Help Needed For FFXI Crashes.

Hi all. Sorry for the long post. I have been consistently crashing for the past 2 months and cannot seem to make any progress correcting it. I could really use some help.

I have an i9-1400HX processor and Nvidia 4090 laptop. Widows 11 and all drivers are up to date. Auto HDR and Optimizations for windowed games is off, and the GPU preference is the 4090 GPU. XBOX mode is off. I installed the NextHD 2022 Beautiful patch and ran it for 1 year with no issues. I occasionally added addons from windower (timers, autocontrol, etc.). In April 2026, I started straight to desktop, usually while in combat. The crashes can occur in town or in combat, and can happen 15 mins apart or 5 hours apart.

Since then, I have reinstalled FFXI, uninstalled / deleted the Next HD patch, reinstalled DGvoodoo, installed DGVoodoo2, repaired Windows, reinstalled Windower, reset my PC (and reinstalled FFXI / Windower again), and updated my motherboard BIOS.

I have completely replaced my job lua's to ensure no spelling errors or calling for gear I do not have. I get sporadic gearswap errors "Gearswap: lua runtime error: Gearswap/equip_processing.lua:62:attempt to index field '?' (a nil value)" which I have asked for help in the gearswap forum. I am not sure this is related to the crashes.

I have tried different variations of addons. If I run only gearswap, the game seems to run for longer periods before crashing. One thing I notice is the Timers plugin seems to guarantee a crash more quickly (less than 10 mins in combat). I suspect Config and FFXIDB can also expedite a crash but have not tested it enough to draw a conclusion.

The addons currently installed are autocontrol, azureSets, craft, DistancePlus, Gearswap, MountRoulette, TPPart, Trusts, and the DelayMeNot plugin. The 4GB patch is installed (I have tried with and without) DGVooDoo is current set to 4096 VRAM (I have tried it at 256 and 2048).

Has anyone else experienced issues like this? I am exhausted trying to troubleshoot it and could use some help. Thanks for the support!

A crash dump is below for thoroughness.

************* Path validation summary **************
Response Time (ms) Location
Deferred srv*

************* Preparing the environment for Debugger Extensions Gallery repositories **************
ExtensionRepository : Implicit
UseExperimentalFeatureForNugetShare : true
AllowNugetExeUpdate : true
NonInteractiveNuget : true
AllowNugetMSCredentialProviderInstall : true
AllowParallelInitializationOfLocalRepositories : true
EnableRedirectToChakraJsProvider : false

-- Configuring repositories
----> Repository : LocalInstalled, Enabled: true
----> Repository : UserExtensions, Enabled: true

>>>>>>>>>>>>> Preparing the environment for Debugger Extensions Gallery repositories completed, duration 0.000 seconds

************* Waiting for Debugger Extensions Gallery to Initialize **************

>>>>>>>>>>>>> Waiting for Debugger Extensions Gallery to Initialize completed, duration 0.015 seconds
----> Repository : UserExtensions, Enabled: true, Packages count: 0
----> Repository : LocalInstalled, Enabled: true, Packages count: 46

Microsoft (R) Windows Debugger Version 10.0.29547.1002 X86
Copyright (c) Microsoft Corporation. All rights reserved.


Loading Dump File [C:\Users\ncond\AppData\Local\CrashDumps\pol.exe.3332.dmp]
User Mini Dump File: Only registers, stack and portions of memory are available


************* Path validation summary **************
Response Time (ms) Location
Deferred srv*
Symbol search path is: srv*
Executable search path is:
Windows 10 Version 26200 MP (32 procs) Free x86 compatible
Product: WinNt, suite: SingleUserTS
Edition build lab: 26100.1.amd64fre.ge_release.240331-1435
Debug session time: Sun Jun 7 10:29:28.000 2026 (UTC - 4:00)
System Uptime: 0 days 20:23:04.314
Process Uptime: 0 days 1:27:18.000
................................................................
................................................................

