From god.000 at gmail.com Sat Dec 3 00:52:07 2005 From: god.000 at gmail.com (Almighty Tallest Cataboligne) Date: Sat, 3 Dec 2005 07:52:07 +0000 Subject: [conquest] Conquest GL client graphics upgrade beta test. Message-ID: To all list members: Take heed, a new era of conquest game play may soon be at hand! For the last several months I have been crafting a graphics update for the conquest GL client. The first gfx goal was adding custom planet images to an upgrade that loaded planet data from a .planetsrc text file upon universe reset. A quick discovery lead to a project to obtain trek ship and planet images. It wasnt long before the images and gfx needs outstripped the purpose of .planetsrc. At this point a .imagesrc was coded with specific animation, scaling and other controls. Not long after I had the core ideas for the full gfx upgrade that now exists. (codes: TOS = the original series, TMP - the motion pictures, TNG = the next gen...) The following goals have been adhered to: usage of TOS actual images, scans, art, and images crafted from TOS data. options to switch on many of these features. replacement of ship images from art crafted based on original models. (very few useable overhead [dorsal] model photos exist.) replacement of photon shots, explosions, and phaser shot from TOS for fed, kli & rom. Orion gfx for this set are conquest originals. inclusion of optional shield visual indicator. this indicator has been designed to correspond as much as possible with available canon trek information. additionaly an option is added to give the indicator a layer effect. an optional tactical background (used with LR scan display) based on TOS and TMP images where possible with filler created by artistic license (i.e. I put in what felt right to make the image work.) there is also an option to shade the background to user requirements. (diff. monitors seem to show this particular update rather differently.) a multi level scaling system. this allows 3 rescalings: scaling is programmed with 10% - 1000% limits. from 50% - 200% is most practical for game play. 1 - HUD scaling. the hud can be scaled up or down and locked at a given scale. (this is especially nice on larger montiors where the up scale makes HUD gfx huge...) This scaling does not change the LR scan or main view scale, however, more will be viewed as the HUD is scaled down and less as the HUD is scaled up. 2 - Main viewer scaling - this allows the main view window to be scaled up and down higher resolution images provide more detailed gfx at larger scales. I added this mostly to compensate for lack of detail with the new gfx set in smaller windows and lower res. monitors. This scaling does not affect the LR scan view. 3 - Ship scaling - the icons for the ships in the main view can be scaled independently of the viewer. I added this just for bringing out details of ship gfx without changing the other aspects of viewer images. the entire HUD has been iconized. an option exists to use the old hud, new hud with old number boxes, new hud only and new hud with hints from old hud labels. (hints are for possible unreadable fed icons at some gfx levels, and translation of race text for non feds.) a ship and shield icon have been added to the hud. an option exists to display conquest status colors with these icons (green = OK, yell = warning, red = trouble) different races have race specific fonts in HUD icons. gauges have quantity lables. each ship icon has a set of alert decals corresponding to: cloak engaged (ship icon is also shaded) armies carried (includes count) tractor beam (also indicates ship in tow) ship under repairs photon load status (displays available torps for firing) engines critical (eng temp in red zone) weapons critical (wep temp in red zone) shield critical (shield in red zone) shield fail (shield < 4 units) hull critical (damage in red zone) warp core breach (damage > 95 units) alert status decals have been added corresponding to: red alert, yellow alert, proximity alert - fed is based on TOS images, and the rest are artistic license. heading has an astrogator icon added with a direction pointer. the fed astrog. is based on the TOS helm console. the other teams are artistic license. the