Removal of trigger lag...

[XyZspineZyX]

Now when UBI says no trigger lag, are they going to do something where all they do is add a client side animation but the bullet appears a second later like AA? Or are they going to add a client-side netcode?(I hope)

Of course the worse thing they could do is add client side bullet effects and weapon effects while using a server-side netcode, which is what BF1942 does(what the hell was DICE thinking...)

[XyZspineZyX]

Now when UBI says no trigger lag, are they going to do something where all they do is add a client side animation but the bullet appears a second later like AA? Or are they going to add a client-side netcode?(I hope)

Of course the worse thing they could do is add client side bullet effects and weapon effects while using a server-side netcode, which is what BF1942 does(what the hell was DICE thinking...)

[XyZspineZyX]

1.3 patch said it fixed bugs, so you go figure...

10
9
8
7
6
5
4
3
2
1
1
1
1
1
1
1
1
1
1
1
1
1

[XyZspineZyX]

BopICOPI wrote:
- Now when UBI says no trigger lag, are they going to
- do something where all they do is add a client side
- animation but the bullet appears a second later like
- AA? Or are they going to add a client-side
- netcode?(I hope)
-
- Of course the worse thing they could do is add
- client side bullet effects and weapon effects while
- using a server-side netcode, which is what BF1942
- does(what the hell was DICE thinking...)
-
-
-
I am no netcode expert, but I can tell you that BF1942 accuracy, timing, and overall gameplay (as far as performance), is right now far superior to Raven Shield...



http://www.changingseasonssale.com/john/sig1.jpg

[XyZspineZyX]

BF1942 is great,but seriously, what the hell was DICE thinking whenthey put ALL client-side effects in a server side game? Where they trying to make it look like client-side, so all the new players get really confused and not get any kills?

[XyZspineZyX]

I think UBI needs to come out with what exactly are they doing. I pray it is a client-side netcode.

Client-side netcode + PB + good game = heaven

[XyZspineZyX]

The naming of Client Side and Server Side netcode is actually wrong.

Client side netcode usually means that the connection is "Peer-to-Peer" which means everyone connects to everyone and there is no server. This netcode has the bad that if one player lags in one server, all will lag as well but it has the good that noonce has the advantage as everything is synchronized. As far as I know only Strategy Games use Peer-to-Peer netcode and some earlier version of the Serious Sam game. Btw Doom 3 is going to use Peer-to-peer as well for the cooperative multiplayer.

Server side netcode is used by almost all the games except the strategy ones in which all players connect only to the server. and there is no direct communication between the players at all. This netcode is good because it uses less bandwidth than peer-to-peer but has the bad that people with better connections to the server always have the advantage.

So the ClientSide or ServerSide Netcode naming is wrong. In reallity we should say Server Side or Client Side "effects" not netcode. As netcode is Server Side always.

ClientSide effects in 1.3 are the player movement and nothing else. With the 1.4 the weapon showing to fire will be also visible.

All other effects should remain Server Side from what I understand and that is good. If they do the effect of bullets really going off the weapon (not the effect of the weapon firing) to be client side then there will be lots of problems with inaccuracy.

[XyZspineZyX]

KaTaFRaKTiS_ wrote:
- The naming of Client Side and Server Side netcode is
- actually wrong.


No, it's quite correct - but you are talking about the wrong thing completely. Counter-Strike uses client-side code, for example. Note the word "side".

What people are talking about (and the terminology is quite correct) is where the computations are performed - specifically the "to hit" ones. The fundamental difference is that with client side you "hit" if you shoot at what's on your screen.. that may or may not be where your opponent (to him) actually IS, that depends to a large extent of your relative pings and the quality of the prediction code. It's usually pretty good out to pings of 250 or so. Server side code is based on you "hitting" an opponent in his absolute position - i.e where the server thinks he is. At anything above very low pings this entails "leading for lag" - i.e shooting at where you think your opponent will be in ping/2 seconds and anticipating accordingly. That's where the "trigger lag" comes in - you fire at time ping/2 and this actually get's shown on your screen after time "ping" (the return signal from the server telling the client it's fired). RvS has an additional problem though.. for some reason, this trigger delay is much, much longer than it should be for a given ping, for some players anyway.

My guess is that Ubi are switching to client-side code - there's no other way of doing it (ya canna break the laws of physics..)



Message Edited on 06/28/0306:25PM by Hertston

[XyZspineZyX]

UBI could just put in the client-side effect of firing the gun, but the bullet appears half a second later, like in Americas Army. That would anger alot of people.

[XyZspineZyX]

Yep. This worries me.

Is UBI going to give us genuine netcode, where the server uses calculations to compensate for lag?

Or is it just going to be an illusion, where the gun fires the moment you press the mouse, but all the underlying game code remains unchanged? Basically, the trigger lag is essentially the same as before but disguised better.

If they are going for the second option, I would rather they just not bother. I would feel less cheated.

[XyZspineZyX]

cmon UBI, we need to know.

[XyZspineZyX]

someone posted in one of the other threads on this.....about UT2K3 using a client side animation to make it LOOK like no trigger lag...while actually leaving the code server side

honestly I'd almost bet thats what they do rather than go to client side

http://members.cox.net/battleratt/GrimJack.jpg

[XyZspineZyX]

ttt hate to say I told you so...

The 3rd iteration of TitoTrinidad

[XyZspineZyX]

They changed it to be exactly like AA:O

Check out this thread with a video experiment I've made:

http://www.ctuclan.net/forum/showthread.php?s=&threadid=394