Loading unloaded module list
..........................
This dump file has an exception of interest stored in it.
The stored exception information can be accessed via .ecxr
(d04.b5c): Access violation - code c0000005 (first/second chance not available)
For analysis of this file, run !analyze -v
eax=00000000 ebx=00000000 ecx=163dfd27 edx=161cfd23 esi=00000003 edi=ffffffff
eip=770ca07c esp=163df3b8 ebp=163df544 iopl=0 nv up ei pl nz na pe nc
cs=0023 ss=002b ds=002b es=002b fs=0053 gs=002b efl=00000206
ntdll!NtWaitForMultipleObjects+0xc:
770ca07c c21400 ret 14h
0:074> .excr
eax=161cfd23 ebx=0b2ba500 ecx=163dfd27 edx=161cfd23 esi=163dfd27 edi=0b2ba4e0
eip=770df95a esp=163dfcf0 ebp=163dfd40 iopl=0 nv up ei pl nz na po nc
cs=0023 ss=002b ds=002b es=002b fs=0053 gs=002b efl=00010202
ntdll!RtlpAcquireSRWLockExclusiveContended+0xeb:
770df95a f00fb137 lock cmpxchg dword ptr [edi],esi ds:002b:0b2ba4e0=????????
0:074> !analyze -v
................................................................
................................................................

Loading unloaded module list
..........................
*******************************************************************************
* *
* Exception Analysis *
* *
*******************************************************************************

*** WARNING: Unable to verify checksum for Hook.dll

KEY_VALUES_STRING: 1

Key : AV.Type
Value: Write

Key : Analysis.CPU.mSec
Value: 421

Key : Analysis.Elapsed.mSec
Value: 571

Key : Analysis.IO.Other.Mb
Value: 0

Key : Analysis.IO.Read.Mb
Value: 1

Key : Analysis.IO.Write.Mb
Value: 0

Key : Analysis.Init.CPU.mSec
Value: 359

Key : Analysis.Init.Elapsed.mSec
Value: 13773

Key : Analysis.Memory.CommitPeak.Mb
Value: 161

Key : Analysis.Version.DbgEng
Value: 10.0.29547.1002

Key : Analysis.Version.Description
Value: 10.2602.27.2 x86fre

Key : Analysis.Version.Ext
Value: 1.2602.27.2

Key : Failure.Bucket
Value: INVALID_POINTER_WRITE_c0000005_Hook.dll!Unknown

Key : Failure.Exception.Code
Value: 0xc0000005

Key : Failure.Exception.IP.Address
Value: 0x770df95a

Key : Failure.Exception.IP.Module
Value: ntdll

Key : Failure.Exception.IP.Offset
Value: 0x8f95a

Key : Failure.Hash
Value: {26b841e3-3725-b36b-3ecb-76d21b7c10c2}

Key : Failure.ProblemClass.Primary
Value: INVALID_POINTER_WRITE

Key : Faulting.IP.Type
Value: Paged

Key : Timeline.OS.Boot.DeltaSec
Value: 73384

Key : Timeline.Process.Start.DeltaSec
Value: 5238

Key : WER.OS.Branch
Value: ge_release

Key : WER.OS.Version
Value: 10.0.26100.1

Key : WER.Process.Version
Value: 1.18.13.0


FILE_IN_CAB: pol.exe.3332.dmp

NTGLOBALFLAG: 0

APPLICATION_VERIFIER_FLAGS: 0

CONTEXT: (.ecxr)
eax=161cfd23 ebx=0b2ba500 ecx=163dfd27 edx=161cfd23 esi=163dfd27 edi=0b2ba4e0
eip=770df95a esp=163dfcf0 ebp=163dfd40 iopl=0 nv up ei pl nz na po nc
cs=0023 ss=002b ds=002b es=002b fs=0053 gs=002b efl=00010202
ntdll!RtlpAcquireSRWLockExclusiveContended+0xeb:
770df95a f00fb137 lock cmpxchg dword ptr [edi],esi ds:002b:0b2ba4e0=????????
Resetting default scope

EXCEPTION_RECORD: (.exr -1)
ExceptionAddress: 770df95a (ntdll!RtlpAcquireSRWLockExclusiveContended+0x000000eb)
ExceptionCode: c0000005 (Access violation)
ExceptionFlags: 00000000
NumberParameters: 2
Parameter[0]: 00000001
Parameter[1]: 0b2ba4e0
Attempt to write to address 0b2ba4e0

PROCESS_NAME: pol.exe

WRITE_ADDRESS: 0b2ba4e0

ERROR_CODE: (NTSTATUS) 0xc0000005 - The instruction at 0x%p referenced memory at 0x%p. The memory could not be %s.

