8 / 8 shorts · 1 / 1
Only while Palworld Dedicated Server was running on Windows 11, Deadlock would not advance past the loading screen after launch, and PUBG showed an "external provider authentication failed" error. When I stopped Palworld, Deadlock launched. After that, I changed Palworld's Query Port from UDP 27015 to 27100, and PUBG could launch while the Palworld server kept running.
Only while Palworld Dedicated Server was running on Windows 11, Deadlock would not advance past the loading screen after launch, and PUBG showed an “external provider authentication failed” error.
When I stopped Palworld, Deadlock launched. After that, I changed Palworld’s Query Port from UDP 27015 to 27100, and PUBG could launch while the Palworld server kept running.
The Palworld game connection was using UDP 8211, but checking the actual listening ports in PowerShell showed that PalServer-Win64-Shipping-Cmd.exe was also using UDP 27015.
Get-NetUDPEndpoint -LocalPort 27015 |
Select-Object LocalAddress, LocalPort, OwningProcess,
@{Name = "ProcessName"; Expression = {
(Get-Process -Id $_.OwningProcess).ProcessName
}}There is also a user report on the official Deadlock forum where the infinite loading problem was caused by port 27015 already being in use. In my case, Deadlock progressed when Palworld was stopped and reproduced when Palworld was started, so it looked like the same port-conflict class of issue.
References: Get-NetUDPEndpoint / Deadlock official forum solved report
I added -queryport=27100 to Palworld’s launch arguments. Before changing it, I checked that 27100 was free, backed up the world, and restarted the server.
PalServer.exe -port=8211 -players=5 -logformat=text -queryport=27100-port=8211 is the game connection port, and -queryport=27100 is the port used for Steam server queries. Steamworks also treats the Game Port and Query Port as separate values.
As of July 23, 2026, Palworld’s official launch-argument list does not mention -queryport. I tried a workaround used by the community and confirmed that, on this PC, Palworld listened on UDP 27100.
References: Palworld Server Guide launch arguments / SteamGameServer_Init / report of the same -queryport workaround
After the restart, the ports were assigned like this:
| UDP port | Process | Purpose |
|---|---|---|
| 8211 | Palworld | Game connection |
| 27015 | PUBG’s TslGame.exe | Acquired after PUBG launch |
| 27100 | Palworld | Query Port |
After moving Palworld’s Query Port to 27100, PUBG acquired the now-free UDP 27015 port and launched. The error message alone was not enough to identify the cause, but on this PC the Query Port change fixed it.
The Hamachi join address stayed the same as before: 25.x.x.x:8211. Participants did not need to change their settings.
This is a private server where people enter the IP address and UDP 8211 directly, so changing the Query Port does not change the join destination. If you publish the server to Steam’s community server list, check that the new Query Port is allowed through the firewall and router.
Related: Windows Firewall settings when Palworld Dedicated Server would not accept Hamachi connections
On Windows 11, I started Palworld Dedicated Server and tried to play with a remote friend through Hamachi. The client could not join, and nothing appeared in the server log. Hamachi showed both peers online, and the server was listening on the default UDP 8211 port. The cause was a leftover Windows Firewall block rule named Pal.
On Windows 11, I started Palworld Dedicated Server and tried to play with a remote friend through Hamachi. The client could not join, and nothing appeared in the server log.
Hamachi showed both peers online, and the server was listening on the default UDP 8211 port. The cause was a leftover Windows Firewall block rule named Pal.
Hamachi’s peer diagnosis said the tunnel and adapter were fine, but the traffic test showed inbound traffic being blocked.
With Windows’ built-in PktMon, I could see packets from the participant reaching UDP 8211 on the server PC, but the server sent 0 replies. The detailed trace also showed these drop reasons:
INET: receive inspection
INET: transport endpoint was not found
Inspection drop“Hamachi is online” and “PalServer is listening on UDP 8211” do not prove that packets can pass through Windows internally.
References: Hamachi peer diagnostics / Diagnose packet loss with PktMon
In wf.msc → Inbound Rules, there were two enabled block rules for PalServer-Win64-Shipping-Cmd.exe.
PalWindows Firewall prioritizes explicit block rules over conflicting allow rules. Microsoft explains that when the first-use app permission dialog is denied or closed, Windows usually creates both TCP and UDP block rules. These two rules had that shape.
Reference: Windows Firewall rules and precedence
After confirming that the executable was PalServer’s, I disabled only the two Pal block rules. Instead, I kept an inbound allow rule narrowed to these conditions:
PalServer-Win64-Shipping-Cmd.exe in my install pathHamachi25.0.0.0/8After the change, the capture showed 1,766 server responses to 10 packets from the participant, and they could join normally.
You do not need to turn off the whole firewall or allow every port. If you changed Palworld’s port, replace UDP 8211 with the actual listening port.
Reference: Palworld Server Guide: Configure the server
Shape of Dreams' Limbo 6 can be unlocked by editing the save JSON, not the game's config file. In this example, I make Bismuth count as having cleared Limbo 5 so that depth 6 becomes selectable. Base Limbo opens with four Evil Lucid Dreams, but starting depth 6 requires all six.
