logoalt Hacker News

xeonmc • today at 6:08 PM • 4 replies • view on HN

    > Does it cause pop-in when peeking?
    The goal is early reveal, not exact last-millisecond reveal.
    CS2FOW predicts using movement and ping, reveals enemies slightly before exact visibility, and keeps revealed enemies visible briefly. This intentionally leaks a small near-corner window to avoid late pop-in.
This fails to address the main point of the "pop-in" issue relevant to fog of war systems, which is that it is the victim of the peek that gets the worst pop-in effect, the peeker much less so. The aggressive peeker gets the benefit of the early-prediction from the server since they're the initiator of the movement, whereas the victim only begins to receive the information after the peeker has already gotten two network roundtrips worth of early prediction.

Replies

the_gipsy • today at 9:49 PM

What is the reason for thinking "early, not milliseconds" would not mean "early enough for the peeked/victim too?

Ultimately the server must decide when to "pop" players, regardless of client interpolation which can only happen after the pop. So why would the server delay the pop for one player?

bakies • today at 6:12 PM

Why wouldn't you just send the positions to both clients in the same tick? Seems trivial to solve.

➕ show 3 replies
webstrand • today at 6:10 PM

Interesting, could it be mitigated by the server doing its own prediction and defogging the peeker early? Or is lag prediction in CS2 not entirely client-sided.

➕ show 1 reply