Update 29 crashes Mac

Voiddreams

Since update 29-the cp reset one- my Mac Pro crashes to desktop whenever I use a wayshrine or try and watch the new video. Any updates?

AngryLorekeeper

I've been having similar issues, I crashed hard last night trying to use a wayshrine to go to Coldharbour. I'm firmly convinced that ZOS doesn't care about Mac users having problems because we are such a small user community. Otherwise they would have at least addressed one of the threads here about problems after the new patch.

ZOS_Adrikoth

Hey there folks,

We're sorry to hear that you are experiencing these crashing issues. If you could start by trying the steps in this article, they may be able to help.

If this issue persists or you need additional assistance, please submit a ticket at help.elderscrollsonline.com and we would be happy to assist you further.

el25pelas

Hello all. A Mac Mini M1 user here. I'm experimenting those crashes too. There's no way to enter any dungeon or delve. I cannot even leave Stonefalls zone or travel home. Game is unplayable at this point.

These problems did not happen before the last update and, although with the already known limitations, it was possible to play smoothly at a stable 30 fps.

I've opened three tickets: one using the ticketing system on the web and two using the /bug command in-game.

The answers offered by technical support are really surprising, even providing adequate technical information, such as the system crash report, where it can be verified that MoltenVK causes a memory allocation error.

Voiddreams

So, a status update, Mar 12, 2001

First off, all their safe mode, windowed screen etc. did not work for me.
If the game crashes, just re-entering the game or fiddling with the settings does not work.

Macbook Pro late 2015
What I saw, was Apple needed to update to Big Sur 11.2.3.
Then I did a launcher update and game repair.
I went to minion and verified all my addons were the latest. I am running about 25 addons.
Then I went into the game, killed all my outdated addons, loaded the character screen and launched my main. It was able to start and show my main.
Then I checked if I could change waypoints without companions. It worked.
Then I loaded my companions and tried a waypoint. It worked.
Then I enabled all my addons, including out of date addons from minion. I refreshed my UI.
Then I tried a waypoint. It worked.
Anytime the game crashes, I do a full reboot of the machine.

Right now the game is working with the same settings and addons as before the update.
I have not tried the game cutscene yet. It died every time and I really want just the game to run.

el25pelas

A final response on one of my open tickets finally arrived:

We will not be able to continue troubleshooting the unsupported hardware. If you need to pursue a refund for the game you can contact Steam's Customer Support.

Unsupported. But game can be purchased on Steam from an Apple Silicon machine.

Seeing the trend at Apple regarding the implementation of Apple Silicon, there will be support for this game on macOS for a year or maybe two.

This is a nonsense.

So far so good ESO. Will wait for the Amazon LOTR based MMO and will request a refund from Steam.

Good luck everybody.

lynzriches

Voiddreams wrote: »

Since update 29-the cp reset one- my Mac Pro crashes to desktop whenever I use a wayshrine or try and watch the new video. Any updates?

Having similar issues. Tried the steps recommeded by ZOS but nothing is helping. Able to play for a bit, then if I try to wayshrine or enter a delve, crashes. Doesn't seem to be *every* time, but often enough to be frustrating.

SeaGtGruff

My crashes have all been while trying to return to the character selection screen so I can choose a different character to do crafting writs on, or trying to return to the server selection screen to switch servers.

Sometimes it works, and sometimes it doesn't.

I don't use any add-ons, just "plain vanilla" ESO. MacBook Air, 2017 model I think.

IIRC, I was able to successfully travel to a guild hall to do a master writ, then take wayshrines from Artaeum to Stormhaven to turn in the master writ, and from Stormhaven to Vvardenfell to return to my current base of operations.

el25pelas

For anyone running ESO an an Apple Silicon, and experimenting crashes when travelling to any delve, dungeon or home: try to banish any minion or pets before travelling or entering a delve or dungeon. It just simply works on my particular vase (no addons, game purchased on Steam).

