If u look at the records in basics of handeling u will se the person who is #1 with a time of 1:40 but when u look at the replay he drives then falls? and he has the top dog achivment![]()
![]()
![]()
Sadly he wont be there for a long time...
http://www.redlynxtrials.com/forum/viewtopic.php?t=300
Our server detects issues like this automatically, but does not do any decisions alone unless we manually investigate it further. Our server admin is at summer vacation this week, so expect this issue solved next week.
So many new players. There are always a minority who try do things like this.
Sanos muuta. You people should check out sebbbis record of The Final Exam. It's quite interesting.Originally Posted by sebbbi
That's a very old known issue from the v1.05. At that time we improved the dynamic physics model of the dynamic objects. We deleted all replays from the dynamic tracks, but didn't remember that someone had put that dynamic tire at the end of the "Final Exam" level. There are other older replays in that track record list that also behave like that (QcChoppers old record for example was one, but he improved it lately). With old physics model the bike rides correctly to the goal (you can verify this with v1.05). This problem is a completely different one. The length of the replay and the time reported match perfectly. And if you estimate how much time it would take to ride the bike to the goal, you also get the same time as the replay lasts after the fall.Originally Posted by Dosto
Same problem in flip-2. We will investigate it also. Seems that we have to run a full replay scan on our server sometime during next weeks to check how widespread this issue is.
Might seem like a stupid question but how exactly do people cheat like that?Originally Posted by sebbbi
A hacked client can report different score than in the replay. The server doesn't check that the replay matches the score immediately as it would take too much processing power (at peak times there are around 50 new high scores and replays received every minute). We can run periodically a simulation on all the replays on the server database and check if all of the replays give identical result (time, faults, etc) than the score reported. This is completely foolproof. There is no way to cheat the simulation, as only button presses are stored to the replay data, and the simulator just runs the game by replaying exactly these same button presses.Originally Posted by AndreSmol
If we find high scores that do not match the simulation, we will delete them. If we find out that some players are cheating on purpose, we can block those players from the track high score server.
If this issue is more widespread than we expect, we can also automate the simulation process, so that every new track high score is automatically sent to a simulation queue, and the server processes the queue one by one when it has free processing power. This way the false records stay on the boards only for a very short time, before the server automatically removes them.
The two obviously faulty records have been removed. We will investigate the issue further before any additional actions.