#arma3_feedback_tracker
1 messages Β· Page 7 of 1
Doesn't hurt to give them. Same with providing a example that can reproduce the crash.
Unplugged my mouse and tried with a keyboard with touchpad and same issue.
we need local variant of "setObjectScale"
there are times when it can become de-synced, and wiki says it has a network component
so client with the wrong object scale has no choice but to clog the network to sync
we don't need it
but it could be nice
supported objects are
- attached ones
- simple ones
the rest is unsupported hack
Well scale is a global property so it can't be local
Scale is "part of" the object rotation (e.g. vectorDir is direction * scale, so it's not normalized)
JIP damage is not synced for scripted damage (HandleDamage) done to pre-placed map buildings: https://feedback.bistudio.com/T172490
Will appreciate if this gets a fix because limiting damage done to pre-placed buildings is very important to KotH as its played in a single city over 1-2 hours and you can imagine it gets leveled pretty quick.
These ruins are also globally simulated even if no player is in the simulation radius π
(Not sure why they are simulated at all after the destruction was done).
Interesting. My original intent was to stop Ruins from spawning at all by only allowing windows to be damaged on the buildings but this ended up desyncing the damage completely. Wish this gets fixed so I can limit number of ruins in the mission.
In Antistasi we store a list of destroyed buildings on the server and then JIP clients hide them. I guess we leave non-destruction damage unsynced but apparently no-one noticed.
i move the engine ruins away and hide them, then spawn in replacement βsuper simple objectβ
@solid marten who do i have to beg for an βattachto/detachβ event handler? very valuable for sandbox logistics stuff
or even just hook the cargo loaded/unloaded so that it works for attach/detach
you could use a scripted EH
I have been thinking about doing that for our ruins in Antistasi. Only problem is if it is worth the time to do it. Have you noticed an improvement?
are you using handle damage EH to deny greater damage to map placed buildings?
Yes, my idea was to make all parts but containing "glass" invincible
perhaps another version of allowdamage command? with filter?
Why not fix HandleDamage to push proper values to JIP?
I'm not sure about internals of it, but it feels like engine always stores\sends engine-calculated damage for JIP, ignoring result of HandleDamage
iβm not sure if this will be a fix, could be not implemented intentionally. Also handle damage on all buildings that fires for every hit point and on minor damage is cruel to the user cpu
I only apply the handler to relevant buildings in playable area that have "glass" hit points, everything else is handled by default, its not that many entities in the end, probably a hundred or a few hundreds.
https://feedback.bistudio.com/T172509
Odd ticket. Appears to be spam, but there's no spam link included. And the person who replied has previously been an active and legitimate user (on the DayZ FT). Some kind of semi-advanced attempt to build rep for the new account by having a compromised old account reply to it? Seems like a lot of effort to go to for, y'know, Arma FT, and they didn't even get a believable topic for it.
it would have to be global, for instance if I want some server code executed on an attach/detach called by the player
the "cargoloaded" "cargounloaded" is perfect for the vehicle cargo attach/detach
Wouldn't it make sense for getShotParents to return [objNull, objNull] instead of [] when used on a non-shot?
Worth a ticket?
Wish there was something like pushBackUniqueFind, so it would return index of already existing element instead of -1
private _array = [1,2,3];
_array pushBackUniqueFind 4; // 3
_array pushBackUniqueFind 4; // 3
//_array = [1,2,3,4]
Revision: 150636
@solid marten Why is Man vs PhysX object (car, etc.) collision is not covered by neither EPEContact* or HitPart? Is there a way to have these EHs trigger on such collisions?
so two generic event handlers which would be handy are, "Simulation" and "Attach"
simulation event to trigger when simulation is enabled/disabled
attach event to trigger when object attaches/detaches
We can do some virtualization with a simulation event
and with an attach/detach we can do some accurate physx mass recalculations
https://feedback.bistudio.com/T169479
I asked for an attached object event and no one got back yet.
for those of us doing logistics stuff which may have to work alongside mods, its valuable
Hmm just found out that BIS_fnc_3DENModuleDescription won't work after the configHierarchy update
Wait for the next dev
Thanks
Another ~~ 2 ~~3 FT tickets with config issues
player unable use his NVG with iron sights
https://feedback.bistudio.com/T172106
IDAP Demining ED-1D drone disassembles into wrong backpack
https://feedback.bistudio.com/T171693
And 1 more FT ticket with config issues 
https://feedback.bistudio.com/T169067
Not sure if worth a ticket, but do3DENAction during mission preview can cause severe issues.
For example: do3DENAction "MissionSave"; accidentally executing this will wipe all data from mission.sqm with no way to get it back if there was no backup
I feel like do3DENAction should have a check that it's in Eden Editor
do3DENAction "MissionLoad" will crash the game
uff! nice find
Not so nice find. GitHub saved me this time...
well, and the other cloud backup, but someone else might be less lucky
it's defo worth the ticket.
Should be somewhat simple to fix I think. Early return if not in 3den.
@solid marten is there a reason the util computes the moon phases again and again? (or is the dynamic UI creation the slow part)
like it would be quite useful, but if you need to wait each time several seconds to try another phase, its makes you not really want to use it π¦
Might be quicker to just extract the code and search for the max
for "_i" from dateToNumber [_year, 1, 1, 0, 0] to dateToNumber [_year, 12, 31, 23, 59] step 1/365 do
{
_date = numberToDate [_year, _i];
_phase = moonPhase _date;
if (_date select 1 > _month) then
{
[_functionDisplay, "RscTextMulti", [format [" %1 %2", _months select _month, _year], "PuristaMedium", 0.06], [0,_cnt * 0.04,0.98,0.07], [0.15,0.15,0.15,1], [1,1,1,1], "", false, _ctrlGroup] call _fnc_addControl;
_month = _month + 1;
_cnt = _cnt + 2;
};
.....
also more handy would be such output imo
setDate [1944, 7, 17, 2, 00];//0.98
the multi select is nice, but could be made more useful if you get the moonPhase value as well
what? no idea what you mean
when you open it, it takes ~20-30 to show the moonphases for all months
if you open it again, it does the same again (no caching)
if you press set, it closes the menu
so you have to reopen it
20 30 seconds?
want a video?
yes because it takes 1-2 seconds on my crap pc
kinda. its ~10-15 seconds
please save us from this requirement to use "entities" whenever we use "allPlayers"
_allPlayers = allPlayers - (entities 'HeadlessClient_F');
allPlayers apply {hasInterface _x} π
What do you need an array of bools for?
i meant select
Someone check which is faster with the simpleVM changes π
Yes plz.