warp indicator also has a chart added the displays a visual indication' of warp speed. the fed warp chart is based on the TOS engineers console on the bridge. the other teams are artistic license. the intro giant "conquest" title has been converted into race specific fonts. the orion team has the most non canon gfx...simply because TOS and TMP contain almost nothing in the way of orion images. Klingon and romulan gfx owe a lot to TMP and the Enterprise show. I hope to have an upgrade package ready for beta testing soon. I'll prob. let Jon take a stab at it prior to general release. I anticipate scaling issues on other OS and monitor / gfx card configs. (I have morphix linux 4.x and dual 19" monitors and a geforce 6600GT [epci].) further plans in sequence: convert any images not already in power of 2 sizes to such. shading on alert decals so shield indicator doesnt swamp them. improved and optimized hud code. live key codes to switch viewer scale and some hud opts so you dont have to call up opts menu. option to mask viewer edges beyond tactical bkg implied viewscreen and possibly reconfigure viewer window to accomodate a more "accurate" viewscreen ideal. smaller size image sets icon text labels converted over the .txf font sets if possible. (this will also make the alert decal hints better...) possibly improving the option set expansion in some way (just no ideas yet.) considering: investigating making the hud a separate "floating" window. I have 2 monitors, so this makes sense to me. also thinking of an total integrated HUD / viewer. HUD data would be arrayed around the viewer in some TOS style fashion with the alert decals / ship icon being a floater... In general any suggestions for improvement will be considered. Here are a few guidelines - Suggestions will be accepted for: scaling and display bugs. ideas for better contrast of icons and text. possible image sources that are (proveable) canon, where I used artistic license. (specifically *detailed* set photos for TOS, TMP, TNG - all races, that better icons could be made from.) Things of aesthetic design or low priority that I'm not interested in changing: layout of icons, alert decals. spacing of gauges, etc. (I followed the old HUD as closely as possible to avoid confusion - if you dont like something or it interferes with _your_ game, well thats why there are options to turn stuff off.) If you dont like the way something looks, I will be providing the source material so you can modify it to your hearts content. Hopefully this will cut down on replies that go like "thats not a bug" or "thats not a design flaw" because I designed it to be that way. What gets included in the official release will be up to Jon since he is the code base maintainer. I hope to have a beta ready within 2 weeks. A few things still need to be looked at. -Roy ---------------------------------------------- 3 4 | 2 \|/ 5--*----> 1 Thing is facing this direction /|\ 6 | 8 7 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jon at radscan.com Sat Dec 3 23:32:03 2005 From: jon at radscan.com (Jon Trulson) Date: Sat, 3 Dec 2005 23:32:03 -0700 (MST) Subject: [conquest] Conquest GL client graphics upgrade beta test. In-Reply-To: References: Message-ID: On Sat, 3 Dec 2005, Almighty Tallest Cataboligne wrote: > To all list members: > > Take heed, a new era of conquest game play may soon be at hand! > > Based on the screenshots you sent, I definitely look forward to the patch! :) BTW: I would recommend making all the textures power-of-two, or many cards/drivers will not be able display them... > For the last several months I have been crafting a graphics update > for the conquest GL client. > [...] > I hope to have an upgrade package ready for beta testing soon. I'll prob. > let Jon take > a stab at it prior to general release. > Thanks! I definitely want to check out the mods and how they will need to factor into the new common block/init stuff. > I anticipate scaling issues on other OS and monitor / gfx card configs. > (I have morphix linux 4.x and dual 19" monitors and a geforce 6600GT > [epci].) > This will be something to check out for sure.. I've got Radeon 9000, 8500 and Intel i915 with XiG Summit drivers. > further plans in sequence: > convert any images not already in power of 2 sizes to such. Please do this soon :) [...] > What gets included in the official release will be up to Jon since he is the > code base maintainer. > Once it's been checked out a little, I'd definitely make a new dev release with this stuff in it... > I hope to have a beta ready within 2 weeks. A few things still need to be > looked at. > I will patiently wait by my computer :) -- Jon Trulson mailto:jon at radscan.com ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962 PGP keys at http://radscan.com/~jon/PGPKeys.txt #include "I am Nomad." -Nomad From jon at radscan.com Sun Dec 4 12:44:07 2005 From: jon at radscan.com (Jon Trulson) Date: Sun, 4 Dec 2005 12:44:07 -0700 (MST) Subject: [conquest] Conquest GL client graphics upgrade beta test. In-Reply-To: References: Message-ID: On Sat, 3 Dec 2005, Jon Trulson wrote: > On Sat, 3 Dec 2005, Almighty Tallest Cataboligne wrote: > >> To all list members: >> >> Take heed, a new era of conquest game play may soon be at hand! >> >> > > Based on the screenshots you sent, I definitely look forward to the > patch! :) > I've put up some of the screen shots at: http://www.radscan.com/~jon/gallery/conquest-newart/ [...] -- Jon Trulson mailto:jon at radscan.com ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962 PGP keys at http://radscan.com/~jon/PGPKeys.txt #include "I am Nomad." -Nomad From johnclute at imagi.net Mon Dec 5 07:27:59 2005 From: johnclute at imagi.net (John) Date: Mon, 5 Dec 2005 07:27:59 -0700 Subject: [conquest] Conquest GL client graphics upgrade beta test. References: Message-ID: <000b01c5f9a8$17fa51b0$0401a8c0@spaceship1> Wow... They look really cool. ----- Original Message ----- From: "Jon Trulson" To: "Almighty Tallest Cataboligne" Cc: "Conquest" Sent: Sunday, December 04, 2005 12:44 PM Subject: Re: [conquest] Conquest GL client graphics upgrade beta test. > On Sat, 3 Dec 2005, Jon Trulson wrote: > >> On Sat, 3 Dec 2005, Almighty Tallest Cataboligne wrote: >> >>> To all list members: >>> >>> Take heed, a new era of conquest game play may soon be at hand! >>> >>> >> >> Based on the screenshots you sent, I definitely look forward to the >> patch! :) >> > > I've put up some of the screen shots at: > > http://www.radscan.com/~jon/gallery/conquest-newart/ > > [...] > > > -- > Jon Trulson mailto:jon at radscan.com > ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962 > PGP keys at http://radscan.com/~jon/PGPKeys.txt > #include > "I am Nomad." -Nomad > > From god.000 at gmail.com Sat Dec 10 01:16:01 2005 From: god.000 at gmail.com (Almighty Tallest Cataboligne) Date: Sat, 10 Dec 2005 08:16:01 +0000 Subject: [conquest] Conquest GL client graphics upgrade beta test. In-Reply-To: References: Message-ID: > To all list members: > > Take heed, a new era of conquest game play may soon be at hand! > > For the last several months I have been crafting a graphics update > for the conquest GL client. Not sure how many players are part of this list, but... I have been using my server as the test platform for the new gfx upgrade. This means I frequently restore it to a controlled set of conditions...thus it is not much fun to play on. Discovering recently that a couple people I have tried, I have set the closed condition until testing is done. Hope this doenst cause much inconvenience. ---------------------------------------------- 3 4 | 2 \|/ 5--*----> 1 Thing is facing this direction /|\ 6 | 8 7 -------------- next part -------------- An HTML attachment was scrubbed... URL: From cyberbillp at yahoo.com Sat Dec 17 15:05:14 2005 From: cyberbillp at yahoo.com (Bill Patterson) Date: Sat, 17 Dec 2005 14:05:14 -0800 (PST) Subject: [conquest] client lockup Message-ID: <20051217220514.55706.qmail@web60316.mail.yahoo.com> I have a problem where the client freezes, then after you kill the client and return to the game, the game says you are already flying another ship. The only way to fix it is to use Conqoper and "k"ill the player. This sux because it locks out the player until I get online to fix them. Shouldn't the server detect that the player is gone and kill the ship or something? Bill Patterson cyberbillp at yahoo.com Owner of: http://ultramaxx.com __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com From dwpayne at gravitic.com Sat Dec 17 16:47:17 2005 From: dwpayne at gravitic.com (David W. Payne) Date: Sat, 17 Dec 2005 16:47:17 -0700 Subject: [conquest] client lockup In-Reply-To: <20051217220514.55706.qmail@web60316.mail.yahoo.com> References: <20051217220514.55706.qmail@web60316.mail.yahoo.com> Message-ID: <43A4A385.2040001@gravitic.com> It would help if you told us what version conquest you are running. On what OS and how you are running the conquestd daemon or conqdriv if you are using an older version of conquest. dave Bill Patterson wrote: >I have a problem where the client freezes, then after >you kill the client and return to the game, the game >says you are already flying another ship. The only way >to fix it is to use Conqoper and "k"ill the player. >This sux because it locks out the player until I get >online to fix them. Shouldn't the server detect that >the player is gone and kill the ship or something? > >Bill Patterson >cyberbillp at yahoo.com > >Owner of: >http://ultramaxx.com > >__________________________________________________ >Do You Yahoo!? >Tired of spam? Yahoo! Mail has the best spam protection around >http://mail.yahoo.com > > > -- #include From god.000 at gmail.com Sat Dec 17 21:05:17 2005 From: god.000 at gmail.com (Almighty Tallest Cataboligne) Date: Sun, 18 Dec 2005 04:05:17 +0000 Subject: [conquest] client lockup In-Reply-To: <43A4A385.2040001@gravitic.com> References: <20051217220514.55706.qmail@web60316.mail.yahoo.com> <43A4A385.2040001@gravitic.com> Message-ID: > > >I have a problem where the client freezes, then after > >you kill the client and return to the game, the game > >says you are already flying another ship. The only way > >to fix it is to use Conqoper and "k"ill the player. > >This sux because it locks out the player until I get > >online to fix them. Shouldn't the server detect that > >the player is gone and kill the ship or something? Normally you can leave a ship vacant in the game and then return to it. There is an issue that can stop this up. You get a different slot when you recconnect. Connect and DONT enter the game. Then look at the '\' list in conqoper. You will have a new slot. When you hit 'e' on your client to enter it switches you to the vacant slot. Sometimes it doesnt want to switch over and just tells you that you are already flying a ship in the game. I have gotten same ship re-entry when this condition has occured, but I'm not really sure how I cleared it up. Also...you may have multiple set. If you have the multiple flag set for your user it will always give you a new ship up to your multiple limit. I've never tested to see if it allows a re-entry once this limit is set. -Roy- ---------------------------------------------- 3 4 | 2 \|/ 5--*----> 1 Thing is facing this direction /|\ 6 | 8 7 -------------- next part -------------- An HTML attachment was scrubbed... URL: From jon at radscan.com Sat Dec 17 23:05:00 2005 From: jon at radscan.com (Jon Trulson) Date: Sat, 17 Dec 2005 23:05:00 -0700 (MST) Subject: [conquest] client lockup In-Reply-To: <20051217220514.55706.qmail@web60316.mail.yahoo.com> References: <20051217220514.55706.qmail@web60316.mail.yahoo.com> Message-ID: On Sat, 17 Dec 2005, Bill Patterson wrote: > I have a problem where the client freezes, then after > you kill the client and return to the game, the game Playing locally or remotely? Your server or someone else's? > says you are already flying another ship. The only way > to fix it is to use Conqoper and "k"ill the player. Ah.. I tried playing your sever last night w/o much luck, so that was probably me you had to kill off. First I tried default UDP. I connected, UDP was negotiated successfully. I entered the game, and wound up just sitting on Murisak. All TCP proto was working but no UDP (ship movement/status updates). I ^\ to exit (and in theory go vacant). But for some reason your server never got the CPCMD_DISCONNECT packet. When I reconnected, I could not get back in because I was 'live'. I then tried TCP only (-u option). This time I could enter and fly around. For awhile. Then the server stopped sending updates (no packets whatsoever). You server seemed to ignore all packets