EXCEPTION_CODE_STR: c0000005

EXCEPTION_PARAMETER1: 00000001

EXCEPTION_PARAMETER2: 0b2ba4e0

STACK_TEXT:
163dfd40 770939e2 0b2ba4d8 163dfd7c 58f9f391 ntdll!RtlpAcquireSRWLockExclusiveContended+0xeb
163dfd4c 58f9f391 0b2ba4e0 0b219020 3a00b5b0 ntdll!RtlAcquireSRWLockExclusive+0x22
WARNING: Stack unwind information not available. Following frames may be wrong.
163dfd7c 58f9f304 0b2ba4d8 00000000 163dfe4c Hook+0xaf391
163dfd8c 58f328b8 0b2ba4d8 1dc834d3 00000007 Hook+0xaf304
163dfe4c 58f326e3 163dfef4 163dfedc 00000000 Hook+0x428b8
163dff1c 58f40688 37bbf8bc 00000000 1dc835a3 Hook+0x426e3
163dff3c 58fa7e78 37bbf8b8 1dc835eb 58fa7e20 Hook+0x50688
163dff74 75bf5d49 3a00b5b0 75bf5d30 163dffdc Hook+0xb7e78
163dff84 770be12b 3a00b5b0 fee0d2bb 00000000 kernel32!BaseThreadInitThunk+0x19
163dffdc 770be0b1 ffffffff 771053f5 00000000 ntdll!__RtlUserThreadStart+0x2b
163dffec 00000000 58fa7e20 3a00b5b0 00000000 ntdll!_RtlUserThreadStart+0x1b


IP_IN_PAGED_CODE:
ntdll!RtlpAcquireSRWLockExclusiveContended+eb
770df95a f00fb137 lock cmpxchg dword ptr [edi],esi

SYMBOL_NAME: Hook+af391

MODULE_NAME: Hook

IMAGE_NAME: Hook.dll

STACK_COMMAND: ~74s; .ecxr ; kb

FAILURE_BUCKET_ID: INVALID_POINTER_WRITE_c0000005_Hook.dll!Unknown

OS_VERSION: 10.0.26100.1

BUILDLAB_STR: ge_release

OSPLATFORM_TYPE: x86

OSNAME: Windows 10

IMAGE_VERSION: 4.7.9.0

FAILURE_ID_HASH: {26b841e3-3725-b36b-3ecb-76d21b7c10c2}

Followup: MachineOwner
---------

0:074> k
# ChildEBP RetAddr
00 163df3b4 74cea00f ntdll!NtWaitForMultipleObjects+0xc
01 163df544 74ce9e68 KERNELBASE!WaitForMultipleObjectsEx+0x18f
02 163df560 75c4c826 KERNELBASE!WaitForMultipleObjects+0x18
03 163df608 75c4c413 kernel32!WerpReportFaultInternal+0x3f5
04 163df624 75c190b9 kernel32!WerpReportFault+0xa1
05 163df62c 74de460b kernel32!BasepReportFault+0x19
06 163df6cc 770be1bd KERNELBASE!UnhandledExceptionFilter+0x27b
07 163df6d4 770ce4a2 ntdll!__RtlUserThreadStart+0xbd
08 163df6e8 770cd960 ntdll!_EH4_CallFilterFunc+0x12
09 163df714 770d24c0 ntdll!_except_handler4_common+0x80
0a 163df734 77105282 ntdll!_except_handler4+0x20
0b 163df758 77105254 ntdll!ExecuteHandler2+0x26
0c 163df820 770cbe9f ntdll!ExecuteHandler+0x24
0d 163df820 770df95a ntdll!KiUserExceptionDispatcher+0xf
0e 163dfd40 770939e2 ntdll!RtlpAcquireSRWLockExclusiveContended+0xeb
0f 163dfd4c 58f9f391 ntdll!RtlAcquireSRWLockExclusive+0x22
WARNING: Stack unwind information not available. Following frames may be wrong.
10 163dfd7c 58f9f304 Hook+0xaf391
11 163dfd8c 58f328b8 Hook+0xaf304
12 163dfe4c 58f326e3 Hook+0x428b8
13 163dff1c 58f40688 Hook+0x426e3
14 163dff3c 58fa7e78 Hook+0x50688
15 163dff74 75bf5d49 Hook+0xb7e78
16 163dff84 770be12b kernel32!BaseThreadInitThunk+0x19
17 163dffdc 770be0b1 ntdll!__RtlUserThreadStart+0x2b
18 163dffec 00000000 ntdll!_RtlUserThreadStart+0x1b

Does it take longer to crash if you enabled the large address aware flag for the game ?

