

De-Sync Bug Seems Worse
#41
Posted 10 September 2014 - 07:20 PM
#42
Posted 11 September 2014 - 02:34 AM
#43
Posted 11 September 2014 - 03:31 AM
Is the game trying to save some data (Positional data/Health/ammocounts ect) on his living mech ect, that is too intense for the current throughput?
Seems to get worse the more people disconnect at any given time.
#44
Posted 12 September 2014 - 09:42 AM
#45
Posted 12 September 2014 - 10:28 AM
Neema Teymory, on 12 September 2014 - 09:42 AM, said:
just messaged you about some possiable assistance with this matter. hopefully it can be sorted soon.
#46
Posted 12 September 2014 - 11:04 AM
Also possible that when getting spectated en mass might also be a problem with desync, I've noticed on a few occasions, (may also be the side effect of being mass targeted) that when loosing forces en mass a desync will occur.
If anyone wants to test add me, I'm down for some custom games and testing, pass ts info.
I guess it could be something to do with targeting computers, and some weird bug that when some portions of the enemy team has different ones, a conflict may occur? Again I have no idea on what loadouts of enemy team had at the times I had desync, testing could be done with different possible combinations. I just really really think that something with targeting causes the issue. GLHF Guys! Hunt those devs!
Edited by Omaha, 12 September 2014 - 11:13 AM.
#47
Posted 12 September 2014 - 07:46 PM
I'm chalking it up to planetary rotation. Which might be funny if I wasn't dead every time I sync up.
Making it harder and harder to want to play. When I don't DC, I am doing well. When the running in place starts up, I'm dead before I can get a shot off. Not. Fun.
Tesunie, on 09 September 2014 - 12:52 PM, said:
Won't make a difference.
The way it works is that your initial connection uses DNS to translate a host.domain.name to an IP address. Once the conversation with the remote IP address is started, you don't refer back to DNS again unless you change sites.
PGI, check your load balancers. If they're F5s, hire me =P
#48
Posted 13 September 2014 - 11:10 AM
Mega frustrating.
#49
Posted 14 September 2014 - 06:31 AM
#50
Posted 14 September 2014 - 06:43 AM
#51
Posted 14 September 2014 - 08:34 AM
#52
Posted 14 September 2014 - 08:37 AM
eta0h, on 14 September 2014 - 06:43 AM, said:
Good job! Now they can see that we're not all doing peyote and making this stuff up.
#53
Posted 15 September 2014 - 03:21 AM
Edited by OZHomerOZ, 15 September 2014 - 03:21 AM.
#54
Posted 15 September 2014 - 12:23 PM
#55
Posted 16 September 2014 - 02:10 PM
Edited by MikeSteel, 16 September 2014 - 02:11 PM.
#56
Posted 17 September 2014 - 07:27 AM
#57
Posted 17 September 2014 - 12:54 PM
the modification that the leatrix program does is "increasing the frequency of TCP acknowledgements"
not too sure what the variables for the desync occurs at this point, but the last 3 games i played seems to desync when i get lrmed or in frozen city 6/6 it started to act up during skirmish, or even in forest colony in the cave when i was alone and no one was around just random ass nonsense.
it almost seems like the game server cpu is reaching some sort of bottleneck
Edited by Draven Darkshadow, 17 September 2014 - 12:54 PM.
#58
Posted 17 September 2014 - 02:40 PM
Edited by Thorqemada, 17 September 2014 - 02:40 PM.
#59
Posted 17 September 2014 - 02:52 PM
I want to thank everyone who volunteered to help out. I am no longer looking for volunteers to reproduce this issue. We have a fix that is currently in testing, and we are hoping to have it out to you guys soon. Everyone's help is very much appreciated and was valuable in reproducing and isolating this issue.
I want to point out that this fix specifically addresses a situation in which the server erroneously stops sending data to the client--the generally perceived symptoms are "moonwalking", teleporting, lack of zoom and vision mode functionality, inability to fire LRMS, etc. This bug is most likely to strike when a user gets disconnected from the match, either intentionally or accidentally. When it occurs, it will affect a random user, and it will affect them for a random amount of time. Connection quality, graphics settings, selected mech, map, mode, rejoin, are not factors in causing this issue.
This bug does not address all rubberbanding, snapping, or hit detection issues, all of which are things that can still occur as a result of your individual connection quality.
With that said, this bug is very widespread, so this fix should address a large portion of issues people have been experiencing. Keep your eyes out for an announcement regarding this issue. I look forward to hearing about everyone's experiences once the fix goes out
#60
Posted 18 September 2014 - 07:01 AM
Neema Teymory, on 17 September 2014 - 02:52 PM, said:
I want to thank everyone who volunteered to help out. I am no longer looking for volunteers to reproduce this issue. We have a fix that is currently in testing, and we are hoping to have it out to you guys soon. Everyone's help is very much appreciated and was valuable in reproducing and isolating this issue.
I want to point out that this fix specifically addresses a situation in which the server erroneously stops sending data to the client--the generally perceived symptoms are "moonwalking", teleporting, lack of zoom and vision mode functionality, inability to fire LRMS, etc. This bug is most likely to strike when a user gets disconnected from the match, either intentionally or accidentally. When it occurs, it will affect a random user, and it will affect them for a random amount of time. Connection quality, graphics settings, selected mech, map, mode, rejoin, are not factors in causing this issue.
This bug does not address all rubberbanding, snapping, or hit detection issues, all of which are things that can still occur as a result of your individual connection quality.
With that said, this bug is very widespread, so this fix should address a large portion of issues people have been experiencing. Keep your eyes out for an announcement regarding this issue. I look forward to hearing about everyone's experiences once the fix goes out
Might want to make a new post or make one in announcements, therefore this message isn't lost

Edited by Saxie, 18 September 2014 - 07:01 AM.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users