Re: Apex Legends Crash no error - PC (apex_crash.txt)

unfortunately after a couple of hours of fun the problem manifests itself again, attached the new crash file.

another one

@OrioStorm  So you think it might just be a server or wifi issue that can be fixed in a later patch? Because today ran for 2 hrs without any disconnecting issues. Other games never give me an issue due to my ethernet cord. Maybe just an Apex issue? the servers are being worked on if im mistaken.

@OrioStorm  Oh wait! Maybe If I switch from Ethernet to Wifi the disconnecting Issues will be Resolved! 🙂

@OrioStorm  UPDATE - I stayed In Main Menus twice with Wireless Wifi and WIred Ethernet, Both Disconnected. Both times I attempted to change data centers to New York (Im In PA) and yet no change. My friends who live near me are playing the game flawlessly with their connections. I am very confused on my next actions, and considering that I might not be able to play this game. ☹️ Plz any advice...

most of the time game crash because I overclock my GPU, but with other game the OC is stable... please fix

New to this forum so sorry if i'm replying "wrong" here.

I've been also getting random crashes since saturday, i've never ever had a crash on this game before and i don't know how to proceed since the logs doesn't really say much, atleast to me.

Attached is the latest crash log.

Would be greatly appreciated if someone can explain why or try to help me out 🙂

Verified the game files and also reinstalled the game, bought a new CPU Cooler as i read from earlier reply's it might have been the cpu overheating, now my cpu is old but it's not overheating..

(Needed a new cooler anyways since the old one was well, really old) 🙂

Thanks!

@OrioStorm  Found out my PC specs, Also Im having trouble finding an actual apex crash.txt  Do you think this might have something to do with my issue not being a crash, but an actual disconnect?

Processor: AMD Fx(tm)-6300 Six Core Processor 3.50Ghz

GPU: Nividia Geforce Gtx1050

Installed Ram: 8gb

and Windows 10

@St4raway, your crash is in some module besides Apex, but Windows didn't tell us the name of the DLL when we asked for it, so I don't know what is crashing.

@B0bbyBiraciall, your issue is a disconnect, not a crash. They're similar from a player's perspective (you can't play the game), but you're right that they are caused by different things and have different error reporting, so you're right that this is why you don't have an apex_crash.txt. In fact, if you don't have that file, it means you've never actually crashed, which is a good thing!

Now, I see you joked about switching from wired ethernet to WiFi to try to fix the disconnects ;-) I didn't know you were on a wired connection, or else I wouldn't have suggested WiFi as the problem. As you know, wired connections are faster and more stable than wireless ones, so you should always use wired ethernet when you can.

That said, I still don't know enough to help you solve your disconnects. It would be helpful to know if you disconnect while playing the game or if you disconnect while loading into a match.

 -- If you disconnect while loading into a match, it could be that it takes you too long to load. If this is when it happens, how long does it usually take you to go from the lobby to the drop ship?

 -- If you disconnect while in a game, that's almost always a problem somewhere between you and your ISP (this is known as "last mile" I believe). I actually don't know how to diagnose and fix these sorts of issues, since I've never had to do it personally or professionally. Maybe try having a web page open and refreshing it when your game disconnects to see if you still have good internet access? I'd recommend using a web page that changes often, like social media or news, so you can be sure it actually refreshed and didn't just show you cached information already on your PC.

@CulturalBaby, your crash is in the 3rd party audio library we use. I'm forwarding the information to them. Thanks for the crash report!

@Ruon-21, this is a crash I haven't seen before. Thanks!

There are two addresses to consider:

0x00007FF7BD29FCE0 -- this is the address of the function we wanted to call.

0x40007FF7BD29FCE0 -- this is the address we actually tried to call.

Under 64-bit Windows, memory addresses starting with 0x4 are invalid, so Windows says you got a memory access violation at address 0xFFFFFFFFFFFFFFFF. This is a magic address that the OS reports whenever there is a memory violation you can detect just by looking at the memory address without even trying to look at the memory that the address points to. An analogy would be the post office recognizing that you put a haiku instead of a recipient address on an envelope, so they don't even bother putting it on a truck to find the house.

