With high ping the updates just arrive delayed. But they come with same precision, rate and everything like they would with low ping. All is just happening a few msecs behind server reality.
Are you talking about jumpy ping or packet loss scenarios, or what is this all about?

I'm just thinking out loud, trying to understand all aspects a feature like this would affect and what exactly would happen. Maybe I'm too sceptical, but it just sounds too good to be true that with a simple trick you could remove the negative effects of lag without any unwanted side effects. Would be nice to hear other devs' opinions about the feature itself and not the abuse, which imo is not that interesting.

I guess I could sum it up like this: I don't want to ever be hit, if I wasn't really hit. If you can guarantee this will never happen, then I suppose I don't have a problem with this.

I would reckon precision doesn't matter, if the information is going to be less accurate than the clients movement prediction.
Otherwise you imply that if a latent player observes someone who is moving in a straight line, the observing client doesn't predict that motion so that they appear where they are in real-time?

Whatever odd scenario you are envisioning through your misunderstanding of the function of player movement prediction could only apply with the originally proposed client-side hit detection. The alternate proposal of backwards reconciliation means that it doesn't matter what you see because the server works out where everyone was when the shot was fired using hindsight.

so after the server decides that you incurred a hit x ms ago, does it apply the knockback to your current position? at your position x ms ago and then applies the possible(valid) effects your movements since that time could have had?

I DID NOT AUTHORIZE THIS DEBATE!!!!

I just learned about this feature tonight, mainly cause i was not active for 1-2 years, during which this "antilag" feature was introduced.

I had heard of this feature before so I wanted to try it some times i remember. But each time i wrote /antilag i got the reply "command: locked". So naturally I assumed this feature was not enabled.

Ooooohhh was I wrong.

Infact this feature is ENABLED by default... and cannot even be disabled. We are talking wargamez, pangela etc. all the servers used for leagues and tournaments.. what the FUCK?

Just tested on a US server + delaypacket... ping 250 vs 25ms... I won ofc and got 40.4% LG
I hearda bout some other disturbing sideeffects from this. That it's impossible to dodge from this antilag. The higher the ping of opponent, the harder it is to dodge. It felt like some weird knockback effect when i tested against another high pinger. Haven't really tested it much.

But after reading this thread it seems that all issues are not quite worked out, and there are several sollutions/fixes to this feature. Then it is UTTERLY INCOMPREHENSIBLE how this feature could be implemented everywhere as default, without even option to disable?

Did admins just sneak this in and hoped noone will notice, so they could come 6 months later "HAH! noone noticed, so you see, everything was ok?"

I guess now I know the secret behind squeeze's improved game and how he can hold his own in div1 against 12-39ms players...

I agree with blAze, if there is even a hint of "aimbot-effects" this at LEAST needs to be disabled untill further development is done.

wouldnt u get more than 40.4% in 13ms pov para?

Anyways. I've noticed something going on in povgames vs hpws sometimes. But i just assumed they had enabled the feature... didnt bother to check if it actually was enabled.

If this is true and has been done intentionally, then whoever is responsible for it needs to be removed from the position that allowed him to make this happen. I have been wondering about some pov games I played, but didn't come to think of this.

had to test some quick povs vs unnamed. The "feature" is enabled and works as designed.

Dont u think this is something different than the rocketdammage 100-120 vs 110? (I mean something to silently test on the big servers, without informing the users. Not saying it was the same person(s) enabling it)

There are two separate issues here, so I made two new threads to discuss them:

The merits of the server side antilag feature

The fact that it has been enabled secretly for months

Well don't you need an updated client for this feature to work? Afaik squeeze uses some older ezq version so I doubt he's been affected by this.

valla no, the new version is purely server side.

Thread closed, continue here.

Note: this thread discussed both client-side and server-side antilag. Client-side antilag is an abandoned scheme at this point. Discussion about server-side antilag continues in the mentioned thread.