No gun glitch survey

[RazY70]

The no gun glitch affects ranked servers as well. Apparently it's not as prevalent but there are threads about it.

I didn't actually count but by going through this thread it seems like some are more affected by the glitch than others. I mean I've almost never experienced the gun glitch (*knocks on wood*), so maybe there are additional factors playing in. Maybe it's some kind of a client/server(s) sync issue, and maybe there's something in the background which interferes with it somehow. Lots of maybes

[Jokerle]

The no gun glitch affects ranked servers as well. Apparently it's not as prevalent but there are threads about it.

sure it was not on playstation? There are lots of playstation threads about a no gun glitch.

Anyway, we do lots of guessing around.... The scientist in me wants to do some real experiments

[RazY70]

The no gun glitch affects ranked servers as well. Apparently it's not as prevalent but there are threads about it.

sure it was not on playstation? There are lots of playstation threads about a no gun glitch.

Anyway, we do lots of guessing around.... The scientist in me wants to do some real experiments

A lot are PS3 but there are also PC. For example here: http://battlelog.battlefield.com/bf3/fo ... 095571/28/

[V_]

The no gun glitch affects ranked servers as well. Apparently it's not as prevalent but there are threads about it.

sure it was not on playstation? There are lots of playstation threads about a no gun glitch.

Anyway, we do lots of guessing around.... The scientist in me wants to do some real experiments

I've been following this thread for a while now, and there are some really good points being made about potential causes. The best thing we can do now is run a few tests. My question is what would you need for them? Would they require a full server, or just a few people? Approximately how many hours should the test run, to see if the fix is really working? The simplest solution would be to try it during the battleday this Saturday, but since it's LoD's attack I have no say in the matter. Are there other tests we could run?

[cairdazar]

it really doesn't matter they took out the dev weapons out or not. they didn't rework the system.
its blaze <---> server <---> client communication.
it doesn't look like its in the blaze <---> server part since it doesn't effect ranked servers.
on locked weapon servers the weapon data received from blaze transferred to the client. it looks that on an unlocked server the data received from blaze is manipulated (unlock all weapons from the DLCs you own) and then sent to the client. i guess the problem is there.

One more bit of information: when a server is set to "all unlocks" it reports all players to have rank 255, when prompted for live scoreboard information. But in game players looks to have been reset to rank 1.

I wonder what happens if a player manages to level up in an unranked round?
If the client - server and server - ea, is any thing like the live scoreboard every thing is transmitted as hex chars with each value having a fixed length. For players the score is 4 chars; kills, deaths, rank, and squad 2. (max value 65535 and 255 respectively).

Can it be a overflow problem... dam, now I want to try this. Any volunteers? you only need 7000 points to get to rank 2

Lots of maybes

this

[ZebraPeps]

Can it be a overflow problem... dam, now I want to try this. Any volunteers? you only need 7000 points to get to rank 2

Lots of maybes

this

Oh, so that's why RazY never get the gun glitch

[Jokerle]

My question is what would you need for them? Would they require a full server, or just a few people? Approximately how many hours should the test run, to see if the fix is really working?

Entirely unclear...

Did it ever happen to anyone during army practice (= less people) ? I cant remember that is does, can someone?

If we find enough volunteers (40 people?) I would suggest:
Go to Karkand, get close to each other, destroy the buildings like crazy (impact on server memory), change kits often, spawn-in together in large groups (impact on blaze-server/client communication).... aahhhmm ... lay down many mines+C4,
lots of kills and revives...

WAIT a minute, that reminds me of something! We have an increase in the revive time! (Player Man Down Time - 150%) What if there is something bugged and if we revive many people at 101% downtime it can cause communication problems with this blaze thingy? Like: The player loadout info is deleted somewhere if its exceeds 100% and suddenly those info are needed again, thus causing communication problems/high load.

[RazY70]

Can it be a overflow problem... dam, now I want to try this. Any volunteers? you only need 7000 points to get to rank 2

Lots of maybes

this

Oh, so that's why RazY never get the gun glitch

Not leveling up comes pretty naturally to me

Seriously though, I was under the impression all players on an unranked server are level 100

[Necromancer]

WAIT a minute, that reminds me of something! We have an increase in the revive time! (Player Man Down Time - 150%) What if there is something bugged and if we revive many people at 101% downtime it can cause communication problems with this blaze thingy? Like: The player loadout info is deleted somewhere if its exceeds 100% and suddenly those info are needed again, thus causing communication problems/high load.

maybe my ping is so high that key packets are dropped on the way and i get a gun glitch? or maybe a strong magnetic field near the server's communication cable mutating the packs cause the bug? i can make tons of these.
come on, back up your theory with something that makes a bit more sense. we need to check the theories that are more likely to cause it.

Go to Karkand, get close to each other, destroy the buildings like crazy (impact on server memory), change kits often, spawn-in together in large groups (impact on blaze-server/client communication).... aahhhmm ... lay down many mines+C4,

useless.
tons of shooting, med/ammo packs, rpgs, nades, dead bodies, alot of revives, broken crap - sounds like a clusterfuck on a 64 metro server. they don't get the bug there, its not likely the cause.