Reviewing the crash log it seems that MoltenVK is trying to access deallocated memory:

Crashed Thread:        12  Dispatch queue: MoltenVKQueue-0-0-1.0-Dispatch

Exception Type:        EXC_BAD_ACCESS (SIGSEGV)
Exception Codes:       KERN_INVALID_ADDRESS at 0x00400000000001d0 -> 0x00000000000001d0 (possible pointer authentication failure)
Exception Note:        EXC_CORPSE_NOTIFY

Termination Signal:    Segmentation fault: 11
Termination Reason:    Namespace SIGNAL, Code 0xb
Terminating Process:   exc handler [906]

VM Regions Near 0x1d0:
--> 
    __TEXT                      104950000-1096c2000    [ 77.4M] r-x/rwx SM=COW  /Users/*/Library/Application Support/Steam/*/eso.app/Contents/MacOS/eso

Thread 0:: MainThrd  Dispatch queue: com.apple.main-thread
0   libsystem_malloc.dylib        	0x00007fff201acfa1 tiny_free_list_add_ptr + 743
1   libsystem_malloc.dylib        	0x00007fff201ac7aa tiny_free_no_lock + 1116
2   libsystem_malloc.dylib        	0x00007fff201ac1f9 free_tiny + 442
3   com.zenimaxonline.eso         	0x000000010763204d operator delete(void*) + 9
4   com.zenimaxonline.eso         	0x0000000106a596b6 ZoMap<VkImage_T*, VmaAllocation_T*>::RemoveNode(ZoMapNode<VkImage_T*, VmaAllocation_T*>*, bool) + 718
5   com.zenimaxonline.eso         	0x0000000106a55bec ZoTexturePoolVk::FreeMemory(VkImage_T*) + 136
6   com.zenimaxonline.eso         	0x0000000106a60182 ZoTextureResourceVk::~ZoTextureResourceVk() + 66
7   com.zenimaxonline.eso         	0x0000000106a602e9 ZoTextureVk::~ZoTextureVk() + 37
8   com.zenimaxonline.eso         	0x0000000106a60354 ZoTextureVk::~ZoTextureVk() + 14
9   com.zenimaxonline.eso         	0x00000001069fd1d0 ZoRenderDevice::ServicePendingFreeRequests(unsigned int) + 142
10  com.zenimaxonline.eso         	0x0000000106a4f51d ZoRenderDeviceVk::ServicePendingFreeRequests(unsigned int) + 101
11  com.zenimaxonline.eso         	0x0000000106a0cea4 ZoRenderManager::BeginNewFrame() + 28
12  com.zenimaxonline.eso         	0x0000000106a0ccb3 ZoRenderManager::Present() + 147
13  com.zenimaxonline.eso         	0x000000010499bed7 -[GameClient macPresent] + 125
14  com.zenimaxonline.eso         	0x000000010499c0f8 -[GameClient macRun] + 303
15  com.zenimaxonline.eso         	0x000000010499ca1e -[GameClient platformMainLoop] + 2164
16  com.zenimaxonline.eso         	0x000000010499bd77 -[GameClient mainLoop] + 61
17  com.zenimaxonline.eso         	0x000000010499bc80 -[GameClient applicationDidFinishLaunching:] + 668
18  com.apple.CoreFoundation      	0x00007fff2046f895 __CFNOTIFICATIONCENTER_IS_CALLING_OUT_TO_AN_OBSERVER__ + 12
19  com.apple.CoreFoundation      	0x00007fff2050b224 ___CFXRegistrationPost_block_invoke + 49
20  com.apple.CoreFoundation      	0x00007fff2050b1a7 _CFXRegistrationPost + 454
21  com.apple.CoreFoundation      	0x00007fff20440546 _CFXNotificationPost + 787
22  com.apple.Foundation          	0x00007fff210e4abe -[NSNotificationCenter postNotificationName:object:userInfo:] + 59
23  com.apple.AppKit              	0x00007fff22bd6b2d -[NSApplication _postDidFinishNotification] + 305
24  com.apple.AppKit              	0x00007fff22bd687b -[NSApplication _sendFinishLaunchingNotification] + 208
25  com.apple.AppKit              	0x00007fff22bd3a72 -[NSApplication(NSAppleEventHandling) _handleAEOpenEvent:] + 541
26  com.apple.AppKit              	0x00007fff22bd36c7 -[NSApplication(NSAppleEventHandling) _handleCoreEvent:withReplyEvent:] + 665
27  com.apple.Foundation          	0x00007fff21110056 -[NSAppleEventManager dispatchRawAppleEvent:withRawReply:handlerRefCon:] + 308
28  com.apple.Foundation          	0x00007fff2110fec6 _NSAppleEventManagerGenericHandler + 80
29  com.apple.AE                  	0x00007fff261bced9 0x7fff261b0000 + 52953
30  com.apple.AE                  	0x00007fff261bc5f4 0x7fff261b0000 + 50676
31  com.apple.AE                  	0x00007fff261b5260 aeProcessAppleEvent + 452
32  com.apple.HIToolbox           	0x00007fff289585a2 AEProcessAppleEvent + 54
33  com.apple.AppKit              	0x00007fff22bcde3e _DPSNextEvent + 2048
34  com.apple.AppKit              	0x00007fff22bcc177 -[NSApplication(NSEvent) _nextEventMatchingEventMask:untilDate:inMode:dequeue:] + 1366
35  com.apple.AppKit              	0x00007fff22bbe68a -[NSApplication run] + 586
36  com.zenimaxonline.eso         	0x000000010499d86c main + 200
37  libdyld.dylib                 	0x00007fff2039d621 start + 1

