| 00:36:06 | vogon_jeltz | w00t... I fixed up a lot of the code to work with SDL |
| 00:56:00 | vogon_jeltz | yo, apathy cream |
| 00:56:13 | vogon_jeltz | is about halfway done with the SDL code |
| 02:11:58 | vogon_jeltz | is away: TV for a couple hours... be back at 9 |
| 03:24:45 | vogon_jeltz | is back (gone 01:12:46) |
| 06:25:02 | vogon_jeltz | heh... this game rules |
| 07:16:26 | captain_proton | hops in |
| 07:16:34 | vogon_jeltz | yo |
| 07:16:40 | vogon_jeltz | you played Progress Quest? |
| 07:16:52 | captain_proton | no |
| 07:17:03 | vogon_jeltz | http://www.nailhead.com/pq/ |
| 07:17:12 | vogon_jeltz | get WineX from TransGaming... |
| 07:17:22 | vogon_jeltz | Progress Quest is, without a doubt, the best RPG ever |
| 07:17:34 | captain_proton | meh |
| 07:17:39 | captain_proton | i'm studying right now |
| 07:17:45 | vogon_jeltz | fine |
| 07:17:53 | vogon_jeltz | just download it, open it, and leave it running in Wine |
| 07:18:10 | captain_proton | leave it running? |
| 07:18:18 | vogon_jeltz | yes |
| 07:18:21 | vogon_jeltz | you'll see |
| 07:18:31 | captain_proton | bleh - maybe tomorrow |
| 07:21:52 | captain_proton | i'm too burned out right now to try to get wine anything to work |
| 07:29:13 | captain_proton | how goes SDL? |
| 07:29:25 | fontenot | vogon_jeltz: I tried that link |
| 07:29:30 | vogon_jeltz | I finished half of the work, and now I'm just gonna take a break for a day or so |
| 07:29:32 | fontenot | vogon_jeltz: all I get is a scam ad |
| 07:29:40 | vogon_jeltz | oop |
| 07:29:44 | vogon_jeltz | nailhead.org... sorry |
| 07:29:46 | captain_proton | lazy biotch ;) |
| 07:29:53 | fontenot | You're in Serious Trouble - It's a Proven Fact! |
| 07:29:53 | fontenot | Deleting "Internet Cache and History" will NOT protect you because any of the Web Pages, Pictures, Movies, Videos, Sounds, E-mail, Chat Logs and Everything Else you see or do could easily be recovered to Haunt you forever! How would you feel if a snoop made this information public to your Spouse, Mother & Father, Neighbors, Church, Children, Boss or the Media? It could easily Ruin Your Life! Solve all your problems and enjoy all |
| 07:29:53 | fontenot | an "As New PC", Evidence Eliminator can Speed-Up your PC/Internet Browser, reclaim Hard Disk space and Professionally Clean your PC in one easy mouse click! |
| 07:29:54 | fontenot | CLICK HERE TO DOWNLOAD NOW! (special 10% discount appears at end of order) |
| 07:29:55 | vogon_jeltz | you know it ;) |
| 07:30:05 | vogon_jeltz | fontenot: yeah, like I said, nailhead.org, not nailhead.com |
| 07:30:16 | fontenot | oh, ok |
| 07:30:55 | fontenot | vogon_jeltz: is this rpg multi-player? |
| 07:31:07 | vogon_jeltz | fontenot: you'll see what it is soon enough ;) |
| 07:31:14 | fontenot | ah i see that it is |
| 07:32:03 | vogon_jeltz | captain_proton: I'm currently a level-3 Enchanted Motorcycle Robot Monk with a Bowie Knife, Macrame Helm, Warped Flannel Hauberk, Macrame Brassairts, Burlap Vambraces, -1 Holey Threadbare Chamois Gauntlets, +1 Macrame Gambeson, Macrame Cuisses, and -1 Macrame Greaves |
| 07:32:37 | captain_proton | threadbare chamois gauntlets? |
| 07:32:55 | vogon_jeltz | yes |
| 07:33:23 | captain_proton | interesting |
| 07:34:03 | vogon_jeltz | ooh, I just upgraded to a Burlap Gambeson |
| 07:34:19 | captain_proton | ROFL |
| 07:34:23 | captain_proton | http://fotm.rotten.com/fotm/ |
| 07:37:37 | vogon_jeltz | l33t... +1 Studded Lace Helm |
| 07:37:50 | captain_proton | ha |
| 07:38:18 | vogon_jeltz | dude, you have to see this thing |
| 07:38:44 | captain_proton | i'll look at it tomorrow when i have some energy left |
| 07:40:25 | captain_proton | is lace stronger than macrame? |
| 07:40:32 | vogon_jeltz | I'm not sure |
| 07:40:44 | vogon_jeltz | I'm up to -1 Canvas Gauntlets |
| 07:41:06 | vogon_jeltz | don't have a shield or any sollerets yet, tho |
| 07:42:20 | vogon_jeltz | I'm about 2/3 of the way to level 4, too :) |
| 07:44:15 | vogon_jeltz | -2 Flannel Brassairts |
| 07:49:10 | fontenot | i would think macrame would be stronger |
| 07:50:06 | vogon_jeltz | finally... Patched Canvas Sollerets |
| 07:57:36 | iLL_Victorious | =) |
| 07:57:37 | iLL_Victorious | heh |
| 07:58:07 | captain_proton | victorious? |
| 08:00:23 | iLLf8d | yeah |
| 08:00:40 | iLLf8d | leader in 2 servers |
| 08:00:46 | iLLf8d | and 2nd places in the other 2 |
| 08:01:20 | iLLf8d | actually I played 5 games |
| 08:01:30 | iLLf8d | one with gerb too was in 3rd there |
| 08:01:41 | iLLf8d | so 2 golds 2 silvers and a bronze heh |
| 08:01:50 | iLLf8d | oh dude uhm wanted to mention a good feature |
| 08:01:53 | iLLf8d | for 1.8 |
| 08:02:14 | iLLf8d | make phantomzone so players who are phantomed can kill each other =) |
| 08:02:39 | iLLf8d | maybe like steamroll or normal shooting or whatnot |
| 08:02:45 | captain_proton | hmm |
| 08:02:50 | iLLf8d | actually wierd shooting would be cool |
| 08:02:54 | iLLf8d | like all shots curve |
| 08:02:55 | iLLf8d | heh |
