8 replies to this topic

[DrummDragon]

This is a global bug that appears regardless of map or game type.

My roommate and I play MWO often, and are connected to the internet via the same wifi network, to a high speed cable internet provider. Well, the fastest for Comcast anyways. Both of us are running Windows 7 on machines that can easily handle MWO.

-Whenever we are in the same group and he presses the "quit match" button, it will force me to quit the match as well.

-This bug does not happen when I quit a game, only when my roommate quits a game, or shuts down his client.

-This bug happens approximately 85% of the time.

-This bug occurs in both 4 player groups and twelve player groups.

-This bug occurs no matter who created the group, even if it was a third party.

-This bug has been a consistent issue since the Open Beta.

-This bug is SUPER ANNOYING...and that's a technical fact.

This is not much of an issue when both of us are dead, but when he rage-quits and I'm still alive, it will force me to disconnect from the match, quite often in the middle of a firefight.


Thanks for making a great game, just please fix this bug so I don't get ejected from a match every time my roommate rage quits.

[Ironwithin]

Really ? Your roommate is such a selfish [redacted] that he ruins YOUR gaming experience ON PURPOSE and REGULARLY ? You need new friends.

Edited by Marvyn Dodgers, 20 January 2014 - 04:30 PM.
Filter bypass

[Troutmonkey]

Ironwithin, on 18 January 2014 - 03:48 AM, said:

Really ? Your roommate is such a selfish [redacted] that he ruins YOUR gaming experience ON PURPOSE and REGULARLY ? You need new friends.

Either your sarcasm is lost in text, or you've entirely misunderstood what the OP is trying to say. I've heard of this bug before, but since my friends play from home I've never been able to test it.

Most likely PGI have made some sort of short cut with their netcode and the disconnect message and it gets forwarded to each client on the same router.

Edited by Marvyn Dodgers, 20 January 2014 - 04:30 PM.
Filter bypass in quote

[Tigreen]

Been happening to me and some friends today. 2 games just lagged out for both teams. But, it is to be expected of a game in this state. They should have just left the beta title on the game for a while longer. Not like removing it changed the state of the game.

[Albert Meyburgh]

Hey could you send me the log files for when this happens? (for all the users affected would be nice) albert.meyburgh@piranhagames.com

It doesn't sound logical--but I've been known to fail to recognize logic from a distance.

[BLOOD WOLF]

Troutmonkey, on 18 January 2014 - 04:57 PM, said:

Either your sarcasm is lost in text, or you've entirely misunderstood what the OP is trying to say. I've heard of this bug before, but since my friends play from home I've never been able to test it.

Most likely PGI have made some sort of short cut with their netcode and the disconnect message and it gets forwarded to each client on the same router.

its not PGI, i never had this problem and still dont, their gonna be running around for a bug that doesn't exist. OR maybe it does, or maybe their are many outside factors, or WI fi is crappy

People need to stop buying high end comps and start getting high end internet service

Edited by BLOOD WOLF, 20 January 2014 - 10:02 AM.

[Troutmonkey]

BLOOD WOLF, on 20 January 2014 - 10:03 AM, said:

its not PGI, i never had this problem and still dont, their gonna be running around for a bug that doesn't exist. OR maybe it does, or maybe their are many outside factors, or WI fi is crappy

People need to stop buying high end comps and start getting high end internet service

Just because it doesn't happen to you, doesn't mean it doesn't happen to other people or means that those people were at fault. It infuriates me when I report an intermittent problem and people come along saying "doesn't happen to me, everything works fine! Your computer must suck!" or something along those lines. Computer programs and networks are affected by hundreds of factors, and even the best programmers might fail to account for some of them.

If the problem was caused by crappy Wi-Fi one of them would randomly drop out every now and again. If it was caused by crappy internet they would both randomly drop out at the same time. This isn't random, and has a diagnosed cause that can be repeated. It still could be cause by a setting on the router, but it's still not the end users fault if other online games manage to work fine without this issue.

Edited by Troutmonkey, 20 January 2014 - 05:11 PM.

[Troutmonkey]

Tigreen, on 18 January 2014 - 05:12 PM, said:

Been happening to me and some friends today. 2 games just lagged out for both teams. But, it is to be expected of a game in this state. They should have just left the beta title on the game for a while longer. Not like removing it changed the state of the game.

Not quite the same problem. The issue as the OP describes it is: 1. He and his mate both log into the game on different computers using the same router. They group up and join a game. 2. When his friend leaves the game and returns to MechLab, his client also leaves the game and returns him to the MechLab.

[Albert Meyburgh]

Routers are notorious for causing issues with UDP NAT traversal... Random example reference: "NAT operating characteristics are not standardized." FTA http://en.wikipedia....P_hole_punching

I could hypothesize blindly that perhaps your router has some sort of SPI that times out a UDP connection for one NAT entry, which would cause the other one to close that was also being used by the other client--though that's a bit of a stretch bahaha

I had an expensive router that would force a DCHP renew on the entire LAN every 24 hours at home for awhile... would disconnect me... took a long time to debug that one because since by the time I go to command line and check out ipconfig and such all the values were fine and I could connect to everything... They called it a 'legacy compatibility' feature... To me that's just a horrible hack that breaks design.

Anyhooo my point is really that if there is some common router out there that has some weird implementation that breaks cryengine's assumptions, then if we could get the logs/repro case, that's what we would need in order to investigate the needle in the haystack.

Or maybe it's our fault somehow and I shouldn't be guessing at routers/cryengine