Hard Light Productions Forums
Modding, Mission Design, and Coding => FS2 Open Coding - The Source Code Project (SCP) => Topic started by: WMCoolmon on December 16, 2005, 01:16:51 am
-
Since it seems that we have a relative shortage of any kind of significant multi support, in the FS community at large (read: HLP), I figured I'd start a thread and see what comes out of it.
So....*starts thread* :p
-
I want a 'capture the flag' mode. :lol:
Seriously. Is there any way to get back the 'lobby'?
-
We have the beginnings of the code, but no one available to fix it up.
We're chronically short-staffed. Probably the best Christmas present we could get would be a few extra coders. ;)
-
press quit? :p
capture the flag would look too weird (imagine a flag sticking out of a ship :p).
At the moment, FS2 doesn't have a lobby function while the game is in progress, and there is no main lobby, although i WOULD like to see both of those happen. :)
Also, what i want is a goddamn voice function for in-mission games, not only the game's lobby and briefings. :p
-
I want a 'capture the flag' mode. :lol:
Do-able in FRED already. The only reason I haven't made this already is the general lack of interest in FS2_Open multiplayer missions. Maybe I posted it on the wrong forum to get noticed or something but I didn't get a single reply on HLP when I posted (http://www.hard-light.net/forums/index.php/topic,37147.0.html) to say that I'd cracked how to get King of the Hill working (In fact the only response I did get anywhere was from DW-Hunter who had already helped me playtest the mission).
Hell after the lack of interest last time I dabbled in Multiplayer the only real reason I worked on KotH was simply for the challenge of doing something no one had done previously and in the hope that I'd get inspiration for single player gimmicks while doing it.
To be honest I think the reason why multi has remained so underdeveloped is that next to no one has any interest in it. The multi community persist in their belief that installing FSO will interfere with their ability to play games on FS2NetD and have their stats accepted (It won't if you use multimod like I've been banging on about for ages now). The GW Multi community (people like MP, Cobra and yourself etc) seem to have completely stopped playing even before FS2NetD went down. I just don't see the current market.
That's why I'm going to port my KotH mission to BSG soon and make a big push to have BSG multi right from the demo. BSG will bring in a legion of fans who actually want to play online. That's pretty much the only audience I currently see for FS2_Open multi.
As for what I want to see in Multi. To be honest I'd simply like to see the singleplayer stuff work in multi. For instance only the host will ever see a variable in a message correctly. The clients only get the default value of the variable. To be honest I'm not certain that the host ever sees any other value as except for anything regarding the hitpoints or subsystem status of a ship any time I've tried to pass a numeric or string variable to a SEXP the client and host have ended up with something completely different.
I'll be adding as many as I can to mantis this weekend. [/rant]
-
I'm always ready to help you playtesting multiplayer missions. Just catch me via ICQ or PM me.
I'm not playing online cause I want something more interesting than the standard stuff. And there is a second reason. My FS2 folder is mostly extremely messed up. I've had a lot of mutliplayer problems due to this and I refuse to install the game again, just to play some old, standard multiplayer missions.
I already have two installations of FS2_open (one for SoL), two of TBP, one of BSG and WC:S and three times Inferno.... OMG... I never thought about this... :eek2:
-
I'm always ready to helo you playtesting multiplayer missions. Just catch me via ICQ or PM me.
I suspected you'd be willing as you helped me out as the last time when I was testing to see if it was possible to have a campaign/mission where you buy ships you were the only one to put yourself forward to play with me. The thing is I'm not going to make a capture the flag mission if you and me are the only ones interested in playing it. Capture the flag is rather **** with only two players. You need a few more to make it fun and the AI really can't be trusted to handle it well enough to be fun.
I'm not playing online cause I want something more interesting than the standard stuff.
You should enjoy the KotH missions then. It's basically just TvT but the danger of leaving someone alone for too long does add a certain something to the mission. :) If you have other ideas for mission types catch me on ICQ and bounce them off of me (or just post them here).
And there is a second reason. My FS2 folder is mostly extremely messed up. I've had a lot of mutliplayer problems due to this and I refuse to install the game again, just to play some old, standard multiplayer missions.
I already have two installations of FS2_open (one for SoL), two of TBP, one of BSG and WC:S and three times Inferno.... OMG... I never thought about this... :eek2:
For the life of me I've never understood why people do this. Even if you're working on a stand-alone campaign and don't want to risk interferance from the game itself why not simply make a folder called Freespace2, move the standard VPs inside it and run FS2 itself as a mod as well? :D
The amount of hassle you'd save yourself from not having to update 9 different folders every time you wanted to update the build you're using would surely make it worth it.
