory. |
| 22:15.35 | JeffM | given the way apple is going we may have to stop building on the mac, unless we give apple 30% :) |
| 22:15.43 | BulletCatcher | It's just more of r22010. |
| 22:16.14 | BulletCatcher | I see no problem in giving them 30% of our sales revenue. |
| 22:16.32 | JeffM | who said anything about sales, I think they want 30% of everytyhing |
| 22:16.50 | JeffM | how many kids you got? cus if it's more then 3 they want one |
| 22:17.56 | BulletCatcher | It is going to be hard to type with only 76 keys on my keyboard. |
| 22:18.10 | JeffM | Gestures |
| 22:18.13 | JeffM | it's all about Gestures |
| 22:18.21 | allejo | BulletCatcher: currently doing "make -k" |
| 22:18.39 | BulletCatcher | I am thinking of a gesture, all right. |
| 22:18.44 | JeffM | heh |
| 22:18.56 | PlasticTank | BulletCatcher, you could always buy a new one from apple ;D |
| 22:18.58 | allejo | is it a hand gesture BulletCatcher? :P |
| 22:19.11 | JeffM | he has his own angry brid |
| 22:19.55 | allejo | lmao |
| 22:25.57 | allejo | BulletCatcher: here's my output with "make -k" do you want the entire output or only the errors? errors only -> http://pastebin.com/7Js53y0g |
| 22:28.56 | BulletCatcher | I need to know which directories contain files that have the error. |
| 22:28.57 | BulletCatcher | So far, I see only src/platform . |
| 22:32.38 | BulletCatcher | Any other directories? |
| 22:33.12 | allejo | This what you want? -> http://pastebin.com/j8JyaPVw |
| 22:33.28 | TimRiker | http://www.youtube.com/watch?v=9BnLbv6QYcA |
| 22:34.11 | BulletCatcher | It seems like src/platform is the only directory. Commit coming... |
| 22:35.37 | CIA-5 | BZFlag: 03bullet_catcher * r22011 10/trunk/bzflag/src/platform/Makefile.am: Add $CONF_CPPFLAGS to AM_CPPFLAGS one more place to use the bundled GLEW on Mac OS X in the absence of SDL. |
| 22:36.21 | allejo | lol TimRiker |
| 22:37.17 | allejo | BulletCatcher: building... |
| 22:39.10 | BulletCatcher | "make -j3" will make the best use of both cores and cut the compile time nearly in half. |
| 22:39.46 | allejo | error -> http://pastebin.com/rDtQS6vX |
| 22:41.24 | BulletCatcher | Yup. It isn't linking with the GLEW library we build. |
| 22:42.32 | allejo | so in order to compile, there would already have to be a GLEW library? |
| 22:43.58 | TD-Linux | JeffM: probably the first step to making customizable tank models would be to make the tanks... models :P |
| 22:44.01 | BulletCatcher | It should have already been built. |
| 22:44.06 | TD-Linux | rather than a sequence of opengl commands |
| 22:44.18 | JeffM | TD-Linux, yeah but that's not hard, I've done that before |
| 22:44.27 | JeffM | the V3 tanks is already a model |
| 22:44.43 | JeffM | well the tank model is a model it's just not hooked up |
| 22:44.45 | TD-Linux | V3 being? |
| 22:44.49 | JeffM | 2.99 |
| 22:45.00 | TD-Linux | last I looked at bzflag it was all opengl (2.0.x) |
| 22:45.10 | JeffM | yep |
| 22:45.14 | JeffM | I don't mean the code |
| 22:45.16 | JeffM | I mean the tank |
| 22:45.25 | TD-Linux | right now bzflag doesn't really support any geometry file formats other than bzw... should that change? |
| 22:45.27 | JeffM | http://my.bzflag.org/w/Image:3_0_tank_test_render.png |
| 22:45.36 | JeffM | yes have an obj loader |
| 22:45.45 | JeffM | a couple of em actualy |
| 22:46.01 | JeffM | I just pulled the code from 2.99 cus it wasn't finished |