If that is the case then you might have a memory leak somewhere


"Gearswap: lua runtime error: Gearswap/equip_processing.lua:62:attempt to index field '?' (a nil value)"

mean that something is trying to read a value that was never set.
like the amount of potatoes in a sack.. but the sack is not even there.
So you want to look at variables
does this happens with items usage/trade because that used to be a bug for me trading keys in abyssea

Considering this started in April 2026, my first guess is that it's related to whatever the new undiagnosed memory leak is. There have been a variety of reports on various channels of running out of memory and failing to load zone textures with configurations that worked perfectly fine prior to the last few patches. Running out of memory can cause failures to allocate. Unfortunately, there's not yet a solution or full understanding of the issue.

The underlying code in gearswap attempts to index item resources and an individual item resource. Line 61 does a nil check, so we know the item exists at the point of crash. However, the later attempt to index it sees a nil. This should not be possible, if res.items[item_id] was nil then line 62 would not be hit. If it was a non-table entry then the error would be different. Assuming you're using the same version on windower's github, and there are no flaws in their resources file leading to a resource not having the indicated data, this hints that your lua state is being corrupted in some way.

We've seen a lot of lua state corruption coming from failing intel cpus in the Ashita channel. Intel has stated that the mobile lines do not have the major production defect that desktop lines did, but consumer opinions are mixed. The desktop lines produced at the same time had severe production issues leading to instability of a similar nature and some users have reported similar issues on the mobile lines despite Intel's statement. In this case, you should consider the possibility that your CPU is toasting itself due to the suspected defect.

A more determinative analysis would require someone with Windower source to evaluate the raw crash dump. It still may not reveal a solution or cause, though.

What version of DgVoodoo2 are you using? Dege is constantly fixing things, which can include the occasional memory leak. Are you turning on Large Address Aware feature in Windower or using the patch? SE occasionally replaces pol.exe and you need to reapply the patch.

After both of those, what is your VRAM setting like? Going past 1024MB without Large Address Aware can cause issues. Also do you have ReShade loaded? Not sure what that Next-HD thing is, it might be using some older canned version of reshade so need to check that.

https://www.intel.com/content/www/us/en/download/15951/intel-processor-diagnostic-tool.html could help in checking for that possibility.

I didn't personally have an Intel CPU fail, but I and an LS mate both had early production 3rd gen Ryzen CPUs that degraded over time and saw a very similar progression of crashes.

There are well known programs to test stability of CPUs, you don't have to guess.

Thanks for all the responses! I lowered the DGVoodoo VRAM to 1GB and crashed on my first pull of 10+ mobs. I have the Large Address Aware patch installed and noticed the flag is also checked in Windower. I was also getting the Gearswap error much more frequently after lowering the VRAM. The error popped while I was alt tabbed away for 3 mins, when I tabbed from enemy to enemy (but not every enemy), while summoning trusts, in combat auto-attacking, etc. I tried again and unloaded gearswap once logged in and crashed on the 2nd or 3rd large pull.

The crash dump (according to CoPilot) states "The game is feeding invalid/corrupted data into the Direct3D 8 rendering pipeline, and D3D8 explodes trying to use it." I can share it if interested.

I then tested running 256 VRAM with no gearswap (had it turned off when I loaded the game) and did not crash after 1 hr of pulls. The Addon list was the same for all tests with the exception of removing gearswap for the last two tests - autocontrol, azureSets, craft, DistancePlus, Gearswap, MountRoulette, TPParty, Trusts, and the DelayMeNot, Config, and Timers plugins.

Tomorrow I will test 256VRAM with Gearswap and 1GB VRAM without Gearswap. I think this will help identify if it is my lua's or DGVoodoo settings or a combination of both. Any other suggestions to test?