Wasn't hasInterface nular? 
did some digging with this as it's particularly noticeable with disposable launchers: https://feedback.bistudio.com/T123193
seems to be related to the loadInventory function that arsenal uses
as saveInventory correctly saves the magazine but loadInventory seems to... lose it?
I think once upon a time, the only way to load a magazine into a weapon was to put the magazine into vest/backpack and then add the weapon to the unit.
BIS_fnc_addWeapon still does it, and breaks in the full-backpack case.
Yeah
is it possible to have sqf manipulate the volume of ogv videos played? worth to make a ticket?
someone suggested it uses the "new" UI volume - there is no sqf command to manipulate that, is there?
what channel it plays it through?
audio channel from audio options i mean
one moment
@untold sky I think one of the most powerful potential optimizations, would be a way to toggle a unit between agent<--->unit
(creategroup west) createUnit ['b_soldier_f',position player,[],0,'NONE'];```
what actually IS the difference
AI
- agents have no groups
the AI in agents is very limited. it can pretty much do nothing but path find
but is it as simple as just "unit disableai "all" + deletegroup unit
all minus anim/path/etc
from your perspective
i stand corrected - no audio volume slider impacts video play. you can set all to zero, and still full volume - so seems like its own audio channel
for reference the call is
["VIDEONAME.ogv",""] call BIS_fnc_titlecard; //does
0 fadeSound 0
which uses
_titlecard = [_video, nil, nil, "BIS_fnc_titlecard_finished"] spawn BIS_fnc_playVideo;
which in the end does
private _videoBackgroundCtrl = _videoDisplay ctrlCreate ["RscText", -1];
_videoBackgroundCtrl ctrlSetBackgroundColor _backgroundColor;
_videoBackgroundCtrl ctrlSetPosition _screenSize;
_videoBackgroundCtrl ctrlCommit 0;
isNil
{
_videoCtrl = _videoDisplay ctrlCreate [["RscVideo", "RscVideoKeepAspect"] select _keepAspect, -1];
_videoDisplay setVariable [VIDEO_CTRL_VAR, _videoCtrl];
};```
is there vanilla .ogv I can try?
via VR in console:
_video = ["A3\Missions_F_EPA\video\A_in_intro.ogv"] spawn BIS_fnc_playVideo;
even with this, you still get it as its source volume
when I change my output device in Audio settings, if there was music being played in the same time, it will be restarted, while any other sound sources (like conversation audio) just continue from the part where they have been stopped. Expected behaviour or a bug?
other thing is that when during conversation when player stops the game with Escape, if stopped for a split second or half an hour the convo audio will be skipped by half a second or so, but I think this has been like that forever and "won't fix"
Here is the ticket about do3DENAction being able to be executed outside of Eden Editor.
Details: #arma3_feedback_tracker message
are there any do3DENAction actions that are supposed to work during preview?
As far as I can tell, there aren't. https://community.bistudio.com/wiki/Eden_Editor:_Actions
Revision: 150668 and possibly next profiling, please test
Thanks. Noted
I know it's been rejected in the past, but I was hoping it would be reconsidered to add a refresh rate video setting in-game.
The reason why is because in my case whenever I go into fullscreen windowed and then revert back to full screen, it sets refresh=85 in the Arma.cfg file. I'm using a 144 Hz GSYNC monitor and wish to play at 144 Hz whenever possible.
I don't know how I managed to make it happen, but yesterday, for some reason, my game suddenly switched to my secondary monitor, capped at 60 Hz. I hadn't noticed until today when I was playing around in the arsenal, but refresh=60 was set in the Arma.cfg file.
Having a setting in-game would permit me to overwrite whatever Arma is autodetecting. It doesn't even have to be an in-game setting, a startup parameter would suffice honestly.
I would make a mod for it, but I don't think there is a command for setting a refresh rate, is there? I know getVideoOptions exists, but I couldn't find a setter.
MaxFPS startup parameter would be nice. That way I could limit fps in Editor and have my GPU stay silent.
+1 to that, vsync isnt always the right option
Speaking of framerate limits, Arma 3 doesn't seem to work with vsync or adaptive vsync (nvidia) when running in borderless windowed mode
No game does IIRC as it's then handled by DWM/windows
With windowed (borderless) the game just renders to a window and windows handlers rendering of the window.
FPS limit is a thing already?
The startup parameters page says it's for servers only
So ogg has own decoder that handles the video, there is no dedicated volume or mixer control, perhaps something could be added at some point when we have fixed all the bugs and have nothing else to do π
ok thanks a lot for checking
i guess we can encode the video 3/5 times with different volumes and link to some audio setting to determine which version of the video gets played π¬
Or play the audio separately?
Since we have an entityCreated EH I figured maybe having an entityDeleted EH makes sense too? to avoid constantly adding a deleted EH to those objs (if you actually want to track the lifetime of objects ofc)
tho arguably since creation is more expensive than deletion I guess its performance penalty might not be justifiable
any thoughts?
if it makes sense I'll make a ticket
think someone requested entitydeleted a short while ago, would def be useful
I once wrote code with EntityDeleted before I realised that it didn't exist :P
I upvote, additional feature to complete the entity EH arsenal
anyone even came across sqf code that ran fine on windows servers, but bugs on linux servers? like not in runtime but during initial loadup parsing/compiling
CO10 Escape had a lot of trouble with that, IIRC.
