I try to post a changelog, and my whole screen gets filled with this.. So I end up having to disable the link embeds completely.

If you could just make that image be a thumbnail (like the GitLab one above), that would already help.

I don't think I can provide a "smooth" 0-1 based on the animation.
It would be a hard flip. Which brings the question, does it flip at the start or end of that animation, and would that be annoying?

"isGunnerView" as name? You do mean "is zoomed in" but.. yeah.

But, what do you want to use that for, would that not cause visual issues?
Your one animation source, would be applied to every weapon, also the weapons of other units that you see infront of you.

The satellite assembly named "arma3launcher.resources.dll, PublicKeyToken=" for fallback culture "en-US" either could not be found

Check that the file exists in "steamapps\common\Arma 3\Launcher\En-us"

If not (or even if it does, to be safe) verify your game files in Steam

if it crashes, you need to submit the crash report, as you were already told in the ticket creation dialog.

See https://community.bistudio.com/wiki/Arma_3:_Unusual_Process_Exit#0xC0000135_-_STATUS_DLL_NOT_FOUND

Have you tried this?

Steam should be installing it all. But maybe the C++ Redistributable is corrupted? Maybe try reinstalling it.

Reason yes, modlist no (too large, too early in connect process)

which mods our audience actually has

Mods should always be loaded to match the server, not having random mods enabled and joining a server with a different modset. I don't see the use for that.

Lots of code depends on a magazine having less or equal to capacity rounds in it.
There seems to be too much code to be checked and fixed manually to respect this, to be worth it.
Not to mention how many mods/scripts expect "ammo" command to return a maximum value of 1.

Yeah normal Korean characters are supposed to work. But what I see in the dump just looks like corrupted garbage.
Thank you, I will try to reproduce with that name

Procdump dump is useful, but does not match your original report of stack overrun. It crashes while trying to check if it should clean up the MPMissionsCache folder, it crashes because something is wrong with the filepath.

I found these in the dump

LOCALAPPDATA=C:\Users\°íÁ¤¸ð\AppData\Local

C:\Users\고정모\AppData\Local\Arma 3\AnimDataCache\
C:\Users\고정모\AppData\Local\Arma 3\Arma3_x64_2026-08-29_22-11-40.rpt

That is a.... somewhat unusual username you have there. That is triggering the crashes.

GitHub is similarly bad (A bit smaller though), but at least their images provide somewhat useful information (Who created it, when, their avatar, how many comments)

Gitlab is nice and compact

The important information to be seen is the issue title, thats what we need in the embed.
A bit of the description can also be useful, but everything else is just wasteful, and makes people want to turn off embeds, which means information isn't visible at all.

This was closed as Done, but the issue is not fixed.

There is still way too much space filled with useless images. Just two feedback tracker links fill my whole screen space, and 42% of the space, is images that provide no value.

Well I can see the "NullReferenceException", and where it occurred.

Sadly, it seems to be nonsense.
It inserts entries (that cannot be null), into a list (that also cannot be null), and throws a null exception.

And this is code that every Launcher runs at Startup, and you seem to be the only person with that issue. I don't know what to do about that.

Spam tagging all possible tags even though they are irrelevant.

A non-descript title, and a useless description text.

That is the perfect way for me to just file your report as spam and ignore it.

The issue seems to have resolved itself (on Microsofts side), confirm?

Perfect! Thanks!

2.22 detects jumping and freezes the weapon, until.. it gets woken up by something else (like a vehicle driving nearby, a ragdoll flying close, or another physx object that might bump it).

If multiple guns are jumping close together, they might wake eachother up

Duplicate of https://report.bistudio.com/projects/arma-3/general/A3G-11

Something went wrong just before the crash actually happened. I cannot see what the cause was.

You could try switching to Profiling Branch, it does some things different, maybe you're lucky and that avoids the issue. But I don't know what the cause is so don't know for sure if that'd fix it.

Fixed on Profiling branch, and will be fixed on Stable in a Hotfix.

Please send the full report, the bottom left button from your screenshot.

System.MethodAccessException: Attempt by method 'MS.Utility.FrugalStructList`1<T>.Add(!0)' to access method 'MS.Utility.SingleItemList`1<T>.SetCount(Int32)' failed.
   at `1.Add(T value)
   at System.Windows.StyleHelper.AddContainerDependent(DependencyProperty dp, Boolean fromVisualTrigger, FrugalStructList`1& containerDependents)

A Microsoft function, fails to call another Microsoft function, from inside Microsofts own code.

I don't know what to do about this. I have neither access to the failing code, nor the ability to fix it.

But the Launcher works for almost all other players, so.. it must be something specific to your System.

Not sure what we should do about Proton not handling newer XAudio versions correctly. Downgrading XAudio for the Linux users is not an option.
But I found we also have support for sound via OpenAL. It is not installed by default (You need to download and install it from the OpenAL website), and we do not currently ship a command line parameter to enable it.

In Profiling Branch v5, I will adda a "-oalsound" command line parameter to switch to it. Maybe that will work better for you.

Windows 7

Windows 7 support has been deprecated since Update 2.20, and now with Update 2.22 we fully dropped support.

The missing DLL is xaudio2_8.dll. XAudio can be downloaded from NuGet, and its 2_9redist.dll in there, can be renamed to 2_8.dll and placed in the Game folder to work around that issue. But you may end up without Game Sound.

We plan on creating a wiki page to document how to do that, but it won't last long until we make more changes that don't support Windows 7.

Please consider not using a Operating System who's support ended 6 years ago.
You can switch to the Legacy Branch (Find the access code here) to play the 2.20 version which still works on Windows 7.

Mh I am unsure now, I found a second ticket with "No, this is a new issue", and that also isn't shown when I filter for it.

So maybe editing works fine, but the whole filtering is broken and just doesn't return any tickets even though it should?

The display does not respect browser locale.
I see "5:02 PM" instead of 17:02

> new Intl.DateTimeFormat(undefined, { dateStyle: 'medium', timeStyle: 'medium' }).formatToParts();
Array(11) [ {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, … ]
0: Object { type: "day", value: "18" }
1: Object { type: "literal", value: "." }
2: Object { type: "month", value: "08" }
3: Object { type: "literal", value: "." }
4: Object { type: "year", value: "2026" }
5: Object { type: "literal", value: ", " }
6: Object { type: "hour", value: "14" }
7: Object { type: "literal", value: ":" }
8: Object { type: "minute", value: "13" }
9: Object { type: "literal", value: ":" }
10: Object { type: "second", value: "26" }
length: 11
<prototype>: Array []

Ticket was closed as done.
But issue still present.
Firefox Nightly 155.0a1

This seems to be fixed now, so can be closed

The text in the top right, that I assume shows the condition, also doesn't show the correct condition.

Here the condition is is NOT feature request, but instead of "!=" the text box shows "=".