On Sunday 07 January 2007 12:37, Mathias Fröhlich wrote:
> I don't believe that this could be done reliably without crude hacks in the
> physics of such a tow.
>
> The problem is the *changing* lag between two multiplayers.
> The lag is optimized to be as small as possible, but sometimes it happens
> that the connections gets slower or just stops feeding packets for some
> time. During that the lag increases dynamically to avoid excessive
> extrapolations. If you get packets just in time, the lag decreases.
>
> If your computers are in a LAN the problem is way smaller.
>
> The alternative to a dynamically changing lag would be to use a fixed lag
> of course. But how would you choose that fixed lag?
> - By the first packet you get? You got that despite of a bad connection
> relatively fast and for the whole multiplayer session you have that
> multiplayer jumping around because of the excessive and most probably wrong
> time extrapolations into the future. That jumps would also fool the physics
> BTW ...
> - By some fixed several seconds to satisfy even worst connections?
> Then forget interactivity if you are in a LAN and probably have dynamically
> adjusted lag to a few msecs.

Yes, this is difficult to get right.  ID Software uses 'prediction' code in 
the Quake series, so clients 'predict' a lot of things from 'last known 
locations'.

If you get severe lag during an online game, and then packets caught up again, 
all of a sudden the client would backtrack (if the prediction was wrong) and 
reset the new location.

9 times out of 10 you ended up dead in this situation as you wasn't where you 
thought you were.

Nick

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Flightgear-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/flightgear-devel

Reply via email to