This is a classic one.

They probably already have, but to find the cause of the problem the development team should take note: one of the best options to find and fix the problem is to use the NSZombieEnabled to true.

When you set NSZombieEnabled then whenever an object reaches retain count 0, rather than begin deallocated it morphs itself into an NSZombie instance. Whenever such a zombie receives a message, it logs a warning rather than crashing or behaving in an unpredictable way. As such, you can debug subtle over-release/autorelease problems without advanced tools or painstaking needle in haystack searches.

To sum up: please use Instruments to profile the game and find memory leaks or bad allocations. The game experience will be much better.

el25pelas

Update: Don't exceed the limits of an area on horseback. The game crashes and you will not be able to use the character again.
I have opened a new ticket to try to get someone from support to move my character to another point.

el25pelas

Update II: After several hours playing the game without a crash on a MacMini M1 I can confirm than MoltenVK seems to try to deallocate continuously already deallocated objects.

Those objects seems to be all that is not the character: minions, pets, horses or any other item.
That makes sense: if game is developed around character (and I bet it is), when you try to load another map you must deallocate all elements and allocate them again when new map is loaded and ready.
For some reason, MoltenVK (or Vulkan) is trying to deallocate an object twice. And this is the reason of the crash. As I pointed on my lasts posts, this could be detected using Instruments for profiling the game on a Mac. This development tool is extremely useful.

Maybe updating MoltenVK to last version should fix it. But I can’t see source code for ESO so it’s just a guess.

So to play again without major problems you MUST remove all minions, pets, horses or whatever BEFORE entering a new zone, delve, dungeon or something similar.
Skyshards, Lorebooks and Spanish language addons works flawlessly too.

Hope this can help anyone.

aldriq

el25pelas wrote: »

Maybe updating MoltenVK to last version should fix it. But I can’t see source code for ESO so it’s just a guess.

So to play again without major problems you MUST remove all minions, pets, horses or whatever BEFORE entering a new zone, delve, dungeon or something similar.
Skyshards, Lorebooks and Spanish language addons works flawlessly too.

Hope this can help anyone.

Ah, that would make sense, I've been getting the crashes when porting to friends/guildies with 'Travel to Player' option, pretty sure I had a pet active. I'll try and see (I'm on an Intel i5 but with integrated GPU).

el25pelas

aldriq wrote: »

Ah, that would make sense, I've been getting the crashes when porting to friends/guildies with 'Travel to Player' option, pretty sure I had a pet active. I'll try and see (I'm on an Intel i5 but with integrated GPU).

Hope it can help.

I've been to dungeons this afternoon. I've done four, and I've experienced two/three crashes. Specifically while trying Fungal Grotto.
I have experimented with summoning minions as soon as my character enter the world and banishing them right after, to immediately travel to the dungeon (it really is a porting to the friend / guildie).

It seems that if you have just entered the game, it works best not to summon anything, to do it right after entering the dungeon.

I'll keep experimenting, but at the moment it seems that the technique of banishing any type of external element to the character just before any kind of travel works quite well, which isolates the problem enough to be fixed with a patch.

Hopefully someone with decision-making power reads this.

Mzav

I encounters this same issue even after trying all of the ZOS recommended fixes + the ones in this thread. Prior to U29 I never had any issues, but now it happens pretty regularly. Same readout about MoltenVK.