I sent to it (raise shields, etc). I ^\ again. Again I could not reconnect due to being 'live'. I have not seen this problem with a server before - I wonder if there is some firewall trouble? I came in as orbit (UDP) and orbit2 (TCP only). Did anyone else try this server and see this problem? Anybody ever seen that on my server(s)? > This sux because it locks out the player until I get > online to fix them. Shouldn't the server detect that > the player is gone and kill the ship or something? > Yes this does suck (especially since it was me :), but it's not a problem I've seen before. Yes, the server should detect this, and always has in the past. When the client exits, a CPCMD_DISCONNECT packet is sent, which causes the ship to go vacant (if allowed). In addition, even if that packet got lost (it's a TCP so it should never get lost) or was never sent (the client crashed), the server will still set you vacant on a socket read error - which it should have received because I was no longer connected. So in short, what was happening, should not have. It seemed like your firewall just 'froze' the connection(s). UDP immediately on entering, and TCP (about 10-15 minutes later). So your server seems to stop receiving packets from the client, stop sending them to the client, and not detect when the client is no longer connected. Very strange. I'm game to try to help figure this out, but it should not be happening in the first place :) -- Jon Trulson mailto:jon at radscan.com ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962 PGP keys at http://radscan.com/~jon/PGPKeys.txt #include "I am Nomad." -Nomad From jon at radscan.com Sat Dec 17 23:07:45 2005 From: jon at radscan.com (Jon Trulson) Date: Sat, 17 Dec 2005 23:07:45 -0700 (MST) Subject: [conquest] client lockup In-Reply-To: References: <20051217220514.55706.qmail@web60316.mail.yahoo.com> <43A4A385.2040001@gravitic.com> Message-ID: On Sun, 18 Dec 2005, Almighty Tallest Cataboligne wrote: >> >>> I have a problem where the client freezes, then after >>> you kill the client and return to the game, the game >>> says you are already flying another ship. The only way >>> to fix it is to use Conqoper and "k"ill the player. >>> This sux because it locks out the player until I get >>> online to fix them. Shouldn't the server detect that >>> the player is gone and kill the ship or something? > > > Normally you can leave a ship vacant in the game and then return to it. > > There is an issue that can stop this up. > > You get a different slot when you recconnect. > Connect and DONT enter the game. Then look at the '\' list in conqoper. > You will have a new slot. When you hit 'e' on your client to enter it > switches > you to the vacant slot. Sometimes it doesnt want to switch over and just > tells you that you are already flying a ship in the game. I have gotten > same > ship re-entry when this condition has occured, but I'm not really sure how > I cleared it up. Yes, you get a temp slot on login. A real slot when entering for battle. This is normal. I have not seen the above mentioned problem before. > > Also...you may have multiple set. If you have the multiple flag set for > your user > it will always give you a new ship up to your multiple limit. I've never > tested to see > if it allows a re-entry once this limit is set. > Don't mess with multiples... No need :) Theyt do not work properly anyway and should be removed. -- Jon Trulson mailto:jon at radscan.com ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962 PGP keys at http://radscan.com/~jon/PGPKeys.txt #include "I am Nomad." -Nomad From god.000 at gmail.com Sat Dec 24 23:51:48 2005 From: god.000 at gmail.com (Almighty Tallest Cataboligne) Date: Sun, 25 Dec 2005 06:51:48 +0000 Subject: [conquest] client lockup In-Reply-To: References: <20051217220514.55706.qmail@web60316.mail.yahoo.com> <43A4A385.2040001@gravitic.com> Message-ID: Merry Christmas to all. I have gotten sounds to work in the glclient. Jon isnt ready for sound code, but we have discussed unofficial releases. Thinking of getting one up when I have the time so I can gather other input on sound clips and such. Yeah, its not much of a present...problem is "a christmas carol" has been taking a lot