| 08:03:05 | iLLf8d | it is a phantomzone |
| 08:03:38 | fontenot | hey you experts - what's the machine-gun good for? |
| 08:03:45 | iLLf8d | killin |
| 08:03:46 | iLLf8d | heh |
| 08:03:56 | fontenot | the range is so short |
| 08:04:00 | captain_proton | for close-in fighting |
| 08:04:02 | iLLf8d | yeah its got its uses tho |
| 08:04:11 | iLLf8d | good to takeout gms |
| 08:04:14 | captain_proton | once someone with mg gets in close you're pretty much done for |
| 08:04:18 | iLLf8d | spiral in then waxem |
| 08:04:23 | fontenot | interesting |
| 08:04:33 | iLLf8d | assuming its a 2-3 shot field |
| 08:04:45 | fontenot | iLLf8d: thanks for your advice on dealing with GM - it helped |
| 08:04:52 | iLLf8d | np |
| 08:04:58 | fontenot | now if only I know how to deal with lasers... |
| 08:05:05 | captain_proton | CL |
| 08:05:06 | fontenot | ppl bouncing the laser all over the map |
| 08:05:06 | vogon_jeltz | stay off the ground |
| 08:05:06 | iLLf8d | whats your callsign? |
| 08:05:11 | fontenot | jacques |
| 08:05:15 | iLLf8d | oh ok |
| 08:05:16 | iLLf8d | =) |
| 08:05:17 | iLLf8d | heh |
| 08:05:23 | fontenot | CL cant be hit my laser? cool |
| 08:05:34 | iLLf8d | laser get off ground is key |
| 08:05:43 | fontenot | makes sense |
| 08:06:01 | iLLf8d | then get close and entice them to shoot atya |
| 08:06:02 | iLLf8d | nheh |
| 08:06:12 | iLLf8d | then pickem off outta the air when they jump |
| 08:06:19 | iLLf8d | use a parallel parking maneuver |
| 08:06:52 | iLLf8d | ie back tank up diagnoally then pull forward and shootem while they shoot one car width to yoru left of right |
| 08:06:59 | iLLf8d | or in this case tankwidth |
| 08:07:11 | iLLf8d | and more importantly |
| 08:07:12 | iLLf8d | when they jump you drop |
| 08:07:21 | iLLf8d | cause you drop faster then they float up and back down |
| 08:07:37 | iLLf8d | so you can beatem to ground and fire then try to swerve |
| 08:07:43 | iLLf8d | assuming you're close enough |
| 08:08:03 | iLLf8d | still talking lasers =) |
| 08:08:26 | fontenot | wow I'll have to develop some more skills before trying that one :-) |
| 08:08:56 | captain_proton | or you could take the easy route and when they jump, shoot towards them and jump yourself |
| 08:09:00 | iLLf8d | yeah theres alot of ways to deal with them |
| 08:09:21 | iLLf8d | that assumes they jump in a straight line and are close |
| 08:09:31 | iLLf8d | newbies do that shit all the time |
| 08:09:35 | iLLf8d | straight up and downs |
| 08:09:36 | iLLf8d | heh |
| 08:10:16 | iLLf8d | you just gotta gauge what a player can do |
| 08:10:34 | iLLf8d | see cap here has a deadly jump spin where he can nail you on the first round over the rooftop (or decktop) |
| 08:10:46 | iLLf8d | so you gotta move around on him so he don't wax your ass |
| 08:10:47 | captain_proton | MWAHAHAHA |
| 08:11:12 | iLLf8d | you'll learn what to watch out for |
| 08:11:25 | iLLf8d | I think i've seen alot of people using other peoples callsigns lately |
| 08:11:35 | iLLf8d | cause one day i'll play someone and they'll be really good |
| 08:11:40 | iLLf8d | the next day they'll suck |
| 08:11:47 | iLLf8d | dunno whats up with that |
| 08:13:20 | fontenot | hmm |
| 08:13:47 | fontenot | som eppl amaze me with their timing skills |
| 08:14:09 | captain_proton | fires up a 1.8 server |
| 08:14:16 | iLLf8d | cool |
| 08:18:49 | captain_proton | care to try it out? |
| 08:19:03 | iLLf8d | yeah I'm there already |
| 08:19:04 | iLLf8d | nheh |
| 08:19:29 | captain_proton | no you aren't |
| 08:33:00 | iLLf8d | yeah I was I had come and gone heh |
| 08:44:20 | iLLf8d | uhm |
| 08:44:30 | iLLf8d | cap whats a unit p and width/height? |
| 08:45:09 | iLLf8d | ie how many p's across the screen and why bother with a width/height spec (and point spec for the type)? |
| 08:45:23 | captain_proton | a p is a pixel |
| 08:45:31 | iLLf8d | what about -4 for example? |
| 08:45:44 | captain_proton | -4 would be -4 pixels offset from the opposite side |
| 08:46:06 | iLLf8d | and the percentages? |
| 08:46:11 | captain_proton | y=-4p is 4 pixels from the top of the screen, x=-4p is 4 from the right |
| 08:46:31 | captain_proton | they're exactly what they seem - 15% of the width of the screen for instance |
| 08:46:45 | iLLf8d | k uhm |
| 08:46:49 | iLLf8d | what about 4p then? |
| 08:46:56 | captain_proton | 4 pixels |
| 08:47:08 | iLLf8d | you said 04 is from top so 4 is from bottom then? |
| 08:47:11 | iLLf8d | sorry |
| 08:47:16 | captain_proton | yes |
| 08:47:21 | iLLf8d | heh |
| 08:47:26 | iLLf8d | I meant -4 heh |
| 08:47:28 | iLLf8d | was a typo |
| 08:47:31 | iLLf8d | ok =) |
| 08:47:37 | captain_proton | in gl numbering goes from the bottom left to top right |
| 08:47:45 | iLLf8d | oh hmm |
| 08:47:54 | iLLf8d | wierd |
| 08:47:55 | iLLf8d | heh |
| 08:48:06 | captain_proton | back-asswards if you ask me but thats how they did it |
| 08:48:19 | iLLf8d | must've been for some reason |
| 08:48:49 | iLLf8d | hmm |
| 08:48:51 | captain_proton | well - it stems from the fact that its 3d |