Init works like 1/4 times or something.
Once it starts correctly then it's fine.
Probably some difference in how init scheduling works but I haven't even figured out the rules there on windows servers.
it makes a lot of sense
Bad conversion: scalar
this is a script problem isnt it? or also engine internal likeArray index out of rangeref: #arma3_feedback_tracker message
what are the chances we can get an export of snap point positions for house/terrain assets that have them?
These are memory points named .snap_xxxxxxxx used by terrain tools to place them, but are stripped out during binarization. The video shows how they are used (Snap in Eden and Zeus) and how I'm getting them out of A2 Samples (remove ., spawn model, selectionPosition of mempoints).
Bug but likely nofix
The only thing I dislike about it, is having to implement it
I have heard of it but never had any on hand that I could repro. Would love to have though
This happens when a script value is read as number, even though its not.
Usually some engine bug as I don't think you can do that in script (unless engine side typecheck is missing)
It already has context enabled, its an error, all errors have context.
If it doesn't print any, it means it doesn't know any context.
But it should happen during script execution, and every script has the calltrace context? So no idea whats going on there
is there an in-engine way to somehow reenable "light glares" like in early alpha, or are they completely out of the game already?
that just looks like edited photo
it was in-game until a certain (alpha? beta?) early access version
you mean without a mod?
yes, that was the game
but you want to change it without a mod?
yes, I want to know if such effect is still somewhere in the render code
to eventually be able to reenable it, either through a command, a mission setting or a startup parameter
I see, that's why I asked π
not sure...
I mean, i'm not sure if the effect exists, changing it without a mod, i'm pretty sure there's no way.
I am asking in this channel to get a way to get it back :p
Avery's work, see if she* uses her* own, or if it's from the game:
https://steamcommunity.com/sharedfiles/filedetails/?id=878502647
From the description of that, the lens flare is just a texture and it can be replaced. The old texture file must exist somewhere since Avery found it (possibly not actually in the current files, but...somewhere). So it's probably not a complicated renderthing that would be a pain to change. Maybe a pain to make a way to change it on the fly though.
try CfgCoreData
its this iirc they changed the lensflare texture
please see: https://feedback.bistudio.com/T172963
kindly provided @loud bear πββοΈ
If player has an object view distance less than 526m the beam from laser pointer will be invisible if aim to the air
how does the game check if an object is a helipad? I looked at the config and the model's named properties but I don't see anything special about them 
are those hardcoded classes? and the game simply does a isKindOf check?
yes
thanks
HeliH
also a "super" simple object helipad doesnt work for AI landings
yeah I was just testing that π
yes, as it's an model only with no class.
We had this issue in CLN where our helipads were simplest object types.
simple ones don't work either btw (which makes sense because isKindOf "parentType" doesn't work on them)
FT ticket
https://feedback.bistudio.com/T172970
(Nevermind, you guys already figured it all out)
I saw that π π
I agree that Alpha lens flare looked sick
A feature request to expand unsafeCVL to allow list of classes that you can createVehicleLocal
a re-creation, as unsafeCVL doesn't exist (anymore) afaik?
I think it does, it was disabled at first due to some issue but been active for years now
π€
Huh, I had unsafeCVL = 1 all this time and thought the feature was there
So there is no way to limit createVehicleLocal at all?
what about https://community.bistudio.com/wiki/Arma_3:_CfgRemoteExec & BE filters?
I guess BE filters is a way, limiting command to certain list of class names sounds much better though
oh I don't diss your suggestion at all π actually, a wait to prevent any command execution from mission with CfgRemoteExec could be nice - the mission maker knows (or should know) what he allows or not (mods are another story)
I guess you mean CfgDisabledCommands?
ah yep
is it ok to send the crash report directly to one of the devs? like the mdmp file or whole crash report
Dedmen collects these, yes
but you'll have to wait till monday so you could just make a ticket with mdmp as attachment
ok
ticket just seems pointles as I dont really have a repo
something wrong with the console, it crashes sometimes
no need to give repro if it's a crash report and you have the mdmp file, just like here: https://feedback.bistudio.com/T172782
ok though the crash report generator found 3 mdmps from somewhere

time to make a new one then
sorry new what?
new mdmp
let me restate - play long enough to make it crash again
no mods?
some..
then it might be caused by a mod rather than the game itself
something in my mod preset brings back the setting for number of audio channels used by game, not interactive but still. Perhaps your mods do something to the base game too
well most of the mods I have are for deving and I've been using them long time now
mods get updated every now and then.
true
yes my guess is though that it has something to do with the new script error EH. because the crashes started after that was implemented. just a guess though
once you manage to get that crash again, you could try switching to perf branch and compare, maybe something in "beta" fixes this and it won't be a mod issue at all
im on prof
then revert to stable
ok
if stable crashes - mod issue. If it doesn't - prof/perf issue
yeah
is it just me or did the order of array obtained with diag_stacktrace change at some point? and is it same for ScriptError EH?
btw spam: https://feedback.bistudio.com/T173040
thanks, removed
AI spam is here I'm afraid
ah no, just copying from other places (https://feedback.bistudio.com/T172283) and inserting the link randomly
Not sure how much time there is left for 2.14, but I'd really appreciate if you could have a look at https://feedback.bistudio.com/T124694 bug. It is a very big gameplay breaker for mods like RHS, because it lets you kill players inside tanks because of their glitched out animation. @untold sky
The bug happens because vehicle doesn't have get in and get out animation setup for the seats and remote clients sometimes (rather often) end up seeing you in this default animation.