of my time. I've seen Scrooge get the christmas spirit at least 15 times. -Roy- ---------------------------------------------- 3 4 | 2 \|/ 5--*----> 1 Thing is facing this direction /|\ 6 | 8 7 -------------- next part -------------- An HTML attachment was scrubbed... URL: From cyberbillp at yahoo.com Tue Dec 27 09:49:22 2005 From: cyberbillp at yahoo.com (Bill Patterson) Date: Tue, 27 Dec 2005 08:49:22 -0800 (PST) Subject: [conquest] full screen Message-ID: <20051227164922.36504.qmail@web60324.mail.yahoo.com> Ooops. Full screen hack doesn't work with the server browser... Bill Patterson cyberbillp at yahoo.com Owner of: http://ultramaxx.com __________________________________________ Yahoo! DSL ? Something to write home about. Just $16.99/mo. or less. dsl.yahoo.com From jon at radscan.com Tue Dec 27 13:04:25 2005 From: jon at radscan.com (Jon Trulson) Date: Tue, 27 Dec 2005 13:04:25 -0700 (MST) Subject: [conquest] full screen In-Reply-To: <20051227164922.36504.qmail@web60324.mail.yahoo.com> References: <20051227164922.36504.qmail@web60324.mail.yahoo.com> Message-ID: On Tue, 27 Dec 2005, Bill Patterson wrote: > Ooops. Full screen hack doesn't work with the server browser... > Interesting... seems to fail with playback (of .cqr's) too. '-f' is working fine, but -geometry seems to get confused. I'll investigate. > Bill Patterson > cyberbillp at yahoo.com > > Owner of: > http://ultramaxx.com > > > > __________________________________________ > Yahoo! DSL ? Something to write home about. > Just $16.99/mo. or less. > dsl.yahoo.com > -- Jon Trulson mailto:jon at radscan.com ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962 PGP keys at http://radscan.com/~jon/PGPKeys.txt #include "I am Nomad." -Nomad From jon at radscan.com Tue Dec 27 14:34:58 2005 From: jon at radscan.com (Jon Trulson) Date: Tue, 27 Dec 2005 14:34:58 -0700 (MST) Subject: [conquest] full screen In-Reply-To: References: <20051227164922.36504.qmail@web60324.mail.yahoo.com> Message-ID: On Tue, 27 Dec 2005, Jon Trulson wrote: > On Tue, 27 Dec 2005, Bill Patterson wrote: > >> Ooops. Full screen hack doesn't work with the server browser... >> > > Interesting... seems to fail with playback (of .cqr's) too. '-f' is > working fine, but -geometry seems to get confused. I'll investigate. > Ok... there seemed to be some confusion with the args and gluts parsing of them. So I removed the -geometry hack and and just implemented a '-g WxH' option. This seems to work at all times :) Please try the attached patch (using -g instead of -geometry) and see if that works. This patch also incorporates the conquestd futex wait hang workaround. You should apply this to virgin 8.1.1 srcs. There is also a patch that removes some ancient (but mostly harmless) cruft I stumbled upon in conquest.h - we aren't trying to use X11/Xt/Motif for the GUI anymore :) -- Jon Trulson mailto:jon at radscan.com ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962 PGP keys at http://radscan.com/~jon/PGPKeys.txt #include "I am Nomad." -Nomad -------------- next part -------------- Index: conquestgl.c =================================================================== RCS file: /home/jon/src/repository/conquest/conquestgl.c,v retrieving revision 1.9 retrieving revision 1.11 diff -u -r1.9 -r1.11 --- conquestgl.c 14 Nov 2004 22:34:04 -0000 1.9 +++ conquestgl.c 27 Dec 2005 21:10:41 -0000 1.11 @@ -2,7 +2,7 @@ /************************************************************************ * - * $Id: conquestgl.c,v 1.9 2004/11/14 22:34:04 jon Exp $ + * $Id: conquestgl.c,v 1.11 2005/12/27 21:10:41 jon Exp $ * * Copyright 1999-2004 Jon Trulson under the ARTISTIC LICENSE. (See LICENSE). ***********************************************************************/ @@ -53,7 +53,12 @@ void printUsage() { printf("Usage: conquestgl [-m ][-s server[:port]] [-r recfile] [ -t ]\n"); - printf(" [ -M ] [ -u ]\n\n"); + printf(" [-f ][-g ] [ -u ]\n"); + printf(" [ -M ]\n\n"); + printf(" -f run in fullscreen mode\n"); + printf(" -g specify intial window width/height.\n"); + printf(" Format is WxH (ex: 1024x768).\n"); + printf(" Default is 800x600.