| 08:48:54 | iLLf8d | pertty cool |
| 08:49:11 | captain_proton | the view in gl assigns (-1,-1) to the bottom left and (1,1) to the top right |
| 08:49:11 | iLLf8d | the fov and theta? |
| 08:49:27 | captain_proton | field-of-view and view angle |
| 08:49:39 | iLLf8d | ooooo |
| 08:49:44 | iLLf8d | so I can change taht too |
| 08:49:46 | iLLf8d | wild |
| 08:49:47 | iLLf8d | hmm |
| 08:49:48 | captain_proton | thats how tim created his rear-view |
| 08:49:58 | iLLf8d | dunno how to turn on rearview |
| 08:50:05 | captain_proton | `set displayRear 1 |
| 08:50:12 | iLLf8d | also is there a way to interactively edit the files? |
| 08:50:20 | iLLf8d | I haerd you meantion command line mode or something |
| 08:50:26 | captain_proton | huh? |
| 08:50:55 | iLLf8d | I mean is there a way to just send the layout commands on the command line |
| 08:51:02 | iLLf8d | ie when in the game |
| 08:51:05 | captain_proton | no |
| 08:51:06 | iLLf8d | theres a console right? |
| 08:51:10 | captain_proton | yep |
| 08:51:13 | iLLf8d | hmm |
| 08:51:23 | captain_proton | thats where you need to do the 'set ...' |
| 08:51:40 | iLLf8d | one other thing does the game need to be restart to reread the bzc files? |
| 08:51:49 | captain_proton | no |
| 08:51:55 | captain_proton | there's a command to do it |
| 08:52:02 | captain_proton | not sure which though - hang a sec |
| 08:52:43 | iLLf8d | does he use the same parser for every file? |
| 08:52:59 | iLLf8d | ie does he have some kinda xml parsing engine (I assume this is xml) |
| 08:53:07 | captain_proton | he's got an engine but its bad |
| 08:53:15 | iLLf8d | lol |
| 08:53:16 | captain_proton | i've got a nice one that i'm probably going to swap in |
| 08:53:22 | iLLf8d | ah |
| 08:53:29 | iLLf8d | I thought that stuff was standardize |
| 08:53:30 | iLLf8d | d |
| 08:53:31 | iLLf8d | heh |
| 08:53:50 | captain_proton | it is - its just crs wrote it badly =} |
| 08:53:50 | iLLf8d | ie there were packages people have been working on for use with xml (ergo the draw)? |
| 08:54:15 | captain_proton | he didn't use one of the xml libs |
| 08:54:27 | iLLf8d | mean any or left one out? |
| 08:54:38 | iLLf8d | oh I see what you mean |
| 08:54:41 | iLLf8d | he made a custom parser |
| 08:54:44 | captain_proton | yeah |
| 08:54:48 | iLLf8d | this is xml tho right? |
| 08:54:57 | iLLf8d | looks xmlish |
| 08:54:58 | captain_proton | yep |
| 08:55:00 | captain_proton | it is |
| 08:55:12 | captain_proton | some of it doesn't meet the standard (yet) but i just have to finish fixing all the tags |
| 08:55:13 | iLLf8d | map files gonna work the same way? |
| 09:05:53 | iLLf8d | hmm |
| 09:05:57 | iLLf8d | that kinda fucked |
| 09:06:29 | iLLf8d | too bad theres no mechanism to make it linked relative screen size |
| 09:07:08 | iLLf8d | those references are to full view not active view so when i selected 100% it ran far offscreen heh |
| 09:07:15 | iLLf8d | cools system tho very cool |
| 11:08:26 | vogon_jeltz | is away: 5133p |
| 12:09:58 | fontenot | hmg |
| 12:10:14 | fontenot | would be nice if 1.7 and 1.8 didnt share a config file |
| 12:10:22 | fontenot | each messes up the config of the other |
| 12:10:49 | fontenot | also, the OO display is different in 1.8 - you cant see the building you are inside at all |
| 12:10:55 | fontenot | it's disorienting after 1.7 :-) |
| 14:29:24 | Chestal | ibot, help |
| 16:53:17 | captain_proton | returns |
| 16:54:33 | Chestal | hi capt |
| 16:55:17 | captain_proton | hallo |
| 16:57:10 | captain_proton | waits patiently for crs to stop by |
| 18:43:34 | crs23 | hey all |
| 18:51:22 | Chestal | hi crs |
| 18:51:39 | crs23 | how's v1.8 working for you at work? |
| 18:52:12 | Chestal | haven't tried the newest version yet |
| 18:52:19 | Chestal | just a moment |
| 18:52:57 | crs23 | k. i wanted to see if performance was okay when texturing/blending/lighting were off. |
| 18:53:10 | Chestal | last time I checked I got about 10fps |
| 18:53:23 | Chestal | I have 10-50 fps with 1.7, dependign on what's in my fov |
| 18:54:06 | Chestal | 1.7 gets slow when it has to a big wall is in view, i.e. I am standgn very close to a buildign and facing it |
| 18:54:15 | Chestal | but it's quite fast on the open field |
| 18:54:19 | crs23 | did you have depth buffer off in v1.7? |
| 18:54:37 | Chestal | hmm, let me check |
| 18:55:37 | crs23 | v1.8 doesn't have the option and probably never will |
| 18:55:43 | Chestal | I have it on, but it doesn't seem to make a difference |
| 18:56:00 | Chestal | I've everythign off except lighting |
| 18:56:29 | Chestal | btw., I think it's a good idea to change the name of the config file for 1.8 |
| 18:56:38 | Chestal | people who want to try 1.8 and play with 1.7 will have problems now |
| 18:56:49 | crs23 | good point. |
| 18:57:11 | crs23 | interesting depth buffer doesn't make a difference. this is software rendering, right? |
| 18:57:30 | Chestal | I get 12fps at most with 1.8 now |
| 18:57:33 | Chestal | yew, Mesa |