| 22:46.04 | TD-Linux | seems like a much better idea to load the tank model from obj in game |
| 22:46.10 | JeffM | yes we will |
| 22:46.11 | TD-Linux | when you start adding stuff |
| 22:46.16 | JeffM | I know |
| 22:46.24 | JeffM | known that for a LONG While now |
| 22:46.37 | JeffM | thats why 2.99 has the "TankGeometryManager" |
| 22:47.38 | TD-Linux | could that be added to the roadmap in a 2.4.x release? |
| 22:48.13 | BulletCatcher | I need some time to catch up to where you are, allejo, so I can see the problem for myself. |
| 22:53.30 | JeffM | TD-Linux, it's implied with the customized tanks |
| 22:53.34 | JeffM | it is |
| 22:53.37 | JeffM | man I can't type today |
| 22:54.14 | mdskpr_ | it's = it is |
| 22:54.17 | JeffM | specifics like that would go on the design document page for the feature |
| 22:54.59 | JeffM | always gets them mixed up |
| 22:55.13 | mdskpr_ | it's the same thing... |
| 22:55.19 | mdskpr_ | it's a contraction |
| 22:55.21 | JeffM | its != it's |
| 22:55.24 | JeffM | that' what I mix up |
| 22:55.28 | mdskpr_ | its == it's |
| 22:55.32 | mdskpr_ | oh ic |
| 22:55.45 | JeffM | no its means "belonging to it" |
| 22:56.06 | mdskpr_ | goes studies english |
| 22:56.12 | JeffM | english is stupid |
| 22:57.00 | BulletCatcher | We prefer the unambiguity of c++. |
| 22:57.33 | mdskpr_ | i suppose that is why there are bugs... |
| 22:59.29 | JeffM | ahh that's what I was thinking of , PlayerAvatarManager |
| 23:00.05 | JeffM | it's suposed to give you back a render model for a specific tank, loaded from whereever |
| 23:00.10 | JeffM | so that's where we hook in model loading |
| 23:00.45 | JeffM | for customized tanks it would build up one for each tank from a model library |
| 23:01.00 | TD-Linux | ok |
| 23:01.06 | JeffM | it just never got done |
| 23:01.13 | JeffM | probalby not the best way to do it now |
| 23:01.17 | TD-Linux | it is a bit complicated, with tanks having moving parts |
| 23:01.32 | JeffM | not really, that's just transforms based on tank state |
| 23:01.50 | JeffM | that's what the scenenodes are for |
| 23:02.07 | TD-Linux | ok |
| 23:03.13 | JeffM | but that's another day |
| 23:04.34 | JeffM | be nice to just to use scripts for tank anims then they could be swaped out as needed |
| 23:04.39 | JeffM | then I can have spider tank :) |
| 23:09.07 | TD-Linux | also now that I'm spouting out random ideas |
| 23:10.18 | TD-Linux | rather than moving to a new font system, might it be better to go to a different GUI system? |
| 23:10.31 | TD-Linux | though I guess 2.99 has the new font system already worked in |
| 23:10.44 | JeffM | it does |
| 23:11.06 | TD-Linux | what are the outstanding bugs on that? |
| 23:11.08 | JeffM | a new GUI system would probably be evalied if/when we looked moving to a third party engine |
| 23:11.17 | JeffM | it's probalby ok |
| 23:11.44 | JeffM | most of the issues were with our GL state management or in FTGL and lot of that got fixed |
| 23:12.33 | JeffM | what GUI system did you have in mind? |
| 23:13.05 | TD-Linux | didn't really have one in mind, just knew that I didn't like BZFlag's :< |
| 23:13.14 | TD-Linux | CEGUI maybe though I've never used it |
| 23:13.22 | JeffM | the hard part with third party ones is getting them to work with with our state managment |
| 23:13.29 | JeffM | cus we do wierd crap |
| 23:13.36 | TD-Linux | I usually write GUIs with Qt which is a bit heavy for bzflag |