You'll notice that the only difference between the correct address and the address we tried to call is that the first digit is a 4 instead of a 0. In binary, that's a single bit difference. I don't know any way that the code that fills in this memory could toggle that bit, so it suggests that some other random code is changing that bit behind our back. This sort of bug is really hard to fix without a repro case and/or lots of crash reports.

@Ruon-21, this is another unique crash I've never seen before. Thanks! I've added it to our bug database.

Suspiciously, in this bug, it looks like RSI is 0x40000290, which is clearly an invalid value and causes the crash. However, 0x00000290 would be a valid value. Just like your last bug, it looks like something changed a 0 to a 4, which is again setting a single bit.

Hi @OrioStorm thanks alot for responding, truly means the world! 🙂

Is there anything i can do to "workaround" this issue?

Just got another crash, looks like the same error codes but attaching none the less 🙂

Is there anyway we will get notified about when these issues are resolved?

@HilariousBabyTTV, I couldn't find any info you've posted about the issue you're having in particular, so I can't give you any advice to work around it. If you have some info, I'd be happy to hear about it.

@OrioStorm  UPDATE- I did some homework on these disconnects and I believe I found a solution. 

Step 1.) Launch Apex (But DO NOT click start on the title screen, just tab out to desktop!)

Step 2.) Open "Services" In windows search bar

Step 3.) Scroll down to Windows Audio Endpoint Builder

Step 4.) Right click this option and select Stop

Step 5.) After processing, Click on Windows Audio Endpoint Builder again and select start

Step 6.) Select the "Windows Audio" file above "Windows Audio Endpoint Builder" and select start (for Windows Audio)

Step 7.) Exit Services

Step 8.) Click start on the apex legends title screen

Step 9.) Play game without any disconnects

Step 10.) Profit 🙂

Just learned this groovy trick today, and there is a full video on youtube.

Sadly this will have to be redone every time I wanna become Champion. I will post an update on my status if any problems occur and will leave this to the developer team for further science. 🙂

-Snake

@B0bbyBiraciall, I just looked up what Windows Audio Endpoint Builder" does, and I don't think it should affect your connection timeouts, unless maybe your PC is frozen / locked up for a long time before the timeout happens.

I'm really happy if that does work, but it would be a mystery WHY.

A related story: back in college we had a program that software emulated a microcontroller. One day, we noticed that the emulator ran faster for a little while after somebody bumped the table, then it would slow down again. Sure enough, bumping the table again made it speed up. Now, programs shouldn't know about the table, and being engineers, we assumed there had to be a logical explanation. After some experimentation, we discovered that bumping the table would jiggle the mouse, and back then the mouse was connected via a serial port (it was a really old PC and mouse), and serial ports could also be used to connect the program to external hardware either running or programming the actual microcontroller (I forget which). So, bumping the table wiggled the mouse which sent a signal on the serial port which made the program stop waiting on the serial port and get back to doing what we wanted it to do, which was emulate.

So, all that to say, even though I don't see how the two are related, there could be a strange series of connections that cause two seemingly unrelated things to be connected in some bizarre way. If so, I'm happy you found a way to avoid your timeouts.

Still getting crashes after the update...