| 18:57:48 | Chestal | 4.1.0-6 |
| 18:58:22 | Chestal | I'm running a 2nd X server with 8bit colour depth for bz |
| 18:58:29 | Chestal | bz runnign fulscreen, 800x600 |
| 18:58:37 | crs23 | the slowness should be due to drawing the pixels and i would've thought depth buffering would slow it down |
| 18:59:04 | Chestal | hmm, togglign depth buffer should work on the fly, shouldn't it? |
| 18:59:15 | crs23 | it does in v1.7, yes. |
| 19:00:06 | crs23 | maybe the hack used to draw the background in 1.7 accounts for some performance |
| 19:00:18 | crs23 | 1.7 just used a screen clear for the sky and another for the ground. very fast. |
| 19:00:53 | Chestal | is this easily odne in 1.8? |
| 19:01:06 | Chestal | for performance testign I could even remove sky and ground compeltely |
| 19:01:22 | captain_proton | wakes up again |
| 19:01:25 | captain_proton | hi crs |
| 19:01:26 | crs23 | yeah, you can try that. |
| 19:01:36 | iLLf8d | framerates didn't display in the last version I cvsd that was a number of days ago tho maybe more then a week |
| 19:01:54 | crs23 | iLLf8d: toggle displayFPS |
| 19:01:55 | captain_proton | crs23: i get a higher framerate in 1.8 than 1.7 |
| 19:02:00 | captain_proton | with all options on |
| 19:02:05 | iLLf8d | cool |
| 19:02:06 | Chestal | ill: did you use toggle displayFPS? |
| 19:02:16 | Chestal | crs23: if I knew how, I would :-) |
| 19:02:16 | iLLf8d | nope I was hitting t or whatever the key is |
| 19:02:21 | crs23 | captain_proton: great! but you've got 3D hardware. |
| 19:02:33 | captain_proton | true |
| 19:02:39 | crs23 | Chestal: okay, edit models.bzg and ground.bzg |
| 19:02:39 | Chestal | crs23: btw, with textures off, the console is still behaving strange. Try switching tetxures off, activate consoel and type 'set' |
| 19:02:42 | iLLf8d | checks his framerates right after he cvs compiles |
| 19:02:51 | captain_proton | hey crs23 i was looking at something |
| 19:03:04 | crs23 | captain_proton: just a sec |
| 19:03:07 | captain_proton | for the models in pixel view |
| 19:03:22 | captain_proton | it'd be awfully nice to have negative values just like in the views |
| 19:03:48 | crs23 | captain_proton: explain? |
| 19:04:03 | Chestal | hmm, I can almost understand ground.bzg |
| 19:04:22 | crs23 | Chestal: easiest thing, edit models.bzg and find <include filename="ground.bzg"> |
| 19:04:23 | Chestal | btw., what type of comments can be used in those xml files? |
| 19:04:27 | captain_proton | well, i wanted to create a border around a viewport |
| 19:04:36 | crs23 | put a # at the start of that line to comment it out |
| 19:04:42 | Chestal | crs: yep, found the 3 includes |
| 19:04:53 | crs23 | currently # is a comment but we'll be fixing that to be <!-- --> |
| 19:05:08 | crs23 | comment all three. now at the top we need to disable the sky... |
| 19:05:34 | crs23 | put a <choice><mask>0</mask> before the <spherical> and </choice> after the matching </spherical> |
| 19:05:54 | captain_proton | and just specifying 0 and 1 as vertices in an lloop doesn't work (the two lines at 1 get cut off by the scissor) |
| 19:05:56 | crs23 | now your display is going to smear *a lot* because the background isn't cleared |
| 19:06:10 | captain_proton | so i was thinking pixel values |
| 19:06:20 | Chestal | 15fps now |
| 19:06:25 | captain_proton | but it'd be nice to have it offset negative values from the opposite side |
| 19:06:26 | crs23 | captain_proton: ahh. yeah fixed sized border are a bit difficult. |
| 19:06:34 | Chestal | I will start a local server, Tim is using a custom map, this might slow things down |
| 19:08:06 | crs23 | captain_proton: the view code can actually handle a percentage size plus a pixel size but you can't specify that in the format |
| 19:08:58 | captain_proton | hmm |
| 19:09:01 | iLLf8d | crs you changed the way it compiles too didn'tya, I used to have to edit one of the config files to get it to include one of my libs but I don't have to anymore |
| 19:09:03 | Chestal | crs: about 18fps, still a lot slower than 1.7. It's hard to read fps now, as the ground isn't refreshed. The sky is solid blue now |
| 19:09:23 | crs23 | that might be useful: layer one view with an additional pixel offset under one without |
| 19:10:20 | crs23 | the sky is blue? |
| 19:10:27 | Chestal | yes |
| 19:10:31 | captain_proton | looks outside |
| 19:10:47 | crs23 | it must be clearing the sky somewhere |
| 19:10:55 | Chestal | I commented out background, cliurs and ground includes |
| 19:11:23 | crs23 | does the sky change color if you change the time of day (- and = keys)? |
| 19:11:25 | Chestal | cliurs is 'clouds' :-) |
| 19:11:28 | Chestal | let me try |
| 19:11:57 | crs23 | iLLf8d: i think i only added auto-dependency generation on linux. nothing about libraries. maybe tim did it. |
| 19:12:34 | Chestal | yes, and I get stars at night, and the sky isn't necessarily solid, at a few times of day I can see colur stripes |