I tried to submit a support ticket but on two separate attempts I got a screen that said my ticket had been submitted and there was no ticket number. I submitted a ticket to get access to the forum and it worked as expected.

el25pelas

Mzav wrote: »

I encounters this same issue even after trying all of the ZOS recommended fixes + the ones in this thread.

Can you please share your hardware and a crash report? Maybe we can find any clue

Are the crashes you found related to travel? How often they occur?

Too many questions, sorry. Just trying to understand.

Mzav

No problem! Also want to get to the bottom of this. It is mostly related to travel, especially to/from dungeons and battlegrounds. Wayshrines appear to work better than map fast travel. It did crash yesterday in Skyreach while fighting a boss. I would say it crashes 2-3 times an hour if I am trying to do dungeons/bg and maybe once an hour if I am just doing overland questing.

Operating System: Mojave 10.14.6
Processor: 3.1 GHz Intel Core i5
Memory: 8 GB 2133 MHz LPDDR3
Graphics: Intel Iris Plus Graphics 650 1536 MB


Here is the info from the crash. Couldn't post the whole log b/c of character limits, but this is the only crashed thread.

Process: eso [7564]
Path: /Applications/Zenimax Online/*/eso.app/Contents/MacOS/eso
Identifier: com.zenimaxonline.eso
Version: 1.0.0 (1.0.0)
Code Type: X86-64 (Native)
Parent Process: ??? [1]
Responsible: eso [7564]
User ID: 501

Date/Time: 2021-03-14 15:21:15.719 -0400
OS Version: Mac OS X 10.14.6 (18G87)
Report Version: 12
Bridge OS Version: 3.0 (14Y904)
Anonymous UUID: D6BB9193-D760-5CC8-161A-7ACD4D949A3F

Sleep/Wake UUID: F71E3176-6D7A-4949-83ED-C522796702D9

Time Awake Since Boot: 150000 seconds
Time Since Wake: 98000 seconds

System Integrity Protection: enabled

Crashed Thread: 18 Dispatch queue: MoltenVKQueue-0-0-1.0-Dispatch

Exception Type: EXC_BAD_ACCESS (SIGSEGV)
Exception Codes: KERN_INVALID_ADDRESS at 0x0000000000000018
Exception Note: EXC_CORPSE_NOTIFY

Termination Signal: Segmentation fault: 11
Termination Reason: Namespace SIGNAL, Code 0xb
Terminating Process: exc handler [7564]

.....


Thread 18 Crashed:: Dispatch queue: MoltenVKQueue-0-0-1.0-Dispatch
0 libobjc.A.dylib 0x00007fff5b6d469d objc_msgSend + 29
1 com.apple.driver.AppleIntelKBLGraphicsMTLDriver 0x00007fff2a539184 SamplerStage::bindSamplerToTexture(unsigned char, unsigned char, bool) + 144
2 com.apple.driver.AppleIntelKBLGraphicsMTLDriver 0x00007fff2a4ed018 SamplerStage::writeSamplerState() + 564
3 com.apple.driver.AppleIntelKBLGraphicsMTLDriver 0x00007fff2a4ecdae IGRenderStateSamplers::writeIf() + 86
4 com.apple.driver.AppleIntelKBLGraphicsMTLDriver 0x00007fff2a4eb10b IGAccelRenderCommandEncoder::programPipeline(sPrimitiveData const&) + 4129
5 com.apple.driver.AppleIntelKBLGraphicsMTLDriver 0x00007fff2a4e9eca IGAccelRenderCommandEncoder::drawIndexedPrimitives(MTLPrimitiveType, unsigned int, MTLIndexType, MTLIGAccelBuffer*, unsigned int, unsigned int, unsigned int, unsigned int) + 486
6 com.apple.driver.AppleIntelKBLGraphicsMTLDriver 0x00007fff2a503a13 -[MTLIGAccelRenderCommandEncoder drawIndexedPrimitives:indexCount:indexType:indexBuffer:indexBufferOffset:instanceCount:baseVertex:baseInstance:] + 347
7 com.zenimaxonline.eso 0x000000010cbd8eb6 MVKCmdDrawIndexed::encode(MVKCommandEncoder*) + 118
8 com.zenimaxonline.eso 0x000000010cbd9c4b MVKCommandEncoder::encode(MVKQueueCommandBufferSubmission*) + 107
9 com.zenimaxonline.eso 0x000000010cbd9aec MVKCommandBuffer::execute(MVKQueueCommandBufferSubmission*, MVKCommandBufferBatchPosition const&) + 124
10 com.zenimaxonline.eso 0x000000010cbeec7e MVKQueueCommandBufferSubmission::execute() + 110
11 com.zenimaxonline.eso 0x000000010cbedacc invocation function for block in MVKQueue::submit(MVKQueueSubmission*) + 28
12 libdispatch.dylib 0x00007fff5ce625f8 _dispatch_call_block_and_release + 12
13 libdispatch.dylib 0x00007fff5ce6363d _dispatch_client_callout + 8
14 libdispatch.dylib 0x00007fff5ce698e0 _dispatch_lane_serial_drain + 602
15 libdispatch.dylib 0x00007fff5ce6a396 _dispatch_lane_invoke + 385
16 libdispatch.dylib 0x00007fff5ce726ed _dispatch_workloop_worker_thread + 598
17 libsystem_pthread.dylib 0x00007fff5d0a3611 _pthread_wqthread + 421
18 libsystem_pthread.dylib 0x00007fff5d0a33fd start_wqthread + 13