| 23:13.43 | JeffM | a tad |
| 23:15.47 | TD-Linux | oh, also, I've had to make HTML guis. what a terrible way to make them :( |
| 23:16.03 | JeffM | I've never seen a game use ether of those |
| 23:16.30 | TD-Linux | yeah I don't write games |
| 23:16.43 | JeffM | I can tell ;) |
| 23:18.35 | Pimpi | i usually use curses for guis ;) |
| 23:18.53 | Pimpi | and i've seen games using curses |
| 23:19.06 | BulletCatcher | nethack FTW |
| 23:20.09 | TD-Linux | I've never really seen the appeal of curses |
| 23:20.15 | TD-Linux | unless you are using a vt100 |
| 23:21.15 | JeffM | heh |
| 23:21.17 | BulletCatcher | Back in the day, a VT100 was top-of-the-line technology. |
| 23:21.22 | bryjen | finds curses handy when using windows ;) |
| 23:21.28 | *** join/#bzflag KTL (~KTL@213.219.154.76.adsl.dyn.edpnet.net) |
| 23:21.31 | mdskpr_ | lol |
| 23:21.39 | JeffM | I've used GEGUI, not a huge fan of it but it's functional |
| 23:21.42 | JeffM | and works with ogre |
| 23:22.00 | Pimpi | & |
| 23:23.24 | JeffM | well you are one of those people that want bzflag to be a 2d game |
| 23:23.26 | JeffM | ;) |
| 23:23.30 | JeffM | cus it mostly is |
| 23:25.41 | JeffM | so for the command line options, what thingy takes the argv and makes it so stuff with spaces and quotes get put into a single item? |
| 23:26.01 | BulletCatcher | The shell does that. |
| 23:26.03 | JeffM | cus I think the server does that but the client dosn't |
| 23:26.10 | JeffM | ok then that will NEVER work on windows |
| 23:26.16 | JeffM | cus it dosn't do that |
| 23:26.18 | BulletCatcher | Correct. |
| 23:26.30 | JeffM | then the -dir bug on windows isn't new |
| 23:26.36 | trepan | bzflag.cpp/WinMain() |
| 23:26.43 | trepan | (end of the file) |
| 23:26.57 | JeffM | aww do I do it manualy? |
| 23:27.00 | JeffM | man I suck |
| 23:27.27 | JeffM | we'll save that for later then :) |
| 23:27.29 | trepan | i remember something about premake automatically setting up a *nix'y main() for winders apps |
| 23:27.44 | JeffM | how does it disable the terminal then? |
| 23:27.54 | JeffM | usualy things that have a main show the terminal |
| 23:27.59 | trepan | i don't know, haven't really looked into it |
| 23:28.07 | JeffM | there may be a build flag |
| 23:28.17 | trepan | ya, was thinking along those lines as well |
| 23:28.22 | JeffM | but if we all use SDL we can just use SDLMain |
| 23:29.09 | trepan | something like: EntryPointSymbol="mainCRTStartup" ? |
| 23:29.19 | JeffM | dunno |
| 23:29.28 | trepan | that's in the BZ/3.0/other_src/premake/src/actions/vstudio/*.lua files |
| 23:29.28 | JeffM | the compiler has all sorts of wierd crap |
| 23:31.29 | JeffM | is not sure why anyone would even need to use -directory on windows |
| 23:31.45 | JeffM | it dosn't use the working dir to find it |
| 23:32.22 | trepan | to devtest without installing? |
| 23:32.29 | JeffM | you don't have to install |
| 23:32.40 | JeffM | bz will run fine from the bin dir |
| 23:32.47 | JeffM | it uses the exe path and looks around it |
| 23:32.54 | JeffM | so it'll run fine from the build dir |
| 23:33.13 | JeffM | thats why it's no big deal to have multiple installs on windows |
| 23:33.16 | JeffM | they are all atomic |
| 23:33.16 | trepan | so what's the working dir? |
| 23:33.21 | JeffM | whatever |
| 23:33.22 | trepan | not the same as the exec path? |
| 23:33.28 | JeffM | we don't use the working dir |