Most of vanilla content is unaffected but lots of modded content is
Is there any possibility of inheritance-compatible array append in config? (https://feedback.bistudio.com/T79801)
It's very hard to make cross-compatible extensions to any kind of array property, because if you want to extend an array that you don't initially define yourself (let's say pylon hardpoints because that's what I'm doing) you have to explicitly list everything. So if I'm adding hardpoints to an existing class, I'll conflict with any other mod that does the same, and if I'm making a new class with more hardpoints, I won't inherit any changes from other mods that extend the base class. A real pain.
I understand if it's too deep in the hellcode to do, but if it is possible, it would be awesome.
if the freeExtension would be fixed and made available at prof branch it would make C# and GO users life easier (https://feedback.bistudio.com/T167202)
how are you generating your c# dll?
aot?
or just dotnet build?
@gray wharf ^ one for you
what's that? π I'm using VS 2022
thanks - at first all the comments looked normal, no weird links, no nothing,
but actually a load of bullcrap indeed - thanks for your eyes ποΈ
i could never get callback working with AOT
You still need to define p/invoke
On the callback function.
I could make an example to show you later
ive no clue what that means
I don't remember touching it since it was added
ok then it must been me who forgot
class A {
arr[] = {1,2,3};
}
class B : A {
arr[] += {4}
}
this?
Yes. Or it could be a new operator or keyword if changing that isn't backwards-compatible.
Should be fine
Just need to make sure to not do the append twice.
Will work:
class A { arr[] = {1,2,3}; };
class B : A { arr[] += {4}; };
class Z;
class A : Z { arr[] = {1,2,3}; };
class B : A { arr[] += {4}; };
Will not work:
class A;
class B : A { arr[] += {4}; };
class Z;
class A : Z { arr[] += {3}; };
class B : A { arr[] += {4}; };
So if it can know the definite value, it will concat them directly. Like Mikeros tools do afaik
Well, that's what we have now. The ticket/question is about making it work with inherited arrays and not just ones you've defined yourself.
So both in the red box I'm fixing then
hmm, testing in description.ext both of these cases seem to work fine, not sure about addons.
my example in the gist is bit more trickier case and not sure how it should work 
description.ext is never merged, thus also no external inhertiance (as there is nothing that doesn't exist inside the one config file)
addon configs are resolved via merge
do you happen to know / are you able to know when += was introduced in A3?
yes I knew that once
Not now
Ah
class A
{
arr[] = {1,2,3};
};
class B : A {
arr[] += {4};
};
So yeah that one already does work
class A
{
arr[] = {1,2,3};
};
class B : A {};
class C : B {
arr[] += {4};
};
This also works, but wiki says it doesn't π€
https://github.com/NikkoJT/DuctTapeFactions/blob/main/DuctTapeFactionsCW/spetsnaz/spetsnazCfgVehicles.cpp
I define a weapons[] array (line 369).
When I append to it in my own inherited class (line 395) that works fine as it's in the same context.
At line 907, if I were to instead have hardpoints[] += { "gm_mi2_pylon" }; rather than a full explicit definition, it would not work - the final array would only contain "gm_mi2_pylon" and not any of the other items - because the inherited array is not defined in the same context.
(I have tested and proved this, it was the source of the request)
Yeah I don't think I can ever get that to work right.
config1 and config2 yes.
But, after the merge was resolved, the array forgets where values came from.
So after config2 is loaded, we have A1, A2, B1, B2
But if you have a config 3 that then does
A += A3
The B one is already resolved, and it won't re-resolve
Fine for me, that gist is more of an explainataion why we have certain things structured the way they are and it's linked in our SOG config.
We expand the muzzles array in multiple PBOs in SOG and it works fine.
I think your issue might be related to load order/ requiredAddons
I'd need some minimal repro (not nested deeply in some vehicle config in a DLC in a file with a thousand lines)
But considering my load order example issue is not fixable, it may not be worth to invest more time into this
https://community.bistudio.com/wiki/Array%2B%3D Don't know whats up with that wiki, the cases that shouldn't work do work, atleast via loadConfig command
The gist example from veteran should work though, when config2 merges, it should see that items is fully defined inside B
The ticket (https://feedback.bistudio.com/T79801) contains an example attachment although it's pretty old and I'm not sure what's in it
weapons_f_vietnam_c
class Put: Default
{
muzzles[] +=
{
"vn_mine_muzzle",
"vn_classic_mine_range_muzzle",
"vn_directional_mine_range_muzzle",
"vn_bounding_mine_range_muzzle",
"vn_classic_mine_wire_muzzle",
"vn_pipe_bomb_muzzle",
"vn_punji_muzzle"
};
weapons_f_vietnam_03_c
class Put: Default
{
muzzles[] +=
{
"vn_water_mine_muzzle"
};
stuff like this works.
Put.muzzles contains all of our and vanilla muzzles.
the page was created on 8th January 2015β
the patch before was v1.36 (4th December 2014)
the big update before that was Helicopters & v1.34 (4th November 2014)
I just checked and my requiredAddons correctly requires the addon it would be inheriting from.
Page was created by mikero, so that could be any time after the feature existed
correct, I just wanted to see if that rang any bell
March 2011
But the support for inherited values was september 2013
Okey found bug
// Find base array
baseArray = parent.findEntry(this.name);
That's veterans bug. It only checks in base class, not in itself.
so 1.00 / 1.02 perhaps π thanksβ¦
what is "March 2011" though? += introduction⦠for what?
March 2011? Isn't it... Futura days?
2011 does for when loading binarized config (only that, not text)
of this style
class A
{
arr[] = {1,2,3};
};
class B : A {
arr[] += {4};
};
No support with merging with other configs
I'll set 0.50 + 1.00 support + 2.14 fix I guess
Person who wrote that, decided to resolve the append when config is loaded. Instead of doing it at merge.