Thread 19:
0 libsystem_pthread.dylib 0x00007fff5d0a33f0 start_wqthread + 0

Thread 20:
0 libsystem_pthread.dylib 0x00007fff5d0a33f0 start_wqthread + 0

Thread 18 crashed with X86 Thread State (64-bit):
rax: 0x00007fc73c7dc600 rbx: 0x0000000169e82090 rcx: 0x0000000000000003 rdx: 0x0000000000000003
rdi: 0x00007fc73c7dc600 rsi: 0x00007fff4cd0bbda rbp: 0x00007000089341b0 rsp: 0x0000700008934188
r8: 0x000000016324d018 r9: 0x0000000000011660 r10: 0x0000000000000000 r11: 0x00007fff4cd0bbda
r12: 0x0000000000000003 r13: 0x0000000000000003 r14: 0x0000000169e820e0 r15: 0x00007fc73c7dc600
rip: 0x00007fff5b6d469d rfl: 0x0000000000010246 cr2: 0x0000000000000018

Logical CPU: 3
Error Code: 0x00000004
Trap Number: 14

Booldi

The remove pets method works perfectly for me, but only in solo play. If I join a group I still crash when changing zones. Anyone found a solution for that?
M1 MacBook Air

Booldi

Crash report when the crash is caused by traveling while in a group:
Process: eso [1616]
Path: /Users/USER/Library/Application Support/Steam/*/eso.app/Contents/MacOS/eso
Identifier: com.zenimaxonline.eso
Version: 1.0.0 (1.0.0)
Code Type: X86-64 (Translated)
Parent Process: ZeniMax Online Studios Launcher [1593]
Responsible: zosSteamStarterMac2 [1592]
User ID: 501

Date/Time: 2021-03-15 19:11:30.448 +0100
OS Version: macOS 11.2.3 (20D91)
Report Version: 12
Anonymous UUID: A14C6FC1-55CE-8686-481F-5B90886E0414

Sleep/Wake UUID: B74004C1-7BD6-4794-B8E5-0B45D5E704FF

Time Awake Since Boot: 11000 seconds
Time Since Wake: 3800 seconds

System Integrity Protection: enabled

Crashed Thread: 32 Dispatch queue: MoltenVKQueue-0-0-1.0-Dispatch

Exception Type: EXC_BAD_ACCESS (SIGSEGV)
Exception Codes: KERN_INVALID_ADDRESS at 0x0000000000000070
Exception Note: EXC_CORPSE_NOTIFY

Termination Signal: Segmentation fault: 11
Termination Reason: Namespace SIGNAL, Code 0xb
Terminating Process: exc handler [1616]