\n"); printf(" -m query the metaserver\n"); printf(" -s server[:port] connect to at \n"); printf(" default: localhost:1701\n"); @@ -90,6 +95,44 @@ return TRUE; } +/* parse the geometry arg. Ensure that dConf init w/h are only + * set if the geom arg was reasonably valid. Format WxH (ex: 1024x768) + */ +static void parseGeometry(char *geom) +{ + char geomcpy[32]; + char *ch; + int w, h; + + if (!geom || !*geom) + return; + + memset((void *)geomcpy, 0, 32); + strncpy(geomcpy, geom, 32 - 1); + + if ((ch = strchr(geomcpy, 'x')) == NULL) + return; /* invalid */ + + *ch = 0; + ch++; + + if (!*ch) + return; + + w = abs(atoi(geomcpy)); + h = abs(atoi(ch)); + + if (!w || !h) + return; + + /* set it up */ + dConf.initWidth = w; + dConf.initHeight = h; + + return; +} + + /* conquest - main program */ int main(int argc, char *argv[]) { @@ -121,10 +164,18 @@ cInfo.remotehost = strdup("localhost"); /* default to your own server */ + dspInitData(); + /* check options */ - while ((i = getopt(argc, argv, "mM:s:r:tP:d:u")) != EOF) /* get command args */ + while ((i = getopt(argc, argv, "fmM:s:r:tP:d:ug:")) != EOF) /* get command args */ switch (i) { + case 'f': + dConf.fullScreen = TRUE; + break; + case 'g': /* to let '-geometry' slide by *HACK* */ + parseGeometry(optarg); + break; case 'm': wantMetaList = TRUE; break; Index: gldisplay.h =================================================================== RCS file: /home/jon/src/repository/conquest/gldisplay.h,v retrieving revision 1.7 retrieving revision 1.9 diff -u -r1.7 -r1.9 --- gldisplay.h 30 May 2005 08:22:47 -0000 1.7 +++ gldisplay.h 27 Dec 2005 21:10:42 -0000 1.9 @@ -1,6 +1,6 @@ /************************************************************************ * - * $Id: gldisplay.h,v 1.7 2005/05/30 08:22:47 jon Exp $ + * $Id: gldisplay.h,v 1.9 2005/12/27 21:10:42 jon Exp $ * * Copyright 2003 Jon Trulson under the ARTISTIC LICENSE. (See LICENSE). ***********************************************************************/ @@ -37,6 +37,8 @@ unsigned int flags; + int fullScreen; + int initWidth, initHeight; /* initial wxh geometry */ } dspConfig_t; #ifdef NOEXTERN_DCONF @@ -235,5 +237,6 @@ void drawQuad(GLfloat x, GLfloat y, GLfloat w, GLfloat h, GLfloat z); void drawExplosion(GLfloat x, GLfloat y, int snum, int torpnum); +void dspInitData(void); #endif /* _GLDISPLAY_H */ Index: GL.c =================================================================== RCS file: /home/jon/src/repository/conquest/GL.c,v retrieving revision 1.23 retrieving revision 1.25 diff -u -r1.23 -r1.25 --- GL.c 10 Aug 2005 15:51:37 -0000 1.23 +++ GL.c 27 Dec 2005 21:10:41 -0000 1.25 @@ -2,7 +2,7 @@ * * Jon Trulson, 1/2003 * - * $Id: GL.c,v 1.23 2005/08/10 15:51:37 jon Exp $ + * $Id: GL.c,v 1.25 2005/12/27 21:10:41 jon Exp $ * * Copyright 2003 Jon Trulson under the ARTISTIC LICENSE. (See LICENSE). */ @@ -37,7 +37,6 @@ #include "glmisc.h" #include "glfont.h" -#define NOEXTERN #include "conquest.h" /* torp animation state */ @@ -918,13 +917,16 @@ } -static void dspInitData(void) +void dspInitData(void) { memset((void *)&dConf, 0, sizeof(dspConfig_t)); + dConf.inited = False; + + dConf.fullScreen = FALSE; + dConf.initWidth = 800; + dConf.initHeight = 600; dConf.wX = dConf.wY = 0; - dConf.wW = 800; - dConf.wH = 600; memset((void *)&dData, 0, sizeof(dspData_t)); @@ -956,20 +958,19 @@ #ifdef DEBUG_GL clog("uiGLInit: ENTER"); #endif - memset(&ConqData, 0, sizeof(ConqData)); - - dConf.inited = False; - - dspInitData(); glutInit(argc, argv); glutInitDisplayMode(GLUT_DEPTH | GLUT_DOUBLE | GLUT_RGBA | GLUT_ALPHA); glutInitWindowPosition(0,0); - glutInitWindowSize(dConf.wW, dConf.wH); + + glutInitWindowSize(dConf.initWidth, dConf.initHeight); dConf.mainw = glutCreateWindow(CONQUESTGL_NAME); + if(dConf.fullScreen) + glutFullScreen(); + glutKeyboardFunc (charInput); glutSpecialFunc (input); glutMouseFunc (mouse); Index: conquest.h =================================================================== RCS file: /home/jon/src/repository/conquest/conquest.h,v retrieving revision 1.2 retrieving revision 1.3 diff -u -r1.2 -r1.3 --- conquest.h 2 Nov 2003 20:53:00 -0000 1.2 +++ conquest.h 27 Dec 2005 21:10:41 -0000 1.3 @@ -1,7 +1,7 @@ /************************************************************************ * - * $Id: conquest.h,v 1.2 2003/11/02 20:53:00 jon Exp $ + * $Id: conquest.h,v 1.3 2005/12/27 21:10:41 