Multipayer-small Question, Not Asking You To Put It In!

[Irenices]

Well as i said i am not asking you guys to put multiplayer in as i know thats not the point of this mod ect.. Im wondering if once its released would it be possible for someone to add it in or have you changed to many aspects of the game where it would be nearly impossible?

 

 

 

Just dreamin about coop as usual =P

[Gildoran]

From what I can tell, it doesn't look like people have been trying to keep code changes safe for multiplayer (and doing so could entail extra difficulties), so trying to make a multiplayer mod based off of TDM might be hard... It'd probably be easier to write a multiplayer mod and import assets/code from TDM.

[Crispy]

I agree with Gildoran. It wouldn't be impossible, but it would be a fairly substantial project. Still, it's not like you'd have to make TDM from scratch all over again.

[Maximius]

Assuming that D3s multiplayer capabilities would serve as a foundation, and with textures and objects and such from the Mod to alter the environment, whats a time frame for such a multi-mod? Are we talking about another couple of years or is this something that could be pulled off quicker?

 

Its a beautiful idea, I know there are other mods out there trying to do similar stuff but IIRC they aren't trying to stay as close to the spirit of the precursor games as the Mod is. My thoughts were that it should be a matter of two teams versus one another in a race to loot a map, as well as face off with the AIs. All sorts of alliances and treacheries could be savored, setting the AIs after your opposing team, taking them on in an alliance, selling your allies out at the last second then racing to jump out the window to freedom and victory...

[Gildoran]

Because it would be able to draw from existing code and art, I doubt it would take as long. Then again, many things would probably need to be programmed again, because they currently assume there's only one player and they don't worry about sending information over the network. So... I have no idea how long it would take.

 

On the issue of alliances/treacheries... I have a hard time imagining it working in an MP FPS game. I suspect players would stick with their teams, and punish any treacheries by repeatedly killing the perpetrator. I suspect anybody who helped the other team would be considered no different from TKers in CS. Sooner or later, most servers would probably end up with automated options for teams to punish any of their members that work against them. In order for alliances and treacheries to have much meaning, you need a much longer lasting game with larger consequences, such as a MUD.

[Maximius]

Because it would be able to draw from existing code and art, I doubt it would take as long. Then again, many things would probably need to be programmed again, because they currently assume there's only one player and they don't worry about sending information over the network. So... I have no idea how long it would take.

 

On the issue of alliances/treacheries... I have a hard time imagining it working in an MP FPS game. I suspect players would stick with their teams, and punish any treacheries by repeatedly killing the perpetrator. I suspect anybody who helped the other team would be considered no different from TKers in CS. Sooner or later, most servers would probably end up with automated options for teams to punish any of their members that work against them. In order for alliances and treacheries to have much meaning, you need a much longer lasting game with larger consequences, such as a MUD.

 

 

Interesting discussion but honestly I wasn't thinking of treachery between teammates, but rather between teams. If you have a difficult map, or a section of the map thats really hard, the two opposing teams could call a truce and agree to work together for the duration of the obstacle. Say two teams are raiding a temple but there is a bad ass monster guarding a doorway both teams have to get through. The teams could say Truce until we kill that thing, and then work together. But in the back of your head you have to watch and listen for the other team turning on you at the last moment, perhaps running off and leaving you to face the beast or who knows. Is this problematic?

[Nyarlathotep]

On the issue of alliances/treacheries... I have a hard time imagining it working in an MP FPS game. I suspect players would stick with their teams, and punish any treacheries by repeatedly killing the perpetrator. I suspect anybody who helped the other team would be considered no different from TKers in CS. Sooner or later, most servers would probably end up with automated options for teams to punish any of their members that work against them. In order for alliances and treacheries to have much meaning, you need a much longer lasting game with larger consequences, such as a MUD.

Anyways, Doom 3 only allows for 4 players at a time. That's fine for co-op, but anything else might be slightly tricky. Doom 3 sends pretty much everything that happens on a client's system, including phys-props. It's great for co-op, but anything slightly more epic than a four-player game requires slightly smarter netcode.

[Gildoran]

How do you imagine players communicating their desire to form a truce? If you see somebody from the other team, wouldn't you be likely to kill them before they have a chance to kill you? Keep in mind that unlike a guard, they're as weak as you are, and it would give your team a significantly better chance of winning if the other team is understaffed or injured. And couldn't it be hard to fill a server with enough players to get past the obstacle? I guess you might be able to scale the obstacle based on the number of players, but I have trouble thinking of interesting obstacles that are scalable.

 

@Nyarlathotep: I'm not sure about this, so don't take my word for it, but... I thought I heard somewhere that it's possible to increase the maximum number of players in a D3 game. Also, it seems like D3 gives you flexibility about what to send or not to send over the network. (for example, I'm under the impression that brittle-fracture physics aren't propagated since they're purely aesthetic)

[Nyarlathotep]

How do you imagine players communicating their desire to form a truce? If you see somebody from the other team, wouldn't you be likely to kill them before they have a chance to kill you? Keep in mind that unlike a guard, they're as weak as you are, and it would give your team a significantly better chance of winning if the other team is understaffed or injured. And couldn't it be hard to fill a server with enough players to get past the obstacle? I guess you might be able to scale the obstacle based on the number of players, but I have trouble thinking of interesting obstacles that are scalable.

 

@Nyarlathotep: I'm not sure about this, so don't take my word for it, but... I thought I heard somewhere that it's possible to increase the maximum number of players in a D3 game. Also, it seems like D3 gives you flexibility about what to send or not to send over the network. (for example, I'm under the impression that brittle-fracture physics aren't propagated since they're purely aesthetic)

That may be true. Unfortunately, I don't know. I only know the reasons why D3 was limited to 4 players. As for brittle-fracture physics (glass-breaking), I had forgotten that it was deemed as special effect--although, it would be interesting if glass actually caused damage.

[Crispy]

I thought I heard somewhere that it's possible to increase the maximum number of players in a D3 game.

Yep. I read elsewhere of a "lagless 10 player game", which bodes well for the performance of the netcode - though you may need broadband for that.

 

That mod I linked to increases the limit to 32 players, which should be plenty for a Thief-like game.

[Nyarlathotep]

That's impressive. Damn impressive.

[kohan69]

a multiplayer mod would be much easier for a player-only game, where no ai is involved.

 

10 guards vs 3 theifs? splinter cell style? possibly or something like the rules of THIEVERY

[Gildoran]

Don't forget that for a multiplayer mod, we have to understand the netcode and design around it. I don't really know how D3's netcode works or what the caveats are, and I'm assuming that's also the case for our other programmers (maybe I'm wrong). Having to take into account things like client prediction seems difficult enough to make me prefer working on a single-player mod, even if it requires AI.