-
:lol:
All you really need to 'install' FS2_Open are the retail VPs and Launcher 5.3 (for the registry entries). For retail FS2, you might need the retail launcher. But the only reason to actually install it from the disks is if you want to be able to remove the registry folder without using regedit. (Well, some programs also try to use the uninstall data to autodetect the FS2 folder)
I would be willing to test missions too...just give me a time/place/build and I'll try and show up, I'm relatively free for Winter break.
I've even kowtowed to using mod.ini so I can evade hacked table errors from mediaVPs. :nervous:
-
Well, i think the interest of FS2Open Multiplayer depend on available support for it. Last few weeks, noone can't simply find anyone else for game, becouse GWmasterserver is down. Calling friends via messagers to play via IP is troubleshooting and in addition, you can play only with ppl who you actually know. There is no possibility to play with someone new. Posting IP on Forums is not even effective. So that is one of the most couses for this interest for last days - there is no server! (Even a server without lobby, some ppl was met on lobby's alternative IRC channels to play and that was good. Not that good as pxo's lobby_20 but not that bad as no server at all.)
Other couse is that the Open's builds influence to game balance which is very important in Multiplayer (especially in TvT and Doggy mode). You can make your 100%-hull-enemy dead by one ram or 2-3 shoots with Kaysers :p. Some ppl wanted to avoid it so they played with OEB - Tom's build. Some gamers on Open's some on OEB... and we have very unbalanced game here.
And finally Multiplayer players need to feel and know that there is some people that really cares the Multiplayer mode and NetD server. People that can make your new multiplayer missions valid at anytime (NetD admins like MatthewPapa) and people making some new features available (SCP rulers :)) that will be based on gamers opionions above all. You are all made very good job so far, but last few weeks / months it was look like a death project.
Karajorma's KotH is a very good initiative. It seems to be a real fun :) I will test it tonight with some ppl via IP. But for more multiplayer interests that kind of initiative must be supported by all cases that I write above.
---
And for the new features in Multi mode, I would say all great things that SCP included in single so far :), maybe full-supported ingame joining (some ideas is at GW, f.e. here (http://www.game-warden.com/forum/showthread.php?t=1156)) and improves communication in game. Voice - more than 5sec msgs, and better quality ;), and text msgs - when you say something only to your team, make it with "Team message: xxx" signature. When to a target, make it with "Target message: xxx" signature when I receive a msg "xxx" from ppl that targets me. That is becouse I really don't know is my team m8 write this msg to all... or just to me.
-
Well, i think the interest of FS2Open Multiplayer depend on available support for it.
I agree. I'm glad WMCoolmon looks like he's interested but what we really need is for a coder to step forward who will make multi his priority, preferably someone from the existing multi community so that they know the pitfalls and problems with multi the way it is now.
Other couse is that the Open's builds influence to game balance which is very important in Multiplayer (especially in TvT and Doggy mode). You can make your 100%-hull-enemy dead by one ram or 2-3 shoots with Kaysers :p.
That's just a table issue really. Shouldn't be too hard to solve that one. There are probably some issues that are code and that's why we need a coder. Sadly all the coders within the multi community seem to interested only in things like OEB and as a result FS2_Open's net code has barely been touched in comparrison to the changes made elsewhere.
Karajorma's KotH is a very good initiative. It seems to be a real fun :) I will test it tonight with some ppl via IP. But for more multiplayer interests that kind of initiative must be supported by all cases that I write above.
The main reason why I haven't FREDded more stuff like that is due to a lack of interest. If I can get feedback on my missions I'm happy to write them. I've got a few more ideas for things to do and I'm happy to hear new ones.
-
we need to make it easyer for mods to work in multiplayer. id like to see server side tables, and the ability to transfer files other than missions. this makes small mods easy to install in multi games. the server side tables means that you can play even if you dont have the mod installed. models and textures could be transfered as well or there could be substitution tags on data files. if the game cant find the data, it tries to look in toe following +substitution: variable. which should point to something in the vps (open or vanilla) that can take its place. not sure if that would work for models. like if somone was running a vanilla herc and somone was running a htl herc, i always wondered if something like that was enough to cause a problem.
-
What are the bandwidth/space/performance requirements of a gaming server? Coul HLP feasibly set one up?
BTW, CtF missions are easier than you'd think. It could be made pretty cool I think. If the flag was a solid, invulnerable object, rather than a nav buoy or something, you could actively hide it with kinetic weapons (around some sort of base - an Arcadia or hecate, or something with a lot of corners) when it wasn't in the opposition's possession as a Freespacey twist on CTF.