| 19:12:46 | iLLf8d | yeah I used to have to edit linux-config or config-linux and add into LDOPTS |
| 19:12:57 | crs23 | hmm, the sky wasn't disabled. did you put the choice node into models.bzg? |
| 19:13:17 | Chestal | choice node? |
| 19:13:28 | crs23 | scroll back and look for my message |
| 19:14:38 | Chestal | ok, didn't get this the first time |
| 19:15:11 | iLLf8d | hey crs did you convert from an old map file or revamp the map file with a new map? (ie are map files in xml now)? |
| 19:15:19 | Chestal | seems to be faster now, can I reposition the FPS display? |
| 19:15:25 | Chestal | I can't read it all that well now... |
| 19:15:39 | captain_proton | iLLf8d: map files haven't changed yet |
| 19:15:45 | iLLf8d | ah |
| 19:15:53 | crs23 | hmm, good point. uh, yeah, edit view.bzc. |
| 19:15:55 | iLLf8d | my framerates I think are better or the same in 1.8 |
| 19:16:51 | Chestal | ill: with hardware accleration I suppose? |
| 19:17:08 | crs23 | find <if name="displayFPS">. before the <text...> line insert <model filename="radar.bzg" />. |
| 19:17:26 | BZFlag | hey all. |
| 19:17:31 | captain_proton | hi timmy |
| 19:17:31 | crs23 | hey BZFlag |
| 19:17:45 | iLLf8d | missles smoke looks cooler now too |
| 19:17:47 | iLLf8d | heh |
| 19:17:48 | crs23 | i'm working on protocol.txt comments now |
| 19:19:01 | crs23 | fyi, everyone, i added a file doc/crs-plans with some ideas. please have a look and comment. |
| 19:19:01 | Chestal | crs: hmm, now the whoel 3D view is black |
| 19:19:19 | crs23 | the <model...> should go inside the <if> block |
| 19:19:30 | crs23 | oops, hang on, my bad |
| 19:19:58 | Chestal | can't I jsut chaneg the coordinates? |
| 19:20:22 | crs23 | yeah. remove the model and move it somewhere that is cleared. |
| 19:20:30 | crs23 | the x and y params to text are the coordinates. |
| 19:20:38 | Chestal | will try center of screen |
| 19:20:48 | crs23 | a `p' suffix means pixels and negative means from the right/top |
| 19:21:04 | crs23 | a `%' suffix means a percentage of the viewport. use 50%, 50% |
| 19:21:15 | captain_proton | starts reading crs-plans |
| 19:21:34 | Chestal | aah, much better, 30-50fps now mostly |
| 19:21:42 | Chestal | similar to 1.7 |
| 19:21:45 | crs23 | wow! |
| 19:21:57 | crs23 | okay, the big smooth shaded sky is the problem then |
| 19:22:04 | Chestal | it gets slwoer, when facing a big wall, I guess it's really the pixels that are drawn |
| 19:22:08 | crs23 | yes |
| 19:22:32 | crs23 | try this. remove the choice node you added to models.bzg then... |
| 19:22:56 | Chestal | ok |
| 19:22:58 | BZFlag | have you all talked about the mountian clipping? |
| 19:23:17 | BZFlag | ie: look diagonal across the world, and the mountians get clipped. |
| 19:23:34 | Chestal | no mountains here :-) |
| 19:23:35 | crs23 | in the <geometry> node in the sky, you should remove all the <color> sections but "early morning/late evening" |
| 19:23:37 | captain_proton | crs23: i *love* the go-style territory idea ;) |
| 19:23:58 | Chestal | for my part, I do not quite get the reversi and/or go style ideas |
| 19:24:21 | crs23 | and remove all but one line from that section. keep one of the `0.43 0.75 0.95 1' lines. |
| 19:24:33 | crs23 | dont' forget to remove the `<color ref=3/>' line |
| 19:24:56 | crs23 | go and reversi are games played on a grid of squares... |
| 19:25:15 | crs23 | in go, if you surround and area you capture every surrounded cell. |
| 19:25:43 | captain_proton | well...go is a little different (array of lines) |
| 19:25:49 | crs23 | in reversi, when you capture a cell you capture all cells between that and you closest cell on the horizontal, vertical, and main diagonals. |
| 19:26:18 | captain_proton | crs23: you play go? |
| 19:26:34 | Chestal | should I keep <vertex> stuff in <geometry>? |
| 19:26:39 | crs23 | no, so i might be mistaken on the details |
| 19:26:51 | captain_proton | ahh - i can help out with that then =) |
| 19:26:56 | crs23 | heh |
| 19:27:22 | crs23 | i can think of two styles: one where you try to make a line to the opposite side of the board... |
| 19:27:30 | crs23 | and the other where you try to capture the most area. |
| 19:27:46 | crs23 | Chestal: yes, for now. |
| 19:27:59 | Chestal | solid blue sky, fps back to 18 |
| 19:28:23 | crs23 | okay. modify the gstate... |
| 19:28:55 | crs23 | change <shading model="smooth" /> to <shading model="flat" /> ... |
| 19:29:17 | Chestal | that's actualyl one thing I thoguht I tried before, but let's see |
| 19:29:59 | Chestal | this helps a lot |
| 19:30:03 | Chestal | about 30fps |
| 19:30:13 | crs23 | maybe <dithering dither="off" /> will help a bit too |
| 19:30:16 | Chestal | it was significantly faster still without a sky |
| 19:30:35 | crs23 | yeah, well filling pixels takes time :-) |
| 19:30:44 | BZFlag | crs23: have you tried: LIBGL_ALWAYS_INDIRECT=1 bin/bzflag ? |
| 19:31:17 | BZFlag | should turn off your hw accel for that instance. |
| 19:31:31 | crs23 | ah, interesting. just a sec. |
| 19:32:19 | Chestal | dithering doesn't seem to make a big difference |