That causes veterans bug, it resolves it at a point where it doesn't even have sufficient information to resolve it.
Then, after it resolved it, it clears the flag and turns it into a normal =
So then later, at the merge step, it doesn't know anymore that it was a append at all
big thonks for the history dive, I like it
also for the explanation π»
So in theeeory, I just remove that on-load resolving, and let the config merge resolve it instead. And it should
?
Chances of things exploding, 75%
hears a distant explosion everything ok?
class A {
arr[] += {1,2,3};
};
class B : A {
arr[] += {4}; // arr = [1,2,3,4]
};
If this is loaded as a standalone config (description.ext, mission.sqm, server.cfg, profileNamespace) then this is totally fine and correct.
But if the config will get merged later, and this config gets merged before it
class A {
arr[] = {5};
};
then its wrong as we now expect [5,1,2,3,4]
But it takes too much work to fix this.
When a config will be merged, we need to skip resolving during load. If it will not be merged, we need to resolve during load.
But when a config is loaded, it doesn't know where its going, it might be merged later, or not.
I could remove it from "during load", and add a postprocess step. So that we either call postprocess or merge.
But, there might be configs that are loaded standalone at first, used standalone, and later merged.
Basically, technical debt says no
@untold sky
This token code is broken as it bears no relationship to any other tokenised output. It 'works' simply because its data format can be ignored.
can be deleted? I have no idea what it means
well, the whole
It originally had a text-only interpreter making it quite impossible to generate config.bin's versus config.cpp's. Later, Binarize was modified to create new token code for this syntax.
This token code is broken as it bears no relationship to any other tokenised output. It 'works' simply because its data format can be ignored.
tbh
β¦thanks, I needed confirmation
Actually I can make change in that case.
The A one is an append.
So I can say "well you clearly expect it to be defined already, so I'll trust you knowledge and skills and expect that this will get merged later"
That would fix alot here, but it breaks if you do this thing in description.ext (which really noone should as it would be wrong..)
tis gud?
I wrong
I missed the text merge
text was supported, but inheritance with merging wasn't
so single configs worked in 2011.
Addon configs that merge into classes defined in other config via requiredAddons, only 2013
aaah okΓ© π
In binarized they do
if (wantMerge) then doMerge;
in text they do
if (!doMerge) wantMerge = true;
confusing
I'll go with this then
https://community.bistudio.com/wiki/Array%2B%3D haaah, neater.
Hah pettka on that FT ticket
The description is correct - this feature was never meant to work with inherited classes. The goal was to create a simple tool to work add into array with unknown previous values (e.g. Throw muzzle with various grenades) and it works just like that.
If it was never supposed to work that way, why was that the way it was implemented in 2011
Config 1:
class A {
array1[] = {"1"};
};
class B: A {
array2[] = {"1"};
};
Config 2:
class A;
class B: A {
array1[] += {"2"};
array2[] += {"2"};
};
Actual result in the game:
class B: A {
array1[] = {"2"};
array2[] = {"1", "2"};
};
lol this one is fun.
In the merge step, someone wrote we don't want to inherit. Because we don't merge into inherited values, we overwrite.
UNLESS, its a array append.
So during merge it checks "does B have array1 when i ignore its parents?"
Well no it doesn't
"Okey then I create a new array!"

As far as I can see, this one is safe to fix
We'll try on prof π€
I don't know which of these fix which of the reported problem cases. We'll find out I guess (I'll need feedback please, gonna be next prof tomorrow or next week, probably next week)
Seems to be working, and seems like we can make a prof tomorrow
great job!
Clicking on a empty CT_COMBO crashes the game, not sure if known or it.
ticket with dump plox, haven't heard of that before
Is there any chance of having an option to disable the "autopilot" that kicks in when you open a display or w/e while piloting added?
found this in my RPT today:
SHIFT pressed during game start. No 'GameConfig'.cfg mods will be loaded!
These no-group no-name not-isPlayer players in multiplayer are annoying as hell
An issue since OA or probably earlier, still a headache
Happens all the time in multiplayer randomly, maybe there can be some kind of trap setup for this kind of bug?
Is this supposed to work like this:
a = createHashMap;
a getOrDefaultCall ["test", {
if (true) exitWith {1};
}, true];
a
// output
[["test",<null>]]

I'd say yes?
Additionally sets key to the code result value if 'set' is true.
Gonna go with no.
Looks like exitWith somehow doesn't work correctly with that one.
exitWith doesn't work with select either for example ```sqf
[1,2,3] select { if(_x == 2) exitWith { false }; true };
exitWith exits loops so that one is expected
But I would expect it to work for this
scopes too
Spam account
It would be very cool if all commands involving classnames were case-insensitive (https://feedback.bistudio.com/T126935 but list may not be exhaustive)
Ah probably the usual exitWith issue thats also present in other stuff like eventhandlers.
So wrap it in call {}
and its probably fine
exitWith exits current scope.
Problem in the above, there is only the root scope
Likely no fix, as its a deep issue
Roger roger
@untold sky Sorry for ping, but I think your comment here is, er, missing something https://feedback.bistudio.com/T173243#2462143
World's biggest spam comment, and...ChatGPT trying to answer the ticket? https://feedback.bistudio.com/T173402#2461968
removed, thanks
π π€£ 
helpn't
https://feedback.bistudio.com/p/DiannCole/
Spam account
b-but they posted in Argo category?!
thanks, nuked π
infinite recursion in circular reference hashmap handling π
// works
a = createHashMap;
a set ["a", a];
a
// RPT and CPU goes BRRRRRRRRRRRRR
a = createHashMap;
a set ["a", a];
a set ["b", a];
a
I think for performance reasons, thats intentional
arrays do a recursive full iteration through the full array on every insert, I didn't want to pay that cost for hashmaps
Just don't fuck up please
Though the to string conversion which you're doing there, should catch it
It's bit annoying it can't be used as I have a valid use case for having a reference to parent hashmap in the children.
