unlaggedzor netcodezor

the server checks back in time to when the HPB clicked to see whether his xhair was on you at that time.

I dont know how exactly the unlagged code works but I know prediction cant be perfectly so the crosshair is on a position of a model which maybe outdated (and the correct position maybe behind the corner).
I think even the server gets my position delayed so when the server is up to decide it was a hit they will have to do it once.


Or maybe they "rollback" the decision when the server got my information about my position so it cant be a hit? just joking ;)

Or maybe they "rollback" the decision when the server got my information about my position so it cant be a hit?

Thats actually the most accurate description.

http://web.archive.org/web/20020320065854/htt...cs_report/

That's rather old and talks about the HL netcode, but the underlying principle is still the same.

/edit

seems like i cant link the direct page, so just look at page 6 (Latency issues)

I think that is what I am talking about

btw I noticed this effect in cs 1.5/1.6 the first time

edit: I think I got it. So the decisions were right. I just got the information delayed because of the ping of the high pinger + my ping.

So the decision is right... but is it a good resolution to be able to run, shoot and so on for a small time?
What about the (worst) situation when my nmy got a 500ms ping?
I will be able to do my stuff (killing my nmy :) and after approx 500ms I see it was a waste of my efforts because I am already dead.

Like diablo2... moving for 1-2 minutes through the map after realising im already dead :(

i think the effect is limited to 100ms

Note that the article oversimplifies of course plus I don't know anything about the actual QL netcode. The way it actually works might be quite different and more sophisticated

I notice this issue so I dont think they improved it significantly

tailored hack prolly.

europe: dodgy netcode
north america: bloody hacker

that about sums it up