Shape of Dreams’ Limbo 6 can be unlocked by editing the save JSON, not the game’s config file. In this example, I make Bismuth count as having cleared Limbo 5 so that depth 6 becomes selectable.
Base Limbo opens with four Evil Lucid Dreams, but starting depth 6 requires all six.
Close the game, then back up the whole QuickSave folder:
C:\Users\<username>\AppData\LocalLow\Lizard Smoothie\Shape of Dreams\QuickSaveThe two files to edit have the same ID:
r_profile<UUID>.json
r_stats<UUID>.jsonIn r_profile<UUID>.json, mark these six pairs as unlocked. If one is already unlocked, you do not need to change it.
| Evil Lucid Dream | Achievement ID |
|---|---|
LucidDream_FishScales | ACH_BARE_SKIN_SUPREMACY |
LucidDream_GrievousWounds | ACH_VIVID_DREAM |
LucidDream_MadLife | ACH_MASTER_OF_EVASION |
LucidDream_MarshOfDestiny | ACH_EMPRESS_OF_PAIN |
LucidDream_Overpopulation | ACH_M_M_M_MONSTER_KILL |
LucidDream_PrudentJellyfish | ACH_WRIST_FRIENDLY_BUILD |
The values I changed for each pair were:
Achievement data:
isNew = true
isCompleted = true
currentProgress = maxProgress
completeTimestamp = current Unix timestamp
persistentVariables = null
Lucid Dream:
status = 2
didReadMemory = false
isNewHeroOrHeroSkill = false
isAvailableInGame = trueIf you only change the Lucid Dream side, startup validation may revert it, so I changed the corresponding achievement data too.
In the same file, set .root.preferredGameSettings.limbo to select Bismuth and Limbo 6.
{
"difficulty": "diffLimbo",
"hero": "Hero_Bismuth",
"lucidDreams": [
"LucidDream_FishScales",
"LucidDream_GrievousWounds",
"LucidDream_MadLife",
"LucidDream_MarshOfDestiny",
"LucidDream_Overpopulation",
"LucidDream_PrudentJellyfish"
],
"allowDejavu": false,
"customData": {
"GameMod_Limbo::depth": "6"
}
}Keep existing fields that are not shown here, such as lobbyName and heroSelectedLoadoutIndex.
In r_stats<UUID>.json, set Bismuth’s cleared depth to 5. Limbo progress is per character, so if you want to open depth 6 for another character, replace the hero name.
.root.heroes.Hero_Bismuth.completedLimboDepth = 5
.root.total.completedLimboDepth = 5After saving, run jq empty <filename> on both files to make sure they are still valid JSON before launching the game. If Steam Cloud reports a conflict, choose the edited local copy. Because this also changes achievement data, it may be reflected in Steam achievements.
Shape of Dreams' Limbo can be unlocked by editing the save JSON under LocalLow so that only the unlock condition is satisfied. The target is not the game's config file. It is this profile save file:
Shape of Dreams’ Limbo can be unlocked by editing the save JSON under LocalLow so that only the unlock condition is satisfied.
The target is not the game’s config file. It is this profile save file:
C:\Users\<username>\AppData\LocalLow\Lizard Smoothie\Shape of Dreams\QuickSave\r_profile*.jsonFrom the game code I checked, the Limbo unlock condition was having at least four Evil Lucid Dream entries unlocked.
Before editing, check which Evil Lucid Dreams are already unlocked and add only the missing ones. In this example, I add MadLife and Overpopulation to reach four.
The Evil Lucid Dream and achievement pairs are:
FishScales: ACH_BARE_SKIN_SUPREMACYGrievousWounds: ACH_VIVID_DREAMMadLife: ACH_MASTER_OF_EVASIONMarshOfDestiny: ACH_EMPRESS_OF_PAINOverpopulation: ACH_M_M_M_MONSTER_KILLPrudentJellyfish: ACH_WRIST_FRIENDLY_BUILDThe basic flow is:
r_profile...jsonACH_* entries for the Evil Lucid Dreams you want to add as completedLucidDream_* entries to status: 2 / isAvailableInGame: trueIf you only change the LucidDream side, the profile validation on startup may revert it. The point is to also mark the corresponding in-game achievement as completed.
This example adds MadLife and Overpopulation. After taking a backup, it uses PowerShell + jq to write to a temporary file, validate it, and then replace the save file. If you do not have jq, manually editing the same values is fine too.
