Quote:
Originally Posted by
Sheleft
My definition of warp possibly differs from what's commonly known as warp. I call any jittering and appearance of frame skipping warp.
I personally knew what you meant, but this could be a source of confusion for others. Consider using more precise terms.
Quote:
I suggest that this was the cause because without setting timenudge my netgraph and connection was cleaner. I think technically it should not cause an actual increase in lost packets but rather change how info is shown to me, ... unless I am mistaken?
You are not mistaken.
Quote:
But for whatever reason my netgraph jumped quite a bit after setting it.
That is very odd. I ran a test to be sure and it does not do that for me.
First of all how to interpret the lagometer:
**************************************
Lagometer is interpreted as follows (Yakumo):
Top bar is prediction.
Blue is interpolation between two valid frames.
Yellow is extrapolation since the last frame (waiting for another).
Bottom bar is ping/packets.
Green is ping for a properly received snapshot.
Yellow is a properly received snapshot, but the previous one was suppressed to stay under the client set rate.
Red is packet loss.
**************************************
Here is what various timenudge values look like with 48 ms, maxpackets 125, rate 25000.
http://img833.imageshack.us/img833/8265/graphs.png
As we can see in this picture the time to interpolate is cut down and out completely as timenudge is lowered from 0 to autoTN (which is < -20).
Ping/packets are unaffected.
Quote:
Wouldn't this visible increase in lag be seen by other players?
Changes in the bottom bar have the potential to have effects that can be seen by other players, changes to the top bar are a different story.
Quote:
If on my screen a player is in front of a doorway and I fire a rail, but on his screen he has passed the opening and behind the wall yet is still hit... who is receiving the correct information from the server?
According to the server, the player in question was not behind the doorway when you fired the rail.
I'd say that it is more likely that he was lagged and you were not, as his client was further out of sync with what was really happening.
Quote:
Shouldn't the player with the lowest ping naturally have a better representation of what's happening shown to them?
As far as I know, they do.
Quote:
I've been on both sides of this now and it just seems odd that with a good solid 30 ping on my cali servers I am railed by someone with much worse ping when from my unlagged pov I clearly beat him to the correct defensive position.
This is a consequence of the anti-lag netcode. It is bound to happen to even lower pinged players, albeit much less than for higher pings. Another thing that could have happened is that you had a lag spike right when you were taking cover.
Quote:
I've read that people can jitter with timenudge set a certain way, is this incorrect?
No!
Quote:
A good player with 10-20 ping and that slight jitter is insanely frustrating to hit for me with my settings.
Nobody likes to play against warpers.
As far as I can tell, fast fluctuations in ping time greater than a certain amount of deviation are what cause the warping. So someone who alternates between 40 and 60 ms really fast is likely to warp, and their netgraph appears very jagged like a saw.
Quote:
With the warpers and feeling like I am playing in mud this timenudge experiment really felt to me like I was on to something that could be behind it.
Understand I am not saying the intended effect of the command is an issue I am saying that if it did indeed cause my connection to spike a bit with increased packetloss then that's the problem.
Timenudge will not cause an increase/decrease in packet loss or ping.
Quote:
Lorf, I assume you are someone who does quite well in games,
Well is a relative term :-)
Quote:
is very hard to hit and has timenudge set differently for each weapon? :)
I don't know if I'm hard to hit or not. I've been told by a few players that I have strong dodging but you know personal testimony is not good evidence.
Right now I'm using -9 for all weapons except for the RG/LG/CG which I have set to 0.
I've only been told that I warp three times that I can recall, two of those were when I was testing a VPN which was deteriorating, and I had plenty of indications client side that I was warping anyways (and the experience was awful). Once by a player who was warping very badly himself, I didn't say anything because he was already like, psychotically upset over losing :-/
Other than that I'm not known as a "warper", and every warper that I've encountered has had the characteristic "sawtooth" spikes in the bottom bar of their netgraph. I do not have these spikes.