VM Regions Near 0x70:
-->
__TEXT 10001a000-104d8c000 [ 77.4M] r-x/rwx SM=COW /Users/*/Library/Application Support/Steam/*/eso.app/Contents/MacOS/eso
Thread 32 Crashed:: Dispatch queue: MoltenVKQueue-0-0-1.0-Dispatch
0 com.apple.AGXMetal13-3 0x0000000126eb9036 0x126849000 + 6750262
1 com.zenimaxonline.eso 0x00000001018fbced MVKGraphicsResourcesCommandEncoderState::encodeImpl() + 653
2 com.zenimaxonline.eso 0x00000001018dad7f MVKCommandEncoder::finalizeDrawState() + 63
3 com.zenimaxonline.eso 0x00000001018d8e55 MVKCmdDrawIndexed::encode(MVKCommandEncoder*) + 21
4 com.zenimaxonline.eso 0x00000001018d9c4b MVKCommandEncoder::encode(MVKQueueCommandBufferSubmission*) + 107
5 com.zenimaxonline.eso 0x00000001018d9aec MVKCommandBuffer::execute(MVKQueueCommandBufferSubmission*, MVKCommandBufferBatchPosition const&) + 124
6 com.zenimaxonline.eso 0x00000001018eec7e MVKQueueCommandBufferSubmission::execute() + 110
7 com.zenimaxonline.eso 0x00000001018edacc invocation function for block in MVKQueue::submit(MVKQueueSubmission*) + 28
8 libdispatch.dylib 0x00007fff201ad5dd _dispatch_call_block_and_release + 12
9 libdispatch.dylib 0x00007fff201ae7c7 _dispatch_client_callout + 8
10 libdispatch.dylib 0x00007fff201b45fe _dispatch_lane_serial_drain + 606
11 libdispatch.dylib 0x00007fff201b50cb _dispatch_lane_invoke + 375
12 libdispatch.dylib 0x00007fff201bec5d _dispatch_workloop_worker_thread + 819
13 libsystem_pthread.dylib 0x00007fff20357499 _pthread_wqthread + 314
14 libsystem_pthread.dylib 0x00007fff20356467 start_wqthread + 15

el25pelas

Booldi wrote: »

The remove pets method works perfectly for me, but only in solo play. If I join a group I still crash when changing zones. Anyone found a solution for that?
M1 MacBook Air

Hi Booldi, I’ve found that if you’re in a group and use the “banish tip” (let’s name it) crashes will occur randomly but not too often. They actually occur very rarely. Game is playable.

A workaround could be to invoke some minions/pets and immediately banish them just before traveling. This worked for me when doing dungeons. Also, don’t try to travel or trespass zone limits by horse, tiger or whatever the mount you use.

Will review this logs tomorrow (if job lets me) in order to find similarities.

el25pelas

Mzav wrote: »

No problem! Also want to get to the bottom of this. It is mostly related to travel, especially to/from dungeons and battlegrounds. Wayshrines appear to work better than map fast travel. It did crash yesterday in Skyreach while fighting a boss. I would say it crashes 2-3 times an hour if I am trying to do dungeons/bg and maybe once an hour if I am just doing overland questing.

It is notable that crashes occur spontaneously. Here I would differentiate between two different types of crash: the one caused by traveling and the spontaneous one.
I have not experienced the last one from a Mac Mini M1 (yet).

Will se crash log to try to put some light on this case. Than you for the information.

Mzav

The spontaneous crash only happened once, but the other crash is regular. The spontaneous crash happened while fighting the mantikora boss in skyreach hold. the fight already seemed buggy before the crash and I had died a few times.

random crashes do happen and it's possible this is a coincidence.

doesurmindglow

I'm having this problem also -- frequent crashes on port since the patch. Running a mac client as well.

I will try the banishing approach tomorrow and see if that resolves it.

Booldi

New clue (maybe):
I was fideling around with settings, and ran dungeons to test. One time the loading screen was short, usually thats when it crashes, but i stayed in game this time with the message “instance is full”. That could be an explanation to why traveling takes away your gold but doesnt take you anywhere when crashing. I suspect this happenes every time you crash, but you can’t see it.

Mzav

It takes your gold because of the order of operations. You pay upon departure, not arrival. Way oversimplified, but imagine Script A takes gold and triggers script B, which loads you into a different cell. Script A goes off, and something in script B crashes. There is no "failsafe" which would go back and re-run script A to refund the gold, because A has already concluded and the game is crashed so isn't running any more scripts.

The "instance full" bug is something that happens occasionally, but does not cause crash. It will usually resolve itself in a minute or two.

el25pelas

Sorry for the delay guys. Having some intense work days.