The ToString thing should detect recursion and abort
as long as you don't ToString it. You'd be fine with recursion besides causing a memory leak (self reference == never unreferenced)
that sounds like you are trying to make a tree out of a hashmap?
Trees are map objects, so it should work, right? :D
So turns out remoteExec JIP is just as unreliable as publicVariable. You randomly end up without some JIP REs or PVars. Seems to happen during round change. No repro of course, only happens on live games.
https://www.youtube.com/watch?v=_tyISRL_2G8
Very old feedback but is this even a feasible/easy idea to implement?
Highly scripted demo, not implemented at all on the MOD nor the game. Used MOD for reference: Arma 3 Aegis
"This video was created using content of Bohemia Interactive a.s."
"Copyright Β© 2019 Bohemia Interactive a.s. All rights reserved."
"See www.bistudio.com for more information."
I wish it is possible to use rtm for every other objects like weapons BTW, IDK
Yeah, we expierence that in Antistasi sometimes
we normal just declare a string like JIP but sometimes just having the object JIP works
it's a hit or miss
That part's deterministic.
You just can't use the same JIP parameter twice.
Weirdest thing I saw with remoteExecs is them occasionally hitting the server after a mission restart. Like they're not mission-fenced.
Would be nice to get it fixed, we have ugly scripted workaround for this in SOG as it was causing our basket backpacks sticking out of vehicles when units die 
Then fix your CDLC!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!eleven
Gimme binary sources and we can think about it 
I support this message.
Thank you, can repro.
And its so ridiculous.
Why does a AI message cause backpack to appear π€£
Oh no... the backpack is up!
But yeah, it is very strange to believe why this bug is even a thing
lol wat.
There is a thing in proxy rendering. Where the backpack is only rendered if the unit has no brain.
(The brain is deleted after the unit is known to be dead)
Aha

The backpack is rendered if the unit does not have a break, or if it is not is not a freely walking unit
break?
Because the "is in vehicle" state is stored inside the brain, and if it doesn't have brain it just assumes its not inside a vehicle
I'm sure we can get that in vehicle state without brain too
so, something like "on KIA confirmed unit gets decoupled from the group and replaced with dumb corpse with different render settings"?
no
As I said, when unit is confirmed dead, its brain is removed, because it isn't needed anymore
So, does this mean the bottom-left group menu is essentially the list of βbrains under youβ?
yea
Sounds like a decades of code does this issue
easy fix for the backpack..
But now the NVG is still there π€
Ah same thing just slightly different :harold:
ahh uff.. I will leave the NVG broken I think. Its not as bad asa the backpack.
It does special magic with vehicle config entries saying whether to show NVG or not. So I need to get the vehicle and not only check if it exists.
I think not worth it for that small visual issue
Hm, I thought whether show backpack or nah is also configurable
no way to show backpack from config entry, only nvgs on everything except ffv
a machine
beep beep
you know it would be handy if you would get error message if you pass non-local argument to command that expects local argument. that way you might get some bugs fixed π
That's so cruel. Poor AI
I got a repro for that USB Audio crash π
Run Arma as single-channel audio device with 48khz, and I get the crashing.
Many of these crappy USB Headsets like the Arctis 7 often run on single-channel mode
Now just gotta figure out why it does that.
Luckily Microsoft made a awesome tool to analyze memory corruption.
Sadly its broken SPECIFICALLY when using it with XAudio, which is what we use, and microsoft said in 2021 that they fixed it, but its still broken
So I have a amazing tool that would tell me exactly what causes the issue here, but I can only use it with Audio disabled π€£
O noice
i get a crash with my bluetooth hearing aids disconnecting that i think is related to the bug -- never been able to get a repro though as sometimes its just fine and other times it crashes
There is a different thing on disconnection I think
This is the one where it crashes when you try to start a helicopter
hmm. So is Arma having to deal with live audio output buffer format changes, or is that a library-level thing?
now that we have lineIntersectsSurfaces with Roadway, would it be possible to fix the Zeus interface issue where objects are placed below surfaces?
e.g. if you attempt to place some units on a carrier, they end up in water. same thing goes for buildings, etc.
works fine in 3den tho
Arma doesn't have to deal with changes.
Because it just stops working and goes mute
ah right, wasnt aware
probably unrelated then sadly, cant rest my head on my hand or they disconnect and arma crashes π
BAM! Got it!
All it took was implementing my own memory access monitoring :harold:
SetupOutputMatrixUnsafe
Unsafe you say? UNSAFE?! Really? 
Y u do dis.
We get an array of two floats.
And we set number one and three π
Amazeballs.
ONE out of bounds write == 10 years of Arma crashing. Man..
which crash, the audio stuff?
yeah
All the crashes that I've labelled "USB Audio crash" over the last years
Well I am not entirely sure that its the "USB Audio crash"
But, running Arma on single-channel audio, reproduces these same memory corruptions
I don't have data on whether the other USB related crashes were also single-channel audio, we only recently added that RPT logging for that
wrap the whole game in a try catch :p
The commit that introduced this bug was "Sound: Fix: Corrupted sounds with certain USB headsets" π Oh the irony
it technically didn't lie :p
Now tracking down the dozens of internal and FT tickets relating to this to close them all 
I think that also caused related crashes like the "map marker texture loading" thing
I found a comment from myself from 2021 pointing at that old commit as "Potentially related"
Hell yeah. Only thing that was missing was me actually realizing that its exactly that
And it still is the best game!
Only took 7 years
hey, he types slowly okay??
Solving: 2h
Testing: 2h
Ticketing: 4h
have a nice day π
solving more like 12 hours, not counting the dozens of tickets that other programmers tried to nibble at over the last decade.