| 23:33.41 | JeffM | when you run from the shortcut the working dir is wherever the shortcut is |
| 23:34.06 | JeffM | if you double click an exe it's where ever the exe is |
| 23:34.19 | JeffM | I use a windows function to get the path to the running module where ever it is |
| 23:34.32 | JeffM | then look arround that for a data dir |
| 23:34.50 | JeffM | it looks in the exe dir first, then up one, then the working dir |
| 23:34.58 | JeffM | installed versions find it in the first one |
| 23:35.06 | JeffM | debuging from the ide finds it in the second |
| 23:35.16 | JeffM | or double clicks from the bin dir |
| 23:35.37 | JeffM | I just ignore the working dir cus I can't trust it to be set to anything I care about |
| 23:35.56 | JeffM | "Ohh your deskop.. THAT's usefull" :) |
| 23:36.31 | JeffM | I belive the mac app does a simialr thing and looks in the .app resource dir first |
| 23:36.43 | JeffM | thats why you can drag it arround all over the place |
| 23:37.21 | JeffM | linux is the only thing that has a "install" step that affects that |
| 23:37.59 | trepan | in *nix land, you can't really trust argv[0] |
| 23:38.14 | JeffM | I can undersand why |
| 23:38.36 | trepan | because execl() and friends can set it to whatever they want |
| 23:38.37 | *** join/#bzflag Anxuiz (~Anxuiz@c-76-22-102-23.hsd1.wa.comcast.net) |
| 23:38.37 | *** join/#bzflag Anxuiz (~Anxuiz@unaffiliated/anxuiz) |
| 23:38.40 | *** join/#bzflag Anxuiz__ (~Anxuiz@c-76-22-102-23.hsd1.wa.comcast.net) |
| 23:38.46 | JeffM | GetModuleFileName(NULL, temppath, MAX_PATH); always gives you the path to the current executable module, .exe or dll |
| 23:38.58 | JeffM | it's the file that the OS loaded |
| 23:47.33 | trepan | hm, wonder how reliable /proc/<PID>/exe is |
| 23:47.41 | trepan | (or something else in /proc) |
| 23:49.05 | trepan | was trying to convince fexecve() to work from a pipe, it didn't like that idea at all :) |
| 23:49.53 | TD-Linux | wait Windows doesn't support arguments containing spaces? |
| 23:51.06 | TD-Linux | this would explain why a lot of the windows IDE's I've used have random space-in-file-path failures |
| 23:51.16 | TD-Linux | and sadly it seems to be the commercial expensive ones too :S |
| 23:52.14 | trepan | thinks that whoever allowed spaces in filenames (way back when), should be flogged |
| 23:54.22 | TD-Linux | it becomes especially painful when "Documents and Settings" means that anything in your home folder triggers the bugs |
| 23:56.31 | trepan | can you use some sort of hard/sym link? |
| 23:56.43 | JeffM | TD-Linux, it's up to the app to do the parsing, most these days do it |
| 23:56.52 | JeffM | trepan, no symlinks in windows untill 8 |
| 23:57.06 | blast007 | BulletCatcher: I don't have a history of clients, unless logDetail keeps that |
| 23:57.17 | trepan | JeffM: ouch, welcome to 1990 :) |
| 23:57.36 | JeffM | yeah that's one thing that is a bit of a pain |
| 23:58.02 | JeffM | windows7 got all the APIs for it in place IIRC with the VFS stuff |
| 23:58.13 | JeffM | tho usualy I don't care about the layout of my disk |
| 23:58.35 | JeffM | so it dosn't actualy affect me in my day to day |
| 23:58.54 | JeffM | but there have been a few cases wer have found where it would be nice |
| 23:59.09 | JeffM | TD-Linux, what IDE do you use? |
| 23:59.19 | BulletCatcher | It looks like client version strings are not kept by logDetail, blast007. It's not a big deal. |