jon Exp $ * * Copyright 2003 Jon Trulson under the ARTISTIC LICENSE. (See LICENSE). ***********************************************************************/ @@ -9,49 +9,6 @@ #ifndef CONQUEST_H #define CONQUEST_H -#include -#include -#include - -#ifdef NOEXTERN -#define C2EXTERN -#else -#define C2EXTERN extern -#endif - #define CONQUESTGL_NAME "ConquestGL" -typedef struct _ConquestDataRec *ConquestDataPtr; - -typedef struct _ConquestDataRec { - Display *dpy; - XtAppContext app; - XtIntervalId timerId; - XtWorkProcId workId; - - Bool rendering; /* are we rendering? */ - - Widget toplevel; - - /* some colors */ - Pixel whitePixel; - Pixel blackPixel; - Pixel redPixel; - Pixel yellowPixel; - Pixel greenPixel; - Pixel bluePixel; - Pixel cyanPixel; - Pixel magentaPixel; - - Pixel defaultPixel; - Pixel infoPixel; - Pixel specialPixel; - Pixel statsepPixel; - -} ConquestDataRec; - - -C2EXTERN ConquestDataRec ConqData; - -#undef C2EXTERN #endif Index: conquestd.c =================================================================== RCS file: /home/jon/src/repository/conquest/conquestd.c,v retrieving revision 1.26 retrieving revision 1.27 diff -u -r1.26 -r1.27 --- conquestd.c 4 Jul 2005 03:18:22 -0000 1.26 +++ conquestd.c 25 Dec 2005 20:24:00 -0000 1.27 @@ -4,7 +4,7 @@ * * conquestd - the Conquest server/client driver * - * $Id: conquestd.c,v 1.26 2005/07/04 03:18:22 jon Exp $ + * $Id: conquestd.c,v 1.27 2005/12/25 20:24:00 jon Exp $ * * Copyright 2003 Jon Trulson under the ARTISTIC LICENSE. (See LICENSE). ***********************************************************************/ @@ -499,8 +499,6 @@ stopUpdate(); - drcheck(); /* check the driver */ - /* update client view of the universe */ updateClient(FALSE); @@ -1635,6 +1633,11 @@ laststat = now; } + clbBlockAlarm(); /* no signals can be accepted when + drcheck is run */ + drcheck(); + clbUnblockAlarm(); + if (didsomething) continue; /* see if there is another pkt */ Index: HISTORY =================================================================== RCS file: /home/jon/src/repository/conquest/HISTORY,v retrieving revision 1.41 retrieving revision 1.42 diff -u -r1.41 -r1.42 --- HISTORY 10 Aug 2005 15:51:38 -0000 1.41 +++ HISTORY 25 Dec 2005 20:23:59 -0000 1.42 @@ -9,6 +9,17 @@ Conquest HISTORY +??? + + - add patch by Bill Patterson adding a '-f' fullscreen option + to conquestgl. + + - fixed localtime() libc futex wait hang in conquestd by + moving drcheck() out of the signal handler. Bad Juju. + Thanks to Bill Patterson for letting me have access to his + server to locate the problem and test a fix. + + Version 8.1.1 (stable) - during playback of a recording, honor the user config From jon at radscan.com Wed Dec 28 15:35:24 2005 From: jon at radscan.com (Jon Trulson) Date: Wed, 28 Dec 2005 15:35:24 -0700 (MST) Subject: [conquest] Conquest 8.1.2 (stable) is available Message-ID: At the usual http://radscan.com/conquest.html This version corrects a couple of potentially serious problems - one where the server could deadlock in libc on linux systems, and one 'reproduced' on Dave's server yesterday that could cause a client to get stuck in Mur when entering a game. It also adds the fullscreen stuff from Bill Patterson. I have received a mondo patch from Cat with the some new icons and such. I will try to release a preliminary version that is mostly working in the next week or so. Here's the release notes snippet: Version 8.1.2 (stable) 12/27/2005 - fixed localtime() libc futex wait hang in conquestd by moving drcheck() out of the signal handler. Bad Juju. Thanks to Bill Patterson for letting me have access to his server to locate the problem and test a fix. - fixed potential problem with ship attributes that could cause ships (including your own) to appear to be stuck in Murisak when you enter a game. Now the clients will issue a CPCMD_RELOAD request to the server upon entering a ship so that data is always accurate. - add patch by Bill Patterson adding a '-f' fullscreen option to conquestgl. Add '-g WxH' option to set the intial window size as well. -- Jon Trulson mailto:jon at radscan.com ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962 PGP keys at http://radscan.com/~jon/PGPKeys.txt #include "I am Nomad." -Nomad