Re: Breaking matchmaking deadlocks

Yes, pretty much any meaningless number can be used.  The reason I suggested using a timestamp is so that the player who started attempting to connect first would be the one more likely to succeed.  Which isn't a huge deal in the grand scheme of things, but does appeal to my own personal (and slightly warped) sense of order.


Blair Holloway wrote:
> Having thought about it, each user *does* have a unique ID akin to a MAC 
> address. Perhaps all we need to break the deadlock is something like:
> 
> bool shouldAcceptIncomingConnectionFrom(USERID fromUser) {
>   if (NotAttemptingConnectionTo(fromUser))
>     return true;
>   else if (this->localUserId < fromUser)
>     return true;
>   else
>     return false;
> }
> 
> I do worry about the implications of relying on a user's id, though -- I 
> sense it is one of those things that, handled incorrectly, could 
> generate one of those reproducible-once-in-a-thousand-times bugs. :/
> 
> - Blair
> 
> On Thu, Jul 8, 2010 at 8:45 PM, Peter Thierolf <[email protected] 
> <mailto:[email protected]>> wrote:
> 
>     Why not use the MAC address ? That is supposed to be unique anyway...
> 
>     Am 08.07.2010 12:26, schrieb Philip Taylor:
> 
>         On Thu, Jul 8, 2010 at 10:41 AM, James
>         Robertson<[email protected] <mailto:[email protected]>>  wrote:
>          
> 
>             All you're looking for is
>             a way to reject one connection and keep the other, with a
>             fallback for the
>             (extremely) rare instances that both times are identical.
>              The timestamps
>             could be years apart and the mechanism would still give the
>             required result.
>                
> 
>         If you just want a meaningless number with minimal chances of a
>         collision, timestamps sound like one of the worst possible choices -
>         people put a lot of effort into making clocks accurate and
>         synchronised, but you'll get more collisions as they get more
>         accurate. And once you get one collision, it's much more likely that
>         you'll get another collision when you try again a little later. An
>         n-bit random ID (seeded by more than just the timestamp) would
>         have a
>         much lower chance of collision than an n-bit timestamp.
> 
>          
> 
> 
>     _______________________________________________
>     Sweng-Gamedev mailing list
>     [email protected]
>     <mailto:[email protected]>
>     http://lists.midnightryder.com/listinfo.cgi/sweng-gamedev-midnightryder.com
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Sweng-Gamedev mailing list
> [email protected]
> http://lists.midnightryder.com/listinfo.cgi/sweng-gamedev-midnightryder.com
_______________________________________________
Sweng-Gamedev mailing list
[email protected]
http://lists.midnightryder.com/listinfo.cgi/sweng-gamedev-midnightryder.com