Lag, freezes, crashes and disconnects


#1

I was just doing the orchard quest with the sinister sprites. Didn’t really have any issues finishing it up. Then a lvl 7 runemage comes along on the same quest as i am about to leave, but is having issues with it getting killed. We chat a bit, and i help him out. Meanwhile a warrior comes along as well and start fighting too (was never sure if the warrior was on the quest, or just tried to help).

At that point or perhaps before the warrior arrived, we start to get lag spikes. Short ones at first, but during his 12 kills the lag get worse and worse, some of the freezeups were very long - like up to 20-30 seconds.

Anyway, we part ways as he needs a bathroom break before turning the quest in, while i head back to guild city. On my way back out i meet him again on the bridge, and i ask him which connection type he is on. He said he is on the websocket one, because the default one was giving him trouble. Sadly that is the same issue i experienced, and i use the websocket one too (i wanted to know if people on the default connection type was getting the lag too).

Regardless, with lag spikes like that playing becomes rather unenjoyable. I have had a few of them earlier today when i was questing alone. Most of them rather short (1-5 seconds), and almost all of them immediately on mob death (i am guessing theres a request for mob loot rolling going on at that point or something).

Long story short, since this showed that i am not the only one having lag issues now and then, i was wondering how widespread this issue is, if it is a known issue and if something is being done about it.

Edit with a few more details:
I am on ethernet, 50mbit up/down. Normally never have any latency or bandwidth issues in any games i can think of. I live in Denmark. By the sound of the runemages voice i helped out, i would guess he is american but i forgot to ask (and obviously that isn’t a guarantee that he is actually located in America either).

Edit 2:
Pinging with " -t game-direct.orbusvr.com" i get a stable 96-98ms, but with occasional timeouts. Scratch that, ping has been running for 10 minutes now with no timeouts except a string with 3 of them in a row during the first minute of running.
Broke the ping. Statistics:
1077 packets setn, 1072 received, 5 lost (0%)
Minimum 96ms, Maximum 100ms, Average 97ms

Edit 3:
I am not using steamVR. I bought the game directly on the oculus store to avoid that.

Edit 4:
Uploading my output_log.txt to here: http://s000.tinyupload.com/index.php?file_id=61400547854872543256
Contains data for both connecting using default as well as websocket.

Interesting errors at lines 96-111, 314-344, 350-364 (and many similar ones to the previous 2 segments).
Long log file and ofcourse i don’t know what to look for, but at line 56.948 there is a
“CURRENT CLIENT LATENCY DIRECT: 16128ms”
Also a few “EvenNotFound” exceptions.


#2

If I’m not mistaken this is hardware lag and not networking. I believe it’s when your CPU gets tapped out. I’ve had this issue before as well but have not experienced it recently. I’ve upgraded my CPU since then.


#3

That doesn’t make a lot of sense to me when two people can be next to eachother (presumably) in very different parts of the world, experiencing the same lag spikes.

Add to that, that i can be standing in Highsteppe with loads of players and monsters next to me, talking, walking, in combat, etc and not have any issues - aka the game being silky smooth.

I will however put a performance monitor up so i can take a quick peek at system performance when i get these freezes to make sure that this isn’t the case.


#4

LIFE uh uh Find a way :wink:


#5

Thanks for your incredibly constructive feedback, don’t the let sarcasm hit you.

(Besides it’s “Life finds a way” - Jeff Goldblum, Jurrasic Park)


#6

wasnt referencing that and ya need to chill, heard the necropolar is pretty chill maybe try there


#7

Come to think of it the other day everyone in a dungeon had tracking issies all at once. Perhaps it is something to do with that. Though I would think low framerate would be caused by local hardware. I’ve had low framerate issues in certain dungeons that even followed me through hardware upgrades and windows installs.


#8

Now that is much more constructive, thank you.

However, i don’t have any framerate issues. When it happens the whole thing freezes up. Much like you would see on a heavy VR app when starting it, except that orbus is already running and otherwise perfectly smooth outside of this. As far as i could tell the other player was experiencing the same thing.


#9

I mean based on your own tests and your output log there was some sort of network interference happening. It could be the actual connection itself (although if you didn’t see any large spikes in the ping outside the game client probably not), or as someone else pointed out your CPU was maxed out, causing the network packet processing to be delayed.

It would be helpful to know your system specs as well as what graphics setting you are running the game on.