70 links to related tickets in internal tracker, and didn't count how many links on that FT ticket 63 FT tickets
anyway, on behalf of many USB headset players, well deserved congrats on that one π
Oh wow, big congratulations π
https://feedback.bistudio.com/T173277#2465402
spam, pls remove
oooh and a sneaky one at it! well spotted, removed - thanks!
Congrats on the audio stuff!
Now a nice QOL thing would be to auto play from laptop speakers when I pull out the earphones rather than having to go into Audio and pressing Refresh
I thought about it but it's not worth the time investment
It would save you the few seconds of pressing refresh. Really not much badness
I fear it would involve a lot of audio output rework?
Easiest way would be to detect this case and Auto press the refresh button for you.
But even that isn't worth the effort. Just live with having to press the button
π§ make a keybind for that button
wait, so when the game loses audio we can make it re-output from the current audio device now?
Yes. Pressing the refresh button in audio options restarts the entire audio system
can't wait for the FPS fix
oooh
if (_username != "honger") then {
_fps = _fps + 10;
}
btw I am not sure how this audio system restarting really works, when I am in a helicopter (or there is one near me) and I change the audio output/refresh it the rotor sound will be played infinitely (lol)
Yeah that's fine with me, I'm sure BI have bigger things on their plate atm
Incidentally, does the refresh button also refresh the input device? And if not, could it? People getting the wrong mic autoselected by Windows, and then having to restart the game after fixing it, has historically been a large source of friction
not really, I hear the helicopter engine even if its away
but I guess your issue is included in mine
something was 'fixed'?
If I'm not mistaken I could get this bug with tank and car engines too, but on current dev I can only get this with helicopters
It might refresh, don't know.
Input device handling is special, which is also why no input selectors
You mean REs from previous round?
Well, mission restart. We don't do rounds.
Meant mission restart by round, yes
I forget whether they were server->server or client->server.
Thinking about it, perhaps JIP RE arrives while player is still loading somehow and gets ignored? Seeing how it behaves exactly like lost PVs under same conditions, its probably a more general issue, messages arrive and get ignored.
Also, setTerrainHeight fails to arrive to players sometimes too, not sure if its JIP or same issue of round restart.
IME JIP RE consistently shows up very early in the client init process.
Like you can't expect anything from initClient to be defined yet.
initPlayerLocal rather.
Hmm, could it be that next mission RE stuff arrives while player is still in previous mission session?
And gets wiped on next mission initialization?
I guess that might be possible.
All in all feels like JIP is unreliable on mission restart and should be replaced by manual JIP request from client side on initialization, as we can't trust the engine.
It's not really an issue for us as we don't do mission restarts unless an admin screws up and gets lazy :P
Won't solve setTerrainHeight unreliability though, maybe check getTerrainHeightASL on a known edited terrain position and kick to lobby if it failed to arrive?
We do 2 hour rounds so mission restart happens very often through ~24h server uptime, so it is very noticeable.
I think the common issue before was that an admin would forget to change a parameter, restart the mission to fix it, and then you'd get the impossible startup errors.
Do you have lobby timer on your servers?
That 90-60-whatever seconds countdown in server lobby with role selection
Our servers have it removed (not sure how actually) so players are instantly thrown into next game
Thought maybe having no lobby delay could influence it
They're generally running on a mission cycle and I never figured out what that does by default.
I don't use mission cycles myself because I don't trust it.
Yeah, mission cycle is impossible to control, we use #mission through serverCommand, not sure if that's the reason for no lobby timer.
Lobby timer is controlled through server config. Setting it to 0 presumably makes it instant.
https://community.bistudio.com/wiki/Arma_3:_Server_Config_File
roleTimeout/briefingTimeout
yes we have a lobby timer
on the community servers
Yeah, must be it
Do you still get that issue of JIP REs and PVs being unreliable on mission restart\change?
Does the lobby timer have any visual effects?
Not sure about "still" because I only ever saw a handful of cases.
It has a timer at the top of the screen, that's it.
I believe the timer is disabled if a logged-in admin is running slotting, but I'd have to double-check that (....not now)
Just noticed there's also a skipLobby parameter
yeah but that one is madness :P
Did run into a couple of people accidentally running it.
We don't use it, it causes players to be randomly ditched off the server during mission restart
Didn't bother reporting it
I haven't seen it recently but, another dev did have to change the way that we did jip because of some reliable issues.
lul
lol get rekt
Not sure whether it makes sense or not, but I somehow expect to have sidecommand also work with <number>
e.g. side 0 // east
same as BIS_fnc_sideIDBIS_fnc_sideType then
I feel like it would be more useful to """just""" have commands with Side inputs accept a number as well. That would allow them to immediately accept config values without needing further interpretation. I suppose that's more work though.
(e.g. units 0)
side 0 == east; // true
side east == 1; // true
```nightmare π
π
Medical Taru has some unretexturable artifacts that can cause Z-fighting
setTerrainHeight and other JIP stuff unreliability example, boats going through the ground, scripting errors due to missing variables, etc.
Indeed happens on mission restart
(There should've been a river made with setTerrainHeight)
Talking about MP mission restart
setTerrainHeight is used during initialization, players randomly end up not receiving it and have to go back to lobby and join again
aaah, that indeed might be interesting. init order issueβ¦
Yeah looks like it gets overwritten somehow or doesn't arrive at all
my guess would be "server sends everything" "here's the JIP" "client connection are (re)initialised"
Yeah, perhaps client state flushes stuff from the server because it thinks its from a previous round?
I have a check that waits until public variables are not nil on clients before letting player into the game, and it still fails, as if variables arrive then get deleted after a bit
Spam https://feedback.bistudio.com/T173638 and more spam https://feedback.bistudio.com/T173666
I'm not an expert on BI FT practices, but is it correct for tickets to be Closed as Resolved before the change comes to stable branch? I thought the usual process was for implemented tickets to be Feedback when on dev branch, and Closed when it's confirmed working on stable, but recently there's been several closed before the change is even on the public dev branch version.
