Loading...
  OR  Zero-K Name:    Password:   
Title: Lobster Club (FREE LOBSTERS!)
Host: USrankShaman
Game version: Zero-K v1.5.1.0
Engine version: 103.0.1-298-gaf1edb1
Battle ID: 443117
Started: 7 years ago
Duration: 43 minutes
Players: 14
Bots: False
Mission: False
Rating: None
Watch Replay Now
Manual download


Show winners



Preview
Filter:    Player:  
sort
#Desync



Consider renaming this map to "Desync dam" -- all the water in the north could be from the tears of the desynced, filling the dam with desync powah.
+0 / -0

7 years ago
something something maps don't cause desync something clear path caches.
+2 / -0

7 years ago
ACtually I asked Brokk whether sea seemed normal for him (he was showing desync messages) and he reply that it was, so i'm wondering whether there really was any real desync....
+0 / -0
develobsters should have implemented path cache clearing / validation before game start right in the lobby...
With invalid path cache player should not even be able to join the gamr
+0 / -0


7 years ago
I don't develop all the lobbies. Use chobby if you want broken path caches to be cleared.
+2 / -2
Sure AUrankAdminGoogleFrog,
but it seems like additional measures required to "help" other lobby developers do smth about it, for example block such lobbies unless they implement this feature(on the proto level)
+0 / -0


7 years ago
They would just spoof the usage of the correct lobby.
+0 / -0
Wouldn't it be possible to send something like hash of the pathCache upon map availability and reporting as ready?
server could trust chobby reported hashes and make a decision from here, i.e. do not allow players with incorrect pathCache hash to join the game(unready player)
i.e. to spoof such hash they would have to have a correct pathCache ;)
+0 / -0
7 years ago
I am pretty sure this is an engine bug, hokomoko can probably fix it if he has time, energy and gets some feedback/logs when desync happens.
Also, it would be interesting to see a few test games on latest spring/develop after kloots recent commit spree, it may give some good feedback to kloot.
+0 / -0
I think you would prefer to wait on engine testing. I'm up for it if you are keen though.



Bottom axis is frame gap buckets, left axis is counts.
+1 / -0