$quickSave = "$env:USERPROFILE\AppData\LocalLow\Lizard Smoothie\Shape of Dreams\QuickSave"
$save = Get-ChildItem -LiteralPath $quickSave -Filter "r_profile*.json" |
Sort-Object LastWriteTime -Descending |
Select-Object -First 1 -ExpandProperty FullName
if (-not $save) { throw "r_profile*.json was not found" }
$backup = "$save.bak_$(Get-Date -Format yyyyMMdd_HHmmss)"
Copy-Item -LiteralPath $save -Destination $backup
$ts = [DateTimeOffset]::UtcNow.ToUnixTimeSeconds()
$tmp = "$save.tmp"
$filter = @'
.root.achievements.ACH_MASTER_OF_EVASION |= (.isNew=true | .isCompleted=true | .currentProgress=.maxProgress | .completeTimestamp=$ts | .persistentVariables=null)
| .root.achievements.ACH_M_M_M_MONSTER_KILL |= (.isNew=true | .isCompleted=true | .currentProgress=.maxProgress | .completeTimestamp=$ts | .persistentVariables=null)
| .root.lucidDreams.LucidDream_MadLife |= (.status=2 | .didReadMemory=false | .isNewHeroOrHeroSkill=false | .isAvailableInGame=true)
| .root.lucidDreams.LucidDream_Overpopulation |= (.status=2 | .didReadMemory=false | .isNewHeroOrHeroSkill=false | .isAvailableInGame=true)
'@
jq --argjson ts $ts $filter $save > $tmp
jq empty $tmp
Move-Item -LiteralPath $tmp -Destination $save -ForceAfter launching the game with this edit, Limbo was unlocked, and the Steam achievement was unlocked automatically too. If you do not want to touch Steam achievements, it is probably better not to use this method.
My Samsung 990 PRO M.2 SSD would sometimes suddenly stop reading and writing during games. Sometimes Windows still did not see that SSD after a reboot. Rebooting once more, or shutting the PC down and starting it again, would usually bring it back.
My Samsung 990 PRO M.2 SSD would sometimes suddenly stop reading and writing during games.
Sometimes Windows still did not see that SSD after a reboot. Rebooting once more, or shutting the PC down and starting it again, would usually bring it back.
Windows Event Viewer showed stornvme controller errors, disk Event ID 51, and Ntfs write failures around the M.2 SSD.
The only thing I changed was updating the SSD firmware in Samsung Magician.
It was 4B2QJXD7 before the update and 8B2QJXD7 after it.
On Samsung’s official Samsung Storage Firmware page, the note for 7B2QJXD7 says: “To address the intermittent non-recognition and blue screen issue.” The note for 8B2QJXD7 says: “To improve read-operation stability.” In other words, a firmware newer than my previous 4B2QJXD7 had already targeted a symptom very close to what I was seeing.
Reference: Samsung Storage Firmware
So far, the SSD has not disappeared or stopped reading and writing during games again.
The Steam version of SAND: Raiders of Sophie failed to launch with Unexpected error occurred. In the end, after setting Sand BE.exe to run as administrator and rebooting the PC, it started launching from Steam. The reboot seems to have been the trigger, so I still do not know whether that setting alone fixed it.
The Steam version of SAND: Raiders of Sophie failed to launch with Unexpected error occurred.
In the end, after setting Sand_BE.exe to run as administrator and rebooting the PC, it started launching from Steam. The reboot seems to have been the trigger, so I still do not know whether that setting alone fixed it.
This was the dialog I was seeing.
Before that, I tried Steam file verification, moving the install from the E drive to the C drive (the drive with the OS installed), and reinstalling BattlEye. At each point, the same error still appeared.
References: Steam Community / tinyBuild support
The last thing I changed was Sand_BE.exe in the game folder.
I enabled Properties → Compatibility → Run this program as an administrator, then rebooted the PC.
After that reboot, SAND: Raiders of Sophie started launching from Steam.
When I have time, I will move the install location back to the E drive and check whether it still launches.
In the PC version of Shadowverse WB, the QR code reader can create a deck from an image URL pasted into the file picker’s “File name” field. You can try decks found on X without saving the QR image first.
In the PC version of Shadowverse WB, the QR code reader can create a deck from an image URL pasted into the file picker’s “File name” field. You can try decks found on X without saving the QR image first.
For an image on X, right-click the image, choose “Open image in new tab,” and copy an image URL like https://pbs.twimg.com/media/xxxx?format=png.
In the file picker, paste the URL into the “File name” field instead of choosing a saved image, then click “Open.” It is just a small shortcut that skips the local download step.
If the URL includes &name=small, QR reading may fail. If it does not work, remove that part and try again.
This is an example X image URL used as the QR image.
After a PC migration, SourceTree may show fatal: detected dubious ownership in repository... for repositories copied from the old machine. I fixed the ownership mismatch by reassigning the target directory owner with takeown.
After a PC migration, SourceTree may show fatal: detected dubious ownership in repository... for repositories copied from the old machine. I fixed the ownership mismatch by reassigning the target directory owner with takeown.
This comes from Git for Windows’ ownership protection: if the repository owner does not match the current Windows user, Git blocks the repository to avoid unintended execution. Treat it as a copied-directory ownership mismatch before assuming the repository itself is broken.
Open PowerShell as administrator and reassign the owner of the target directory. Replace C:\github with the directory where you keep your repositories.
takeown /F "C:\github" /R /D YAfter running it, SourceTree could open the target repository again. On shared PCs or directories that contain other users’ files, keep the target path as narrow as possible.