#10

Hey Riley,
Well to answer that, no i didn’t notice any lag outside of the game client. But on the other hand i wasn’t running anything else besides orbus at the time that i would notice. I have however been running the windows resource monitor constantly since my previous post 3 hours ago. And while CPU usage is high, it isn’t capping out. It is around 75-95% both during normal gameplay and when the freezes occur.

I was just about to type that my system specs are already posted up above, but must have been in another thread then. Anyway, here’s my system specs, and while they are not optimal i am not on the lowest end either:
i5 core 2500k (old quad core @ 3.3ghz)
gtx970, 4gb
16gb ram, nothing fancy
Slow old harddisk, but that doesn’t seem to be an issue other than on startup
Not on wifi but on ethernet (fiber) at 50mbit up/down
Windows 7 home premium (yeah, i don’t plan on windows10 until i really have to)

Here is a fun fact though:

The past 2+ hours i have played with the default connection type. I thought that it was strange that i was unable to play with this setting, so try-try again you know. So, i started orbus up and did that again. And after 3.16something minutes according to the output_log.txt the game crashed. Sadly, there is nothing abnormal to see in the output log, not even a mention of anything going wrong, so i am guessing it fully crashed without any ability to write in the log.

Windows wrote a crash dump somewhere, and i actually considered looking for it and downloading something to look at it with, but being stubborn i tried booting up orbus once more on default connection.

And lo and behold, this time the game did not crash, and has been running smoothly for the past 2+ hours of gameplay during which i have been fighting, questing, and traveling between highsteppe, the desert and the jungle. Not a single hitch. Now isn’t that strange?

So i have a few cases of the game crashing on default connection. I have a few cases (the first night after buying the game a few days ago) of the game freezing up every few minutes on default connection. And now i have 1 example of 2+ hours of smooth gameplay on default connection. The ironic thing is that the game crashed again after having stolen the book from the university island archives. I pulled out the journal outside where i actually had some light and was reading the flavor text. In other words nothing was going on around me at the time.

This is the last entries of the output_log.txt straight after that crash:

GOT NEW MESSASGE STRING
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
GOT NEW LOGOUT TIMER
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
SEND PING!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
PING SIZE: 14
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CURRENT CLIENT LATENCY DIRECT: 61ms
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
TOTAL TIME PLAYED SO FAR: 126.1153 min.
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CURRENT MEMORY USAGE IMPOSTERS: 0.4839935
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
GOT NEW ZONE TIME REMAINING
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
0
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
GOT NEW MESSASGE STRING
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
GOT NEW LOGOUT TIMER
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)

The only odd thing up there is perhaps the line with a single “0” on it?

Anyway i also have several cases of sessions using the websocket, which i used after my first night with the regular freezeups. During those sessions some have been smooth with minor freezes occasionally. And then the episode descriped in the top post where it suddenly froze up constantly and for long periods of time during which a player close to me was experiencing the same thing.

Could it be that there are 2 issues at play here perhaps?

I am thinking that there may be some initialization issue going on when i start up the game on default connection, that sometimes makes the game on my end crash, occasionally freeze up or in the latest case run smoothly.

And i am thinking the other issue may have just been actual bad lag from my location to the server where the other player may have had a similar routing on the end of his chain.

I’m not a network/netcode guy, that’s just the immediate thoughts i am having. My main goal is ofcourse to be able to play the game smoothly, and if we can figure out why it is sometimes not smooth for me and solve the issue, then chances are that i’m not the only one having the issue. Win-win for the early access stage i would say :smiley:

Anyway, if there is any more information i can provide to help narrow it down, please do ask for it. As far as i can see pings and tracerts look pretty “normal”, at least there is nothing i can see out of the ordinary there.


#11

Oh btw, during crash or freezeups the vrclient process sometimes also shows up in red (aka unresponsive) in windows resource monitor. You probably can’t use this information, but in case it is important i thought i’d post it anyway.

Edit: Changed text to reflect that during freezeups the vrclient doesn’t always show up as unresponsive


#12

A little update:

I logged in using default connection, and again i experienced consistent freezes 5 seconds to 1 minute apart. Upon a few game exits and restart, the game got smooth again.

I then advanced the MSQ with the execution in guild city. Upon meeting Fassith in the center of guild city and going through the dialogue, i get into “black space” with a white screen. I guess i am supposed to be seeing a small clip as i have had a few other times during the MSQ so far.