The no gun glitch affects ranked servers as well. Apparently it's not as prevalent but there are threads about it.

there are several 64 servers that fully populated almost 24/7 and they NEVER get a gun glitch. we get it almost every battleday. is there no difference? reducing the bug appearance to the same levels as on ranked servers is fine by me

The simplest solution would be to try it during the battleday this Saturday, but since it's LoD's attack I have no say in the matter.

its always someones attack/defense. one of the sides can always claim they lost just because of that. isn't it the same now with gun bug? isn't each side thinks 'we had it worse on our side'?
at least locked weapons impact both sides equally, unlike playing all 6 hours on the NA server...

@cairdazar gives bits of potentially valuable information.
points another reason on how 'unlocked weapons' are "messing up" the "normal" (like in ranked) data.
can you catch what exactly the client is receiving upon getting the gun glitch and getting the weapon back?

and on final note - test it on battledays as we need more then 40 players.
last campaign most EUBOs were played with ~20 players per side, and we saw the bug rarely but we get it practically every battle day.

having one full battle day without a glitch is a good start, but i don't think its an indication, might just be 'luck'. playing 2 battle days without it makes a stronger conclusion.

[Jokerle]

WAIT a minute, that reminds me of something! We have an increase in the revive time! (Player Man Down Time - 150%) What if there is something bugged and if we revive many people at 101% downtime it can cause communication problems with this blaze thingy? Like: The player loadout info is deleted somewhere if its exceeds 100% and suddenly those info are needed again, thus causing communication problems/high load.

maybe my ping is so high that key packets are dropped on the way and i get a gun glitch?
come on, back up your theory with something that makes a bit more sense. otherwise its just 'everything can cause it, no way to tell what it is, drop it. its never going to be solved'.

Dont be so negative against all theories! Critical? sure... but negative...?
Knockout arguments like yours kill all discussion.

Player Man Down Time is a setting that differs from most servers.
I just aimed at getting as much load on the server as possible in a concentrated area. Simulating a situation like in a battleday.... Of course we can all sit in spawn and eat cookies for hours if you like that idea better....
Not correctly managing variables inside the server software to take an increase into downtime into account at one point in the code, but having still the default values set at another point in the code does NOT sound so unlikely.

And face it: yes, everything can cause the glitch if there is a bug somewhere in DICE's server software.

I am all for testing weapon unlocks this Battleday, but thinking about alternatives is always good.

[RazY70]

The no gun glitch affects ranked servers as well. Apparently it's not as prevalent but there are threads about it.

there are several 64 servers that fully populated almost 24/7 and they NEVER get a gun glitch. we get it almost every battleday. is there no difference? reducing the bug appearance to the same levels as on ranked servers is fine by me

Well, you don't know it NEVER happens at all. All you know for sure is that it never happened to you. Obviously reducing the glitch's occurrences on the GC servers to the same level would be nice .

There's definitely something which causes that glitch to occur more frequently on the GC servers. What that Something might be is the big question. As far as I can tell GC servers have two unique settings which differentiate them from most others - PLAYER MAN DOWN TIME and PLAYER HEALTH. Plugins might be a factor but if I understood correctly procon was disabled and the problem still occurred for a few players. As a side note and while on the subject of setting, EU server has KICK IDLE PLAYER AFTER SECONDS = 300 instead of 0.

[Necromancer]

@Jokerle
well, then the bug would only happen upon late revive.
how would you explain weapons disappear when alive and conscious?

And face it: yes, everything can cause the glitch if there is a bug somewhere in DICE's server software.

i do, but i don't intend to search everything. i'd try to narrow my search looking at other servers that do/dont suffer from it and then testing things that i think may be the cause.

@RazY70
1) i know more then 'its not happened to me' - i know its probably doesn't happen to anyone on the server because nobody is complaining about it in the chat. seeing how many people are effected by it in GC and not one reporting it in the game chat pretty much means nobody is experiencing it. plus if full 64p where to suffer from it extensively their popularity would drop.
2) PLAYER HEALTH - there are 64p ranked servers with 80% health (hardcore mode) and they don't suffer as much as we do.

[RazY70]

1) i know more then 'its not happened to me' - i know its probably doesn't happen to anyone on the server because nobody is complaining about it in the chat. seeing how many people are effected by it in GC and not one reporting it in the game chat pretty much means nobody is experiencing it. plus if full 64p where to suffer from it extensively their popularity would drop.
2) PLAYER HEALTH - there are 64p ranked servers with 80% health (hardcore mode) and they don't suffer as much as we do.

1) That still doesn't mean it never happens. I for one wouldn't bitch in the chat about getting the gun glitch since no one gives a crap anyway. I would either wait for my gun to reappear or leave the server.
2) In Hardcore mode player health is at 60% which is pretty standard, not 80%

[Necromancer]

if the bug is factor of player health %, then wouldn't it need to hit either 100% or 60% servers harder then us?

[RazY70]

if the bug is factor of player health %, then wouldn't it need to hit either 100% or 60% servers harder then us?

Not necessarily. 100% and 60% are standard settings (Normal; Hardcore) while 80% is customized.

Either way, those are the two settings which I believe are different on the GC servers. I have no idea if either one leads to the glitch in anyway, but that's a good place to start. Maybe after the campaign is done we can arrange a test scrim with them at normal level, starting with PLAYER MAN DOWN TIME. If the bug still occurs frequently that would rule that theory out.