Well I've been experimenting a little bit more and I found that the "banish tip" works as expected. At least on 90% of cases. There's a 10% of cases located on specific places. For example, it is virtually impossible for me to enter on Lady Llarel's Shelter (Deeshan) or Fungal Grotto (Stonefalls), directly using the door.
On this places crashes always occur, no matter the tip or trick I'm trying to apply.

Fungal Grotto is a case to study because I can enter and play the dungeon without major problems if any friend is already inside. Of course, never after two or three crashes.

Speaking of the crash reports that colleagues have sent, I cannot add much more than what has already been clarified. But I will try to give a more specific point of view:

@Mzav reported a crash that seems to be the same as reported on this thread: https://forums.elderscrollsonline.com/en/discussion/486515/elsweyr-keep-crashing
Specifically, it seems that the crash occurs when calling the bindSamplerToTexture method on the Metal layer of the system. I sense there may be a zombie problem here again.
Sending messages to objects no longer allocated in memory produces this type of crashes.
Here, and seeing the complete trace of the crash, it could fit the possibility of a bug in Apple's Metal library. But I can't assert anything for sure without debugging.

@Booldi reports a different type of crash. It's dying at MVKGraphicsResourcesCommandEncoderState::encodeImpl()
Probably related to buffers and their creation and destruction. I have no further information in this regard, although this MoltenVK bug from 2018 seems to be quite similar.
https://github.com/KhronosGroup/MoltenVK/issues/189

At this point I suppose someone in support should speak up. We are offering more than enough information to debug the game and correct these problems that do not allow 100% play on Mac.
Not only in devices with M1 chip but in others with Intel processors although integrated GPUs. As far as I know, the latter hardware is supported and is also failing at the moment.

I regret not being able to provide more information.
Can someone from ZOS speak please?

Booldi

I’ve found the sad solution.....play on stadia

mav983

Since update 29 I have found that I crash on load screens with my 2015 iMac with great frequency. It seems to be much worse when logging into the game for the first time in the day rather than after a crash. I am not having as many in-game crashes as I was with update 28, but that may be because I am having a much harder time getting in to begin with.

aldriq

Only 2 crashes to report so far since I banished all pets on Sunday - one leaving the Imperial City Sewers and one logging out of a char (both seem on the same 'deallocate stuff on exit' category at least). But crucially, I haven't got any crashes porting in and out of houses or the guildhall, so the workaround does seem to work well for those cases.

Let's hope once the dust settles on the U29 console release we can get some word from dev or support to acknowledge this issue.

el25pelas

Just downloading new patch for macOS client. Hope it fix all that crashes

https://forums.elderscrollsonline.com/en/discussion/566664/pc-mac-patch-notes-v6-3-6#latest

Ohtimbar

el25pelas wrote: »

Just downloading new path for macOS client. Hope it fix all that crashes

https://forums.elderscrollsonline.com/en/discussion/566664/pc-mac-patch-notes-v6-3-6#latest

It didn't. At least, not for me.

el25pelas

To be fair, it is true that in some (few) places that I could not enter before, now it is possible to do so.

However, I have observed reproducible behavior:

When trying to complete a mission that ends with a quick trip (Simply Misplaced, for example, that ends up reading a book), I have tried to do it with my pets and minions. I have not banished them.

As has happened other times, the game has crashed. I have re-entered the game and now my pets and minions are gone. I tried to complete the mission (remember, it has a fast trip) and it crashed again.
When I re-entered my pets and minions were still missing, which is correct. In trying to complete the mission I have been able to do it without major problem.

This indicates that a memory trace of these pets and minions may still exist even after they have been banished, and therefore the problem still exists.

To be fair, I have to say that now the behavior in other types of transitions (such as traveling through wayshrine) seems to be different.
When taking the wayshrine, the minions banish themselves and, after the trip, are automatically summoned again. You can see the animation perfectly. This was not the case with the previous update.

Some other sites are still not accessible. Without going any further, it is still impossible to enter Fungal Grotto (I) "manually" (accessing from the door in Stonefalls). The game crashes 100% of the time with or without pets and minions.

In summary, it seems that some work has been done to alleviate this effect, but seeing the crash logs, an EXC_BAD_ACCESS continues to occur for the same reason: insertion or deletion of non-existent elements. Smells a lot like race condition.

Seriously, crash logs provide a ton of information. Seeing the logs of the two crashes described at the beginning of this post is very enlightening.