Also, I have the basics of a Freespace soccer multiplayer mission somewhere, where the goal is to get your (invulnerable) cargo container into the opposition's jump node. It wasn't that hard to FRED either - I had a rectangular field defined with the get_ship_x/y/z for the container, and it just stopped moving when it got outside those regions (every time, get ship x, ball, greater than 1500 or something, set ship x, ball, 1500). Score with distance from waypoints in the centre of the nodes, and kept with variables. You equipped everyone with kinetic weapons and either blasted away at the players (massive respawn counts) or tried for the ball.
Probably would have been quite fun with some tweaks, but alas, never finished.
-
I'd had thoughts about doing the latter of those two myself :)
-
If we could get a good ingame menu for joining games, and a reliable server, I would definately play multiplayer again. (I did briefly back in 2000). I think those two are the biggest issues to be conquered.
-
The server itself is fairly reliable. It's only down at the moment because MP is switching from Windows to Linux.
-
Been trying to play fs2 for a week online now and no luck. I have the retail version of fs2 and i think i got it working using the fs2 open launcher(version 3.6.5), i've added the modified multi ai cinfig file for pxo and the fs2 open 20050301 exe file, but i dont see any games in multi. Never got a confirmationemail from FS2netD after I registered my pilot and I cant logon. Fixed the "connection failed" error i was getting by doing a fresh install of FS2, without patchint to v 1.2. Shouldn't I be able to add ip addy's and play from regular launcher just by using multiplayer thats built in and checking the box pxo version isn't checked? Just thought that was easier for ppl with the retail game.
-
You can still play IP games with both retail and FS2_Open. Nothing is stopping you doing that. The problem is finding multi people to get an IP address from and run a game.
Which is why FS2NetD is so useful.
-
I want a 'capture the flag' mode. :lol:
Well, that comment inspired me to try something wierd, ...
..., I have started making a wierd Soccer map, with a poor Hermes shuttle as ball and two x two VACs as Goals, the hermes is pushed around useing the Morningstar.
The current *version* has some issues, though,(If it even could be called a version, ...), mainly that the only thing triggered by a goal is a instantenous supernova (just for testing the goal balance), but also with the goal detection (the problem is that I have found no way to check wether something is in a cubic area, the only way to check the goals is useing distance to a waypoint, the problem should be solved with a cluster of waypoints ...)
-
Or you could use a custom jumpnode model with different colors for each team. If you open up a mission in notepad and do a search for "$Jump Node:", after each entry you can add:
$Jump Node: 1.0 100.0 40.0 ;;Position
$Jump Node Name: Epsilon Pegasi ;;Duh
+Model File: jumpnode.pof ;;Special model file for this jump node
+Alphacolor: 0 255 0 255 ;;Jumpnode color (RGBA)
+Hidden: NO ;;Can you see the jumpnode?
There are of course SEXPs that do the same thing.
Unfortunately these have never been added to FRED, so if you save the mission from it, the new fields (ones with "+") will be lost.
-
My skill with maths is not what it once was but it seems to me that using a cubic goal area defined by waypoints is going to be a complete nightmare to get working. Personally I'd use the get-object-x,y and z SEXPs in FS2_Open instead. You can directly compare the location of the goal with the location of the ball that way. Saves you having to faff around with distance sexps completely.
-
Or just have a spherical goal.
Edit: Actually, I could probably code an is-within-box SEXP.
is-within-box
-- object to check
-- top front right x
-- top front right y
-- top front right z
-- bottom back left x
-- bottom back left y
-- bottom back right z
Or:
is-within-box
--object
-- center x
-- center y
-- center z
-- box width
-- box height
-- box depth
It'd require a fraction of the work needed for it with SEXPs and would be much faster too.
-
Or just have a spherical goal.
Far too easy :) I'd already thought of that and dismissed it as not challenging enough :D The SEXP would be nice though :)
-
SEXP is in CVS. I changed it a bit:
num-within-box
-- box center X
-- box center y
-- box center z
-- width
-- height
-- depth
-- ship or wing
-- ...
Note that if a wing is specified, it will only count as _one_ object for the count. So specifying "Alpha" and "Galatea" would give you 2, assuming everyone in the wing and Galatea were in the box.
-
The is-within-box returns a boolean right?
Well if the goal would be a Ganymede, it could be spherical as well...