crash:
{
	module@00007FF849D00000: 000000000036878A
	EXCEPTION_ACCESS_VIOLATION(read): 000000000000B880
}
cpu: "AMD Ryzen 5 2600X Six-Core Processor           "
ram: 16 // GB
callstack:
{
	KERNELBASE: 000000000008667C
	ntdll: 00000000000A81CB
	ntdll: 000000000008FD56
	ntdll: 00000000000A477F
	ntdll: 0000000000004BEF
	ntdll: 00000000000A34EE
	module@00007FF849D00000: 000000000036878A
	module@00007FF849D00000: 00000000003697B2
	module@00007FF849D00000: 0000000000CCC5D2
	module@00007FF849D00000: 0000000000389BCF
	module@00007FF849D00000: 0000000000389A76
	module@00007FF849D00000: 00000000003816BD
	module@00007FF849D00000: 0000000000BED65A
	module@00007FF849D00000: 000000000126191C
	KERNEL32: 0000000000017974
	ntdll: 000000000006A271
}
registers:
{
	rax = 0         
	rbx = 0x000001E326609FA0
	rcx = 0         
	rdx = 0x000000DBC70BF260
	rsp = 0x000000DBC70BF1F0
	rbp = 0x000000DBC70BF2F9
	rsi = 0x000001E382D60330
	rdi = 0         
	r8  = 0         
	r9  = 0x000001E40B0941C0
	r10 = 0x000001E375C95640
	r11 = 0         
	r12 = 0         
	r13 = 0         
	r14 = 0x000000DBC70BF360
	r15 = 0         
	rip = 0x00007FF84A06878A
	xmm0  = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm1  = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm2  = [ [-0, 2.1861458, 0, 0], [0x80000000, 0x400BE9D0, 0x00000000, 0x00000000] ]
	xmm3  = [ [136, 0, 0, 0], [0x43080000, 0x00000000, 0x00000000, 0x00000000] ]
	xmm4  = [ [1, 0, 0, 0], [0x3F800000, 0x00000000, 0x00000000, 0x00000000] ]
	xmm5  = [ [18, 0, 0, 0], [0x41900000, 0x00000000, 0x00000000, 0x00000000] ]
	xmm6  = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm7  = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm8  = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm9  = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm10 = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm11 = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm12 = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm13 = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm14 = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
	xmm15 = [ [0, 0, 0, 0], [0, 0, 0, 0] ]
}
build_id: 1557879477

@OrioStorm Auch, wellwell,

Do you have any awnsers regarding when we might get some of these errors/bugs fixed?, i know the majority is different issues, but is there any like "Trello Board" or something we can see the current status?,

I'd love to know when i can the play the game again without having random crashes, really bums me out to start a game, get some good fights, and than just having the crashes..

Oh also, i noted that i got a windows update the same day the crashes started appearing, also a longshot but from my experience, windows updates tend to screw up alot.

Will uninstall that update and do some games and get back to ya!

KB4494441 - might help the sound guys.

EDIT2: Uninstalled the KB, jumped into a game and got a crash just now..

Not the same error as earlier though. Attached the crash

Should i just reinstall windows and hope that solves the problem? ^^

@Sptz87, that's something besides Apex crashing and taking down Apex with it. Unfortunately, Windows didn't tell us the name of the DLL that crashed, so I can't tell you WHAT is crashing.

@OrioStorm  So what you are explaining Is that there must be some logical explanation on why Windows Audio Is somewhat connected to Apex Legends.. Very weird. Well It only takes 1 minute to do. I'll make sure to update any other changes. But for now, as weird as it may sound, changing this command and switching it back on seems to have resolved my issue that Ive been struggling with since launch. That or I'm just lucky and using the Windows Audio command as a placebo effect 🙂 . Y'know one of the two.

@Sptz87 this is a crash in a 3rd party DLL. I can't tell what DLL for you, but for a couple other people it seems to have been their graphics driver. It could also be antivirus, in-game overlays, and a host of other things.

@OrioStorm So far no CPU Parity Internal errors or crashes since the patch.  Great work!

But is Intel actually going to acknowledge that you found a hardware flaw in their processors? 

Been crashing for a while and been following all of the self helps available, still crashing about 20-30% of the time, if you have any suggestions id appreciate it, attached is my most recent crash report

@Falkentyne, thanks for the confirmation!

As for Intel acknowledging the hardware flaw... I personally doubt that will ever happen. They weren't that interested in looking into it for us without an easy repro. And you saw how they dismissed your report with "it's their software problem, our CPUs are great" (to paraphrase). My skeptical self thinks they are a large corporation that would see admitting a rare flaw as only harmful to themselves.

So, we called them out a bit in our patch notes, but I doubt it will go any further than that. Which I think is a shame, really. They had a repro case to find and fix a flaw in their designs, even though it rarely triggered. Looking into it could make all their future CPUs even more solid.

But even if Intel does fix this flaw in firmware, I'm glad we have a proven workaround. I expect that most people don't bother updating their CPU firmware, unless Microsoft forces it on them with an automatic Windows update.

I did leave a way for the game to run the old code that would crash Intel chips, but I didn't investigate whether the compiler generated the same assembly. You can get the old C++ code with setting "old_gather_props" to 1 in the console, or by passing "-force_old_gather_props" on the command line. I'll hit up Intel again and see if they care to fix this; if they don't, I'll remove the old behavior in a future patch.