Needs serious multiplayer improvements

[NullProgrammer]

I cannot comment on the other two because I've only played EA's RTS's a few times, however the times i've played them my games have never just stopped even if the host left mid-game. I've never had a blizzard RTS end prematuraly even if the host has left. Warcraft 2, Starcraft, or Warcraft 3. It seems the problems on your end because i've never had a problem.

So it seems theres a lot of people here content with the game simply just ending when the host decides its over then? No one cares about actually solving or want to provide support to solve the problem even though its a major game block? Even if you mentioned two games that do Chose to fail to work correctly I don't see why Multiwinia needs to. I will never accept bad game design. Maybe thats why the game community has just gone down as of late. People simply accept bad game design.

[SunKing]

Is it really that much of a big deal? It's certainly not a "major design block". As far as I'm concerned, if the host leaves during the middle of the game then I win. Simple as that. Is it that important that the computer tell you that you win?

[Phelanpt]

I would still like a reply to my post. You keep mentioning EA and Blizzard, which are huge software houses. What about small companies?

[Shwart!!]

I would still like a reply to my post. You keep mentioning EA and Blizzard, which are huge software houses, and they still can't do it right. What about small companies?

Fix'd.

Shwart!!

[gulfy32]

I have played many games in WC3 and been disconnected because the host left. Many times the host dies, he leaves, and everyone gets a pop up saying "you were disconnected."

[The GoldFish]

On principal, I agree with NullProgrammer.

IVs architecture is not server-client or peer to peer, really. It's actually more the case that the "server" is running both a server and a client, all the clients connect to their server, and then the only things that are communicated is chat, user input and game checksums. All the clients are smart clients and have all the data.

The result of this, is that, if the host drops, the game will be over. If there is a sync error, the game will be over.

There should be no disconnects, and there should be no sync errors, but, of course, there always will be. There is no accounting for server disconnects, either malisciously or through no fault of their own, the game could, actually, pause, analyze if there's anyone else everyone can connect to, and choose them as the new server, and just carry on. The netcode is, in fact, pretty ideal for this purpose. If there are sync errors that result from 2 events not yielding the same result, then it should try to actively rectify that.

Neither of these are not especially high priority. Unlike a 2-3 hour long strategy extraveganza, Multiwinia games last only about 10-20 minutes. What this means in practice is you get slightly less annoyed when the host quits, but that this event can happen more often.

Small company or no, this issues should be dealt with; not because big companies do it, or because "every other strategy game ever" does it, but because it's a simple concept and a good idea for the enjoyment of the players. Because fun is the actual ultimate reason for playing.

[Borgatt]

I'd tend to agree that it seems weird. I dont play any RTS other than this, but I imagine there should be a way to say that "host quit, " instead of connection lost. If there could be a confirmation along the lines of "if you quit you forfiet" for any player if you try to leave that would be cool. It's anal, I know, but I'm not any good at most games, and its not often I get to see "Borgatt wins" in fancy lit up letters. It vain, but its less frustrating than the vauge "connection lost" message as soon as I take somebodies second last spawn point.

[Nimbus]

C&C fan here; RA2 and Generals are completely P2P: The game continues when the host leaves.

[Major Cooke]

Please include passwords... 'Nuff said.

Oh, and kick/ban options.

[NeoThermic]

Incorrect, Warcraft 3 and Starcraft use peer to peer, or similar server/client systems where if the host leaves the game does NOT End. The hosting is either transfered to another client which now becomes the server, or it uses flawless peer to peer. I think you should relook at your list.

The docs say it's client/server. They also fail to mention anything that hints that the gamestate is transferred to another client for it to become the server.

You also need to learn to make a coherent argument that doesn't faultier from your original stance. I quote:

You cannot have an RTS thats runs on the server/client. There is no RTS in existence which runs on server/client for that Exact reason.

Yet we've just shown that there are RTS games that do.


Next you're going to complain that the game runs over UDP, or that the packets are encrypted, etc.

Should it be changed? Possibly. Does ranting with a viewpoint that alters quicker than the value of the USD value help? No.

NeoThermic

[Shwart!!]

C&C fan here; RA2 and Generals are completely P2P: The game continues when the host leaves.

Again, not from my experience. Though in these cases I've never had a game continue with the host gone, not even once like Warcraft- which, according to NeoThermic, shouldn't even have happened there.

Shwart!!