I asked for help on the Gearswap Lua in the support thread if anyone can look at it. I verified it does not have any spelling errors and I own all of the gear. It was written by AI so it may have a critical flaw.
Ceowolf Gearswap

Couple of other things to note: Reshade is not installed, my processer passed the Intel diagnostic tool, DGVoodoo is v2.87.2, the Next HD package I was referring to is the Next Game's "Make FFXI Beautiful in 2022" package. My DGVoodoo settings are the same as in that video. I have no issues with other games (FFXVI, Avowed, and Subnautica 2).

Thanks for all the support!

Hand optimized assembly code can take the breath out of modern CPU's
This is a stable in my testing toolset

If you upload the actual dmp file we can look at it and see specifically what function crashed.

Can either post a link here or drop it in the windower support discord

Did some more trials today with and without Gearswap at different VRAM level with the follwing results. The addons were the same as before:

GS=Gearswap
1. VRAM 256 w/ GS - crashed witin 5 minutes in Norg, could not log back in until changing VRAM to 512
2. VRAM 512 w/ GS - crashed within 10 minutes of combat. For this crash, I got a connection terminated error from POL before it crashed to desktop.
3. VRAM 512 no GS - no crash. After ~20 mins of combat, I enabled GS and crashed 10 mins later

I still get the "Gearswap: lua runtime error: Gearswap/equip_processing.lua:62:attempt to index field '?' (a nil value)" error and it occurs more often the lower the VRAM. I used AI to try and find the source of the error but have been unsuccessful so far. Will keep trying.

PS For the avoidance of doubt I had Windower reinstall GS and then added my Lua's to the data folder.

would be interested to see if you can find a fix as i also have been experiencing massive DC's since changing over from laptop to desktop.

I'm back after a couple days of travel and did a few more trials. I focused on the Gearswap error I was getting by creating a super simple lua script but still crashed. I also tested without windower with DGVoodoo 1GB VRAM and crashed.

I plan to test vanilla no addon FFXI but am tired of entering my password for the day.

It has been suggested my CPU might be overheating. Temp monitoring shows spikes to up to 95-100 C range on a couple of cores. Average in the 60 C range. The ambient environment has also been warmer since the start of summer (but was not when the crashed started). I may try lowering the CPU voltage if I crash with any addons.

stop what you're doing and fix this

I am back trying to fix this and have had some success but still have an issue. Sorry for the long post. I did a clean install the January's Nvidia driver and, per SE's suggestion, set the game to run ad administrator in the compatibility tab.

After a crash, Windows help popped up asking if I want help with compatibility of an older program. I said yes and it made some changes (I am not sure what it changed). After the changes, I was able to play vanilla FFXI for 3 hours with no issue. I added back in Windower and gearswap and ran Apollyon 2x solo with no crashes. I continue to get "Gearswap: lua runtime error: Gearswap/equip_processing.lua:62:attempt to index field '?' (a nil value)" errors.

It was looking good until I went to the Central Basement with a large group of people. I crashed 2-3x while trying to tab to enemies.

Later, I was in Whitegate tabbing around NPC's and I crashed. Just prior to the crash, I got the Gearswap error above. I was able to recreate the error a few times. The gearswap error would pop abter approximately 3-4 mins of tabbing and running around the town. I would usually crash after about 5-6mins. I swapped to an old lua and was able to replicate the crash. I then changed to a super simple lua {no functions, 4 sets) and I got the error but was not able to get the crash. When in Gearswap debug mode, the older lua's show "entering user function xxxx" at every NPC. The simple lua does not do that. No gear was swapping during any of this.

Anyone have issues or errors like this? Why would this gearswap error pop while in town when no gear is swapping?

Thanks for the help.

Remove reshade if you installed it during the dgvoodoo process.

Thanks for the response. I do not have DGvoodoo or reshade installed at this time. I am considering reinstalling DGvoodoo since I have windower active again.

What I do not understand is why I randomly get the '...equip_processing.lua:62...' gearswap error while tabbing targets in town.

Additionally, the character select screen looks like it is at the wrong resolution (characters look short and fat) but in game everything looks normal. Where is the setting that governs the character select screen?