| 19:32:38 | Chestal | btw., the full screen view in 1.8 has a different aspect ratio than the cinemascope-like view in 1.7 |
| 19:32:41 | crs23 | Chestal: i'm not surprised |
| 19:33:05 | Chestal | can I think of the 1.7 view beign 1.8 view with top and bottom clipped? |
| 19:33:07 | captain_proton | likes the new fov better - he's been using it for a long time =) |
| 19:33:11 | crs23 | BZFlag: didn't work. is that Mesa only? i'm using nvidia libs. |
| 19:33:27 | captain_proton | Chestal: yes you can |
| 19:33:28 | crs23 | Chestal: yes |
| 19:33:50 | crs23 | Chestal: btw, that's one reason it's slower -- you've got more pixels to fill per frame |
| 19:34:27 | captain_proton | is anxious to put up iLLf8d's boxy world with chessboard textures |
| 19:34:55 | iLLf8d | oh the evil one? |
| 19:34:56 | iLLf8d | heh |
| 19:35:04 | iLLf8d | I got another like that too I made recently |
| 19:35:07 | iLLf8d | mor maneuverable |
| 19:35:08 | BZFlag | crs23: hmm, guess it's mesa only. I didn't think so though. perhaps nvidia choose not to support it. |
| 19:35:11 | iLLf8d | but good too |
| 19:35:24 | iLLf8d | has alot of qbert style shit up top |
| 19:35:26 | captain_proton | iLLf8d: you might have sent it to me - i like the old one better |
| 19:35:33 | iLLf8d | this is different |
| 19:35:41 | iLLf8d | its more decks ish |
| 19:35:50 | captain_proton | considers putting models of pawns around on top =) |
| 19:35:53 | iLLf8d | well tho not decks at all don't wanna mislead you witht hat comment |
| 19:35:55 | crs23 | i had a libGL.a floating around a few weeks ago that i was unintentionally linking against... |
| 19:36:29 | iLLf8d | I'm gonna reload a box today so I can put up a flag server in my dmz |
| 19:36:30 | crs23 | when i moved it aside my top independent triangle rate on the stanford bunny went from 700k/sec to 7M/sec |
| 19:36:40 | crs23 | maybe that was the Mesa lib. good riddance! |
| 19:36:54 | captain_proton | crs23: which card? |
| 19:37:03 | crs23 | GeForce2/GTS 4x AGP |
| 19:37:18 | captain_proton | ahh |
| 19:37:58 | crs23 | that's around 22M vertices per second and nvidia claims only 25M, so it was pretty close to max |
| 19:38:34 | crs23 | could probably get faster by playing some nasty games with vertex buffer extensions |
| 19:38:53 | crs23 | but a quick hack made it slower so i stopped trying |
| 19:39:02 | BZFlag | heh |
| 19:39:05 | iLLf8d | uhm I have those in these paths /usr/share/nvidia-glx/diversions-X11R6/libGL.a /usr/share/nvidia-glx/diversions/libGL.a uhm there some way for me to eek some more speed out? |
| 19:39:47 | crs23 | i removed the libGL.a to go faster. i now link against /usr/lib/libGL.so |
| 19:40:06 | iLLf8d | hmm |
| 19:40:42 | crs23 | that's from the NVIDIA_GLX-1.0-2313 RPM |
| 19:41:32 | crs23 | it's important to note that the stanford bunny has 70k triangles so there's little overhead to pushing triangles... |
| 19:41:55 | crs23 | bzflag has much more overhead with switching models, drawing just a few triangles per model, doing transforms, etc. |
| 19:43:09 | Chestal | crs23: the 1.7 sky/ground hack will work as long as view is parallel to ground, I assume? |
| 19:43:12 | captain_proton | we should try to minimize that |
| 19:43:35 | crs23 | yes, only when the horizon is horizontal. |
| 19:43:50 | Chestal | currently this is the case at all times |
| 19:44:00 | captain_proton | oh there was something weird i noticed |
| 19:44:03 | crs23 | captain_proton: yeah, i'm already merging all the buildings into a single pre-transformed model |
| 19:44:23 | captain_proton | if you get up high and look beyond the moutains, at the low points there are drawing errors (not drawing anything?) |
| 19:44:50 | crs23 | yeah, i know. same thing if you turn the mountains off. the sky doesn't meet the ground. |
| 19:44:59 | captain_proton | err - why not? |
| 19:45:21 | crs23 | the sky is part of the local world but the sky is infinitely distant |
| 19:45:33 | crs23 | erm, /sky/ground/ |
| 19:45:51 | crs23 | so the ground moves down when you jump but the sky does not |
| 19:46:02 | captain_proton | ok... |
| 19:46:24 | captain_proton | hmm |
| 19:46:37 | crs23 | i think we can make the sky move but the sun/moon will then appear to move within the sky when you jump |
| 19:46:56 | captain_proton | ick |
| 19:47:03 | crs23 | that's only noticeable when you've got the sunset color gradient to compare against |
| 19:47:52 | crs23 | another solution is to put a skirt around the bottom of the sky |
| 19:48:14 | captain_proton | thats what i think ought to be done =) |
| 19:48:32 | captain_proton | or - could we just extend the sky down further? |
| 19:48:57 | iLLf8d | yeah but gradient issues again right |
| 19:49:19 | crs23 | right, gradient issues. the skirt, uh, skirts the problem |
| 19:50:35 | captain_proton | ugh...i just took a sip of some rancid coffee |
| 20:00:04 | Chestal | tries to put the 1.7 fast-sky/ground=drawing into 1.8, but is getting strange result |
