PING

Thanks?

Just to clarify:

Any ping under 350 is perfectly acceptable as the game has built in simulated-LAG for this amount.

Any SIM Speed over 1 is perfectly acceptable, as your SIM Speed is the number of times faster you're capable of running the game at that moment - SIM Speed of 5 means you can run the game 5 times faster than it is currently running. This is a residual feature from the Supreme Commander engine.

 

My fave is when someone tells me I am the source of lag because my sim speed is 5 and his is 9.  Please.

This made me lol!

Haha yeah, I've seen players get flammed alot for having SIM Speeds of 4 in the end game. Get tired of correcting people some days

man i played 5v5 with american players with 380 pings that lag less then 3v3 with 200 ping...why cause it depends on how stable ur conn is...if somone sayts they got a stable con give it a go? whats the wrost u remake OMFG

Demigod now has a remake option?

i think hes just glad he got that one off his chest...

 

 

rehost...remake?

Speaking of Australian gkrit has no problem playing in my games unless there are people from russia, america and australia in it. If its just gkrit and a bunch of americans there is no problem. His ping is even better than some Americans.

mmmmmmm.....:P

 

I thought P2P was chosen because it is more reponsive and has less lag than client/server? Now its the case there is an artificial lag of 350ms?

Every RTS has an artificial lag built in. StarCraft has 250 ms in Battle.Net for example (100 ms in LAN). This is made to provide fluent gameplay.

"Every RTS has an artificial lag built in. StarCraft has 250 ms in Battle.Net for example (100 ms in LAN). This is made to provide fluent gameplay."

1.  Every RTS has built in lag.  Are you certain of that?

2.  Not really sure how that is relevant, anyway.  Can we never hear the argument "Demigod uses peer to peer because it requires split second timing" again?

Quite certain. In case of StarCraft there is even a tool to announce LAN games to Battle.Net in order to trick StarCraft into thinking it's a LAN only game and thus using a netlag value of 100 ms instead of the default 250 ms over Battle.Net, even though you are still able to play over Battle.Net. iirc WarCraft III uses 350 ms just like Demigod.

Right, so Starcraft and Warcraft have it, so every RTS must have it?  And again, don't really see how that is relevant.

No, that's not what I tried to say. I think I posted the netlag values of other games like World in Conflict and C&C in another thread here on the Stardock forums, maybe I can find it again.

Right, and what does that prove?

You were wondering about the default net_lag of Demigod and made it sound like this is something out of the ordinary and I just wanted to clarify that.

Nope, I'm saying we were told the justification for Demigod using peer to peer was because there was less latency than client/server in order to make split-second actions.  Introducing simulated lag just shows thats BS.

The peer to peer system is used for two main reasons:

• The Moho Engine runs a synchronized simulation. Meaning every client calculates everything and the calculation results are compared with each player directly. This makes cheating without exploiting bugs in the system virtually impossible. If someone tries to force a cheat on to the system, results will mismatch and the game will desynchronize (this of course can be exploited to avoid losses, but that's another story).
  
• Latency between players is reduced, since they all have a direct connection to each other. In a server client system, the information from one player to another always goes through the server first.

I don't think anyone has phrased it like that. And even so, 350ms is still a fraction of a second, technically .

Without a "simulated lag" every RTS game that uses some form of timed data synchronization would stutter all the time, no matter how low your ping is, regardless of whether it's peer-to-peer or client-server.

"The Moho Engine runs a synchronized simulation. Meaning every client calculates everything and the calculation results are compared with each player directly. This makes cheating without exploiting bugs in the system virtually impossible. If someone tries to force a cheat on to the system, results will mismatch and the game will desynchronize (this of course can be exploited to avoid losses, but that's another story)."

That's a lazy dev's answer.  If cheating is going to be a problem, put in an anti-cheat.

 

1)      "Latency between players is reduced, since they all have a direct connection to each other. In a server client system, the information from one player to another always goes through the server first.""

2)      "I don't think anyone has phrased it like that. And even so, 350ms is still a fraction of a second, technically"

The second sentence renders the first moot.  If you're adding 350ms to the latency, the advantage of using a direct connection is taken away.

The synchronized simulation is the anti-cheat system. It's not a lazy dev's answer... such a system is quite complex.

You still don't understand. It doesn't matter how low your latency is. The game would still stutter if no netlag is used in the system. The default value of 350 ms is arbitrarily set, so that many people around the world can play with each other fluently. If it was set to 50 ms for example, you could only play with people in your neighborhood for example.

The netlag simply amplifies the range of people who can play with each other and so does the p2p system. If the system wasn't peer to peer and the netlag value would stay the same, less people would be able to play with each other without stutter.

"The synchronized simulation is the anti-cheat system. It's not a lazy dev's answer... such a system is quite complex."

It is the lazy devs answer if the synchronized simulation is creating a huge problem in itself... lag.

"You still don't understand. It doesn't matter how low your latency is. The game would still stutter if no netlag is used in the system. The default value of 350 ms is arbitrarily set, so that many people around the world can play with each other fluently. If it was set to 50 ms for example, you could only play with people in your neighborhood for example."

Nope, you are the one missing the point.  I'm not saying whether the game would or wouldn't stutter - I'm saying there is no advantage to using p2p over client/server from a latency point of view.  This is what we were told when the game was released. 

"The netlag simply amplifies the range of people who can play with each other and so does the p2p system. If the system wasn't peer to peer and the netlag value would stay the same, less people would be able to play with each other without stutter."

The logic behind this is flawed but there is no point in even explaining why, when I can just point to games like Heroes of Newerth and League of Legends (both client/server) which run 5 vs 5 games with no lag and Demigod which (95% of the time) can't (what are the chances of getting 10 people in a pub game all with great connections?)

95% of the games played in HoN and LoL are 5vs5.  Why is it not the same in Demigod? Because the players have no confidence they will be able to get a smooth, robust game.

This is a post from Frogboy back in April:

 

"In Demigod, it is peer-to-peer. This was necessary because it needed to have the least amount of latency possible since your moves are directly sent to the other players rather than to a server first.  Because the action in Demigod focuses a great deal on Melee in the third person (i.e. players can see their hits) a lot of traditional methods to mask latency aren't possible. Peer-to-peer reduces the problem of "lag"."

 

Yeah, right up until you stick 350 ms artifical ping onto it!