However, instead looking at windows resource monitor CPU usage goes to maximum, the game freezes up for a while and nothing really happens. After a while, an audioclip starts playing with Fassiths dialoque. However, the audio switches to my desktop and the in-VR setting is still frozen up in black space with a white screen.

Not wanting to spoil whatever is supposed to happen (or possibly breaking the MSQ) i quit orbus at this point through Oculus Home (which runs fine as always when switching to it although orbus is frozen up). I have tried several times on both default connection and websocket in case there was any difference. And there was not. I have also tried switching to Oculus Classic Home which consumes less resources in general. Exact same thing happening, although when switching to oculus home that runs just fine and smooth as butter.

I also tried restarting my oculus software and service a few times (no change).

It think it is odd seeing my CPU max out (the graphs for disk, network and memory all look normal during this) when there doesn’t seem to be anything going on and knowing that i previously have seen these “cutscenes” which i suspect i am supposed to see here.

Here is the relevant information from the output_log.txt from my last attempt before writing this:

TRIGGERING PLAYER INTERACTION!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
RE-POSITION DIALOGUE
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
PLAYING MOVIE!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
2
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
PLAYING VIDEO
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
[AVProVideo] Initialising AVPro Video (script v1.7.3 plugin v1.7.3) on NVIDIA GeForce GTX 970/Direct3D 11.0 [level 11.0] (MT True) on WindowsPlayer
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
[AVProVideo] Opening C:/Program Files/Oculus/Software/Software/orbus-online-llc-orbusvr/vrclient_Data/StreamingAssets\Movies/Custscene3.mp4 (offset 0)
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
Coroutine couldn't be started because the the game object 'MediaPlayer' is inactive!
(Filename:  Line: 769)
SETTING FMOD VCA: Master Volume
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
[AVProVideo] Using playback path: DirectShow ([email protected])
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
[AVProVideo] If video fails to play then it may be due to the resolution being higher than 1920x1080 which is the limitation of the Microsoft DirectShow H.264 decoder.
To resolve this you can either use Windows 8 or above (and disable 'Force DirectShow' option), resize your video, use a different codec (such as Hap or DivX), or install a 3rd party H.264 decoder such as LAV Filters.
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
GfxDevice: creating device client; threaded=1
GfxDevice: creating device client; threaded=1
GfxDevice: creating device client; threaded=1
SEND PING!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
PING SIZE: 14
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CURRENT CLIENT LATENCY DIRECT: 75ms
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
TOTAL TIME PLAYED SO FAR: 0.6700122 min.
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CURRENT MEMORY USAGE IMPOSTERS: 0.04982758
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
SEND PING!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
PING SIZE: 14
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CURRENT CLIENT LATENCY DIRECT: 87ms
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
TOTAL TIME PLAYED SO FAR: 1.003686 min.
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
SEND PING!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
PING SIZE: 14
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CURRENT CLIENT LATENCY DIRECT: 64ms
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
TOTAL TIME PLAYED SO FAR: 1.335976 min.
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CURRENT MEMORY USAGE IMPOSTERS: 0.04982758
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
Releasing render texture that is set to be RenderTexture.active!
(Filename:  Line: 697)
SEND PING!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
PING SIZE: 14
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CURRENT CLIENT LATENCY DIRECT: 78ms
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
TOTAL TIME PLAYED SO FAR: 1.669587 min.
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
ON APPLICTION QUIT
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
[AVProVideo] Shutdown
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
Setting up 2 worker threads for Enlighten.
  Thread -> id: 122c -> priority: 1 
  Thread -> id: 1814 -> priority: 1 
ON DISABLE
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CLOSE WS CONNECTION
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
WEBSOCKET CONNECTION CLOSED!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
1005
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)

Looking at the log, it would seem the culprit is to be found here, and that it is unrelated to the previously posted problems in this thread:

Coroutine couldn't be started because the the game object 'MediaPlayer' is inactive!

I will go ahead and try to skip the cutscene and see if any other issues crop up.


#13

Having the occasional crash too, usually only when starting my game session for the day, need to re-boot a couple times due to vrclient un-responsive. After that it runs normally, allso want to add that there seem to be a 50/50% chance to loose connection, total loading freeze or crash when exiting player house, guessing there is lots to load in Highsteppe so gonna get a new house next time.

Cant remember pc-specs, but acording to steams vr test, Im well beyond max score.