| 20:00:18 | Chestal | hmmph, why are there no man pages for gl functions on my system? |
| 20:00:29 | crs23 | mine, neither. try opengl.org |
| 20:00:30 | captain_proton | the only systems with gl man pages are SGI |
| 20:01:26 | crs23 | win32 too |
| 20:01:38 | crs23 | well, with visual c++ |
| 20:02:08 | crs23 | one of the rare cases where the visual c++ docs don't suck :-) |
| 20:02:21 | captain_proton | starts looking into where to put the particle engine |
| 20:03:47 | captain_proton | it should be a SceneNode right? |
| 20:05:06 | crs23 | right. i'd like to see a proposal written up once you've got a handle on how to add it. |
| 20:06:00 | captain_proton | well - i'm thinking it can share timing code with a SceneNodeAnimate |
| 20:06:44 | crs23 | probably. particles can have more complex motions than animate can do, though. |
| 20:07:05 | captain_proton | true |
| 20:07:27 | captain_proton | but at their core they just update their position each 'tick' |
| 20:07:49 | crs23 | they also get created and die |
| 20:08:01 | captain_proton | individual particles do |
| 20:08:04 | crs23 | right |
| 20:08:08 | captain_proton | but the particle system as a whole controls all that |
| 20:08:20 | captain_proton | eg. SceneNodeParticleSystem |
| 20:08:33 | captain_proton | all you'd have to deal with is that |
| 20:08:49 | captain_proton | and it handles creation, updating, and death of particles based on its parameters |
| 20:09:21 | Chestal | at least I got my screen all blue and green now :-) |
| 20:09:21 | crs23 | so particles would not be vertices in a geometry node? |
| 20:09:44 | captain_proton | no |
| 20:09:58 | crs23 | Chestal: a screen clear for the ground will make the shockwave look funny (it won't get clipped to the ground). just fyi. |
| 20:10:11 | crs23 | captain_proton: makes sense. |
| 20:10:34 | Chestal | crs: at the moemnt, I can't see anything except sky and ground |
| 20:10:43 | Chestal | crs: is this sw bug also present in 1.7? |
| 20:10:44 | crs23 | oh my. |
| 20:11:10 | crs23 | i don't think so. iirc, v1.7 uses glClipPlane to do that clipping. |
| 20:11:38 | Chestal | is hackign around without any knowledge of GL whatsoever :-) |
| 20:11:59 | crs23 | don't worry about it for now. |
| 20:12:20 | captain_proton | thats what i was doing at first Chestal =) |
| 20:12:26 | crs23 | when you get the world back, let me know the FPS with clearing vs. drawing solid color polygons. |
| 20:14:05 | crs23 | captain_proton: okay, go ahead with the particle system node. i suggest using bzgview for testing. |
| 20:14:43 | BZFlag | crs23: you been on two:5190? |
| 20:14:43 | captain_proton | oh crs23 - care to take a look at this XMLTree class and see if it might be useful? |
| 20:15:01 | crs23 | BZFlag: no, not yet. i'll have a look. |
| 20:15:08 | crs23 | captain_proton: where? |
| 20:15:14 | captain_proton | crs23: let me make a tarball |
| 20:15:23 | BZFlag | elevated bases have 0 height. of course the current protocol does not send one anyway |
| 20:15:55 | crs23 | captain_proton: i'm using xchat now so DCC should work (i think) |
| 20:16:11 | BZFlag | 1.7e4 has a hack for 1unit height bases if they are elevated, so perhaps we should not worry about it till we get to the newer protocol. |
| 20:16:46 | BZFlag | also I'm thinking if bases are elevated, they should be solid on radar, not just the outline. |
| 20:17:32 | BZFlag | perhaps they should be solid all the time. =) |
| 20:18:01 | captain_proton | i like the outline but imo they shouldn't fade (or at least not as much) |
| 20:18:05 | BZFlag | crs23: have you tried many custom worlds? two:5190 is running my HiX ctf world. |
| 20:18:25 | crs23 | i've played on a few. |
| 20:18:26 | BZFlag | captain_proton: an elevated world that's faded is very tough to see, |
| 20:19:18 | BZFlag | I think you need a way to tell that the base is not on your elevation. so fading does make sense. |
| 20:19:38 | Chestal | oh well, I can't get it to work. I jsut tried taking the glScissor / glClearColor calls from 1.7, but I obviously am doing soemthing wrong. Wherever I put them, they are draw above everythign else |
| 20:21:41 | BZFlag | rear view looks nice mirrored btw. ;-) |
| 20:21:53 | crs23 | Chestal: try src/bzflag/playing.cxx. find the first glClear(). put it before the view->render() in the next if block. |
| 20:22:30 | crs23 | BZFlag: mirroring wasn't quite as trivial as i hoped but wasn't too bad. |
| 20:24:10 | captain_proton | is a mirrored rearview in cvs? |
| 20:24:59 | Chestal | crs23: that's exactly the palce I tried last |
| 20:25:16 | Chestal | but my world is all blue now |
| 20:25:25 | crs23 | captain_proton: yes. `toggle displayRear' |
| 20:25:51 | crs23 | Chestal: hmm. |
| 20:26:02 | captain_proton | ahh pretty |
| 20:26:15 | vogon_jeltz | is back (gone 09:17:45) |
| 20:26:23 | BZFlag | the wider view makes more sense too. nice. |
| 20:26:31 | crs23 | BZFlag: btw, i made the rear view wider to match the aspect ratio. |
| 20:26:38 | crs23 | ah, you saw it |