-
It would be kinda pointless to do it with a sphere since you can just use a waypoint in the centre of the sphere. Unless WMC fanices giving us a is-distance-from location SEXP too. I'm sure I'd eventually find a use for it that wasn't just elimination of an unneeded waypoint :)
BTW I suppose I should point out that I like the first version of the proposed SEXP as by the look of it that allows non-rectangular boxes. :)
-
Edit: Sorry misunderstood your post ...
// You know that there is a distance SEXP, but you need to use comparing operators (<, >, =) with it (well rather logical considering that it gives a integer...), when done in an event it returns a kind of sphere region (gonna try and mess with the proposed SEXPs, though, a rectangular goal would be way easier)...
Anyways, I will try to get a first playable version up by today or tommorow.
-
It would be kinda pointless to do it with a sphere since you can just use a waypoint in the centre of the sphere. Unless WMC fanices giving us a is-distance-from location SEXP too. I'm sure I'd eventually find a use for it that wasn't just elimination of an unneeded waypoint :)
BTW I suppose I should point out that I like the first version of the proposed SEXP as by the look of it that allows non-rectangular boxes. :)
Ehh, you should be able to do the same things with one that you could do with the other. Either way, all angles would be assumed to be 90 degrees. Basically if you want to convert: width = (top_front_right - bottom_back_left)/2, center_x = (top_front_right - (top_front_right - bottom_back_left)/2) and so on for the other two dimensions.
However it makes it harder to calculate because you'd have to throw in checks for if someone swapped sides, as 'left' 'right' and such are all relative.
-
Oh. I see. I made a mistake reading the post :)
Yeah the second one is better :)
-
I can think of a use for such a sexp already.
"Hold fire until you are within range X of the targets. Even if they're already shooting at you. I repeat, hold fire until range X!"
-
You can use Distance for that though surely.
The only case I can think of for needing a spherical SEXP would be if you had something like a mission where you had to be within a certain range of the front of a moving vessel. In That case you couldn't simply use distance from a subsystem because that would still be true if you were behind the ship. Using the SEXP you could work out where the front was and then check if someone was say 1000m in front of it.
It's a bit of a stretch though.
-
I remember the thread about doing a soccer game from ages ago, I still think that'd be fun, especially considering how deadly tackling would be ;)
-
More multiplayer features that I wish to see (and not mention in my previous post in this topic) is able to record 'demo' / 'replay' from the game, and play it anytime you want. Just like it is in Doom / Quake series and many other games. It can't be hard to do this (n00b programist think :p).
Another (which is mentioned by Karajorma yet) is to improve variables in game, that clients can get modified values as good as host do now. I play Karajorma's King of the Hill and it was very good and fun :) (my feedback is here (http://www.game-warden.com/forum/showthread.php?t=1382#21) @ GW). Some code-work on variables can improve such ideas like KotH much.
Third for now: is to improve standalone application. Make it just a little bit more stable and add advance options like for example: DaBrain's idea about 24h in-game join server (http://www.game-warden.com/forum/showthread.php?t=1156) and more :)...
-
I think recording demos/replay might prove quite hard if the hooks to record everything aren't already in place in the code. (PS don't say something is easy if you don't want to be told to do it yourself :) )
The problem with getting variables fixed for multiplayer is that there is a very real chance that trying to get variable changes to update the clients could actually break multiplayer. We won't know till someone tries though.
As for DaBrain's 24h server I think I've already made clear I've got some very big plans for such a beast :)
-
I think recording demos/replay might prove quite hard if the hooks to record everything aren't already in place in the code. (PS don't say something is easy if you don't want to be told to do it yourself :) )
There are hooks already actually, they just aren't complete or optimized in any way. I had at one time cleaned up the demo recording/playback code in order to do large and complex runs through a memory debugger but I no longer have that modified code. And the demo save files were quite large, upwards of 35meg per mission, so it would need a lot of work to be generally usable.
The problem with getting variables fixed for multiplayer is that there is a very real chance that trying to get variable changes to update the clients could actually break multiplayer. We won't know till someone tries though.
Goober asked me about this not too long ago. It's easily possible to do it without breaking old clients, however those old clients would ignore the variable changes. Adding it wouldn't be difficult (at least for me since I know pretty much what to do) so I'll probably end up making the changes for this before too long, I just need the time. I gave Goober the simplified rundown down on adding it though so he may decide to brave the multi code and add the changes himself before I get the chance.
-
Excellent :)
It's not really a problem if older clients ignore the variable updates for now. I can just warn everyone to have a recent build and I've designed the missions I've made so that the default value for any string variable is something which is at least readable :)