#14

Another update:

Went back in, skipped the cutscene and all was fine. Was following the signs, but got those frequent freezeups again (default connection). Exited orbus, started it again. Upon load instead of being just outside guild city i was in highsteppe, level 0. I was about to take a screenshot when i noticed there was nothing around me. At the same time a “connection closed by remote host” appeared, and i was booted back to the login screen.

Edit with output log specifics before they are lost:

SEND PING!
 (Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
PING SIZE: 14
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
45 seconds or more since we sent our ping and no response...timeout?!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
ON APPLICTION QUIT
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
ENET DISCONNECT!
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
CONNECTION ERROR
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
Connection closed by remote host. Please restart the game to reconnect.
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
LOGOUT REQUEST COMPLETE
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
LOGOUT REQUEST COMPLETE
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/DebugBindings.gen.cpp Line: 51)
Unloading 19 Unused Serialized files (Serialized files now loaded: 0)

I would like to stress the part about the connection being closed by the remote host.


#15

And yet another update:

Been playing again on default connection. Smoothly aside from a few loading hitches, but have gotten disconnected twice although my general latency according to the log is between 50-100 ms. Once during the donkey escort quest and rather suddenly at that since i was speaking with a guy also doing it. And once during an elite at the end of the donkey quest (monster with a long tongue, didn’t catch its name).

Both times the with the “Connection closed by remote host” message in the log.

So right now it seems there are 3 connection issues in this thread so far:

  • the original posted one with massive lag/freezes but no disconnects
  • the frequent freezeups after login using default connection, which is mysteriously fixed by multiple login attempts
  • the “Connection closed by remote host” issue i’ve encountered today

#16

Hey Tommy,

Thanks for providing such detailed information. I’ll certainly review it and there might be a few things we can tweak to help, but at the end of the day I think the primary problem is that your CPU is below our minimum requirements for the game. Other folks with older, slower CPUs have reported similar issues in the past, and upgrading their CPU has fixed them.

Most likely, these two issues would be solved by that:

As for the third, that might just be something to do with the server hiccuping on our end, I’ll investigate that further.


#17

Thanks Riley,

But i don’t understand how upgrading my CPU - which has a static speed (well, an upper cap as all CPUs do) would have an effect on sometimes logging in with freezes, and other times with the exact same setup like f ex right now, writing this post with the game in the background i have an active session going on 138 minutes without any freezes.

My point is that i can probably recreate the problem logging out and back in a few times right now and get the frequent freezes again without any soft- or hardware changing on my end.

(edit: there is no doubt that upgrading my cpu will benefit me in general, i am not disputing this at all, i just doubt that it has anything to do with the above)


#18

Yeah I’m not totally sure either, it has something to do with the way our game engine (Unity) works and manages multiple threads when the CPU is under heavy load I think. All I know is that as I said we have around 5-6 previous cases of similar issues that were solved via upgrading the CPU so that it met the minimum requirements.

For example, see here: No more websockets for me!

It’s worth noting that one of our primary things we’re working on in the next 4-6 months is making the game run faster and better in general, via our art asset upgrades, engine upgrades, and performance optimizations to our shaders, etc. So likely this will get better long-term. It’s just not something I can provide a simple and easy fix for right now.


#19

I hear you and i know that optimization isn’t something that is just done overnight. It takes time to do mesh combines, chunk optimizations, texture atlasses, the right culling(s), mesh optimizations, light optimizations, thread balancing and so on and so forth :smiley:

Anyway, i hope you find a thing or two on your end regarding the connectivity. I think the issue may be a little more common than posts here leads to believe :slight_smile:

For now i’ll consider the matter as closed until either i stumble on new information for you or you do :slight_smile:


#20

Just duo’ed the first dungeon with a helpful guy that turned up to my dungeon request (mad props to you ICEHatchet!)

Anyway, a few issues during an otherwise smooth (apart from the duo difficulties) run which may be related to the network/connection issues:

  • ICEHatchet started out as warrior (lvl 9). At the start of the run his shield disappeared, but i could see it. I told him that happens to my poison arrow once in a while too. After some fiddling it reappeared for him.
  • Mid run i stopped being able to hear him. At first i thought he turned his mic off intentionally. Later on, using gestures we both relogged after which we could hear eachother again. It turns out that before we did this (and during the first barely failed attempt on the second and last boss) he could hear me all the time, but i couldn’t hear him. In turn, he couldn’t see me, but he could se my dragon scooting around.

These might ofcourse be completely unrelated issues, but who knows. It could also be hiccups in the connectivity so i thought i would post it anyway.