| 20:27:36 | captain_proton | SceneNode::visit is what actually pushes all the triangles? |
| 20:28:48 | crs23 | SceneVisitor derived classes traverse the scene graph. SceneVisitorRender and SceneVisitorSimpleRender do rendering. |
| 20:29:23 | Chestal | uh, I forget to save a file |
| 20:30:09 | crs23 | SceneNode::visit() just forwards to a visitor. this lets the visitor know the exact type of object being visited. |
| 20:31:07 | crs23 | there are two problems. the first is my mistake; i should've used a separate iterator class for traversing the graph. |
| 20:32:14 | crs23 | the second is that adding a new node type means updating all the visitor types to handle it. this is a consequence of the design. |
| 20:32:42 | crs23 | however, it's not too bad since SceneVisitor can often supply a useful default behavior. |
| 20:33:26 | captain_proton | hrm |
| 20:34:30 | captain_proton | maybe i'll create a SceneNodeParticle thats hidden from everything else... |
| 20:35:06 | captain_proton | unless - can i push multiple triangle strips in a single node? |
| 20:36:45 | crs23 | no, no, it's easy to add a new node. really. derive SceneNodeParticle from SceneNode... |
| 20:36:52 | crs23 | add a visit(SceneNodeParticle*) to SceneVisitor... |
| 20:37:21 | crs23 | add and override that visit() method in SceneVisitorRender to make the stuff to render |
| 20:37:54 | captain_proton | hmmm i'll have to learn how this works more |
| 20:38:04 | Chestal | ok, got it working now. Maybe I'm only imagining, but the actual framerate I perceive seems to be much less than the one which displayFPS shows |
| 20:39:31 | crs23 | captain_proton: try walking through SceneVisitorSimpleRender and/or SceneVisitorRender::traverse() |
| 20:40:31 | crs23 | the visitor calls root->visit(this). the root node turns around and calls the visit() method on the visitor that takes the root node's type. |
| 20:41:01 | crs23 | the visitor then handles that node and, for nodes with children, descends into those children by calling visit() on them and the process repeats. |
| 20:42:32 | captain_proton | starts reading code |
| 20:42:34 | crs23 | XMLTree doesn't appear to add any functionality (except the minor fixes to be strictly XML). the main difference is that it stores the parsed file whereas ConfigRead processes on the fly. |
| 20:43:22 | captain_proton | yes |
| 20:43:50 | captain_proton | i've found it useful in that you can test the name of an XMLTree and if it matches just pass the tree to the object |
| 20:44:44 | crs23 | yes, that has it's advantages. |
| 20:45:13 | captain_proton | and that tree class is a nice STL-style tree =) |
| 20:45:15 | crs23 | the ConfigReader based code can look rather ugly. |
| 20:45:50 | captain_proton | i still need to go through and determine why it doesn't use std::map |
| 20:46:11 | crs23 | i was wondering that myself. a vector, eek! |
| 20:46:22 | captain_proton | well vectors are ok |
| 20:46:27 | captain_proton | but it uses its own Map class |
| 20:47:23 | crs23 | it uses a vector in the map and does linear searches. eek! |
| 20:47:47 | captain_proton | true... |
| 20:48:02 | captain_proton | but in any one node the number of name:value pairs will be small |
| 20:48:09 | crs23 | granted. |
| 20:48:44 | captain_proton | its probably faster to do an O(n) search than an O(nlogn) sort and an O(logn) search |
| 20:49:38 | crs23 | depends on the number of searches. in any case the difference is likely to be vanishingly small. |
| 20:49:46 | crs23 | for most inputs :-) |
| 20:49:49 | captain_proton | heh |
| 20:50:02 | Chestal | hmm, without lighting the buildings have a reddish look, and teleporter borders are yellow |
| 20:50:06 | Chestal | with lighting on, all is grey |
| 20:50:35 | crs23 | yes. i didn't make a special version of the world for lighted but untextured. |
| 20:59:39 | captain_proton | creates a basic particle system node |
| 21:15:52 | Chestal | crs: if I set displayDebug, my framerate looks a lot more even, it's difficult to describe, I guess it's more smooth |
| 21:16:21 | Chestal | withotu displayDebug, time for a single frame varies greatly, i.e. one slow frame and then 3 incredible fast frames |
| 21:17:21 | crs23 | displayDebug forces each frame to complete before the time is computed. |
| 21:17:50 | crs23 | without it, it's possible for a frame to still be rendering in the background. |
| 21:18:14 | crs23 | displayDebug can greatly reduce the frame rate, so that's why it's an option. |
| 21:18:28 | Chestal | without displayDebug it looks like only every nth frame will hit my screen |
| 21:18:33 | Chestal | it loosk much better with Debug for me |
| 21:21:02 | Chestal | without Debug, the fps displayed is higher, but it looks 'faked', actually the perceived framerate gets much lwoer |
| 21:21:47 | crs23 | that's weird. well, whatever works! |
| 21:22:05 | crs23 | gotta go mail a package. bye. |
| 21:22:21 | captain_proton | ttyl crs23 |
| 21:22:28 | Chestal | I'm heading home, maybe later |