Just found this in the cient RPT.
2:28:27 CWeaponSlotsManager::ProcessRemovePendingObject: NULL PTR in magazineOwnerCont! Manager owner: [Exile_Unit_Player] - pending object: [5Rnd_127x108_APDS_Mag]. Maybe source container(vest, uniform, ...) was removed during reload phase or this could be a sign of a serious bug(probably something in MP)
Is context needed and maybe a repro ?
I'd prefer if we can label this a one-time fluke and ignore it
That's a rabbit hole that will go so deep that getting into it will melt your skin off
@sinful kettle https://feedback.bistudio.com/T172770 check if gud
Yeah thanks a lot! 
btw what if a mod is inside another mod?
I mean two cfgPatches in one pbo sorry 
CfgPatches doesn't matter
addons is per pbo, not per CfgPatches
Currently I'm detecting the matching mod, by comparing the prefix of the bank.
I think there should never be two pbo's with same prefix so sthat should be fine π€
Also this makes that command quite a bit more expensive.
The hash is actually read from the pbo on disk.
But I consider this to already be a expensive command that is rarely executed (really you'd only need to exec it once per game start)
Yeah exactly
Hmh nice feature to have
Btw I was wondering if there's a reliable way to tell which addon does a file belong to. e.g. if I do getModelInfo to get the path, I don't think there's a way to get the addon is there? Your best bet is getting the addon whose prefix is similar to that of the path
But then, I guess there could be cases where the addon prefixes overlap:
a3\data
a3
so now you can't tell which addon to pick for a path like:
a3\data\model.p3d 
Could that even happen?
Hm latest #biki_changelog reminds me of my wishlist: crossfade music
What's it going to do if there are two PBOs with the same prefix?
I would actually like to detect this case for idiot-prevention reasons.
What's it going to do if there are two PBOs with the same prefix?
Doesn't sound unlikely to me.
One of the pbo's wins and eradicates the other
Can what I said happen btw?
the overlap? I guess get the most specific one. Simplification: the longest one π
I don't want to research and answer your question. I think it probably can't
That's the idea right now
if same prefix is found, later loaded gets skipped
Revive system doesn't recognize Global Mobilization / other modded medical items
https://feedback.bistudio.com/T152608
@solid marten can this be fixed for 2.14 please - important for all CDLCs
ticket has code that should address the problem
Yeah, waiting for work approval
In the meantime you can add an example config for non-vanilla med items
+1 config issue, bell sounds from Enoch has a maximum hearing distance of 20m
https://feedback.bistudio.com/T173688
Sorry I know nothing about these values, perhaps @sterile nova can clarify?
are those for very small bells?
RingTheBell could be perhaps increased to 100 meters
but "small" and regular bells are for those located in the church towers
https://youtu.be/cYB5n2ZiWaM
idk is this a very small bell, but the max distance must not be 20m
sth like this?
class CfgPatches
{
class TEST_FAK_Subclass
{
units[] = {};
weapons[] = {};
requiredVersion = 0.1;
requiredAddons[] = {"A3_Data_F_Oldman_Loadorder"};
};
};
class cfgWeapons
{
class ItemCore;
class FirstAidKit: ItemCore
{
class ItemInfo;
};
class Medikit: ItemCore
{
class ItemInfo;
};
class TEST_Medikit: Medikit
{
scope = 1;
// scope = 2;
displayName = "TEST Medikit";
class ItemInfo: ItemInfo
{
allowedSlots[] = {801,701,901};
mass = 40;
};
};
class TEST_FirstAidKit: FirstAidKit
{
scope = 2;
displayName = "TEST FirstAidKit";
class ItemInfo: ItemInfo
{
mass = 8;
};
};
};```
yes thanks
Is the createMarker channel broken?
_marker = createMarker ["markername", player,currentChannel];
_marker setMarkerType "hd_dot";```This doesn't send the marker to MP clients
While without channel argument it does
Hm if currentChannel is 0 does work
π€When I want to specify the channel creator is also mandatory to have?
Also noticed that sometimes the markers don't respect the given channel
Despite I've told them to be make in Side channel some are in Global
check that you have side enabled
for "_xx" from 1 to 16 do {
for "_yy" from 1 to 16 do {
private _marker = createMarker [format ["%1/%2",_xx,_yy],getPosWorld player vectorAdd [_xx,_yy],currentChannel,player] ;
_marker setMarkerType "Mil_Dot" ;
} ;
} ;```Repro
I've figured out what was the issue: the slash in the name
It is because the USER_DEFINED markers use slash to store some data and looks the game confuses it
_USER_DEFINED #2/1/1" if I place one in Side channel, same thing is happening for createMarker'd markers
And was the issue in this context, because I generated name with hashValue that could return /
allMapMarkers should return like ["1/1","1/2","1/3","1/4"...] in this script, and if I try sqf allMapMarkers apply {markerChannel _x}does return only an array with full of 1 (or other channel ID)
So I'd assume markerChannel is overwritten with /1 thingy
Even though if I create those markers with 1 aka Side channel, it does respect the visibility for other players (Fig 1: BLUFOR client who made the markers, Fig 2: OPFOR client who can't see them)
And not to mention that if I create markers with 0 are visible for both clients
So, Dedmen, pls fix.
I'd expect it to use # as separator
but if it doesn't then, no fix because backwards compat
Well since the way to avoid is just not to use / so ez fix for my scripts too
just use \ instead
worth a biki note though @alpine tulip
I'll do it when I'm back and if I remember
https://community.bistudio.com/wiki/createMarker
Noted. Not sure my description made sense enough
Grrr another DX11 out of memory error, just by flying around in an empty map
Caused a hard driver crash
what GPU are you on?
1660 Ti
i'm on 1660S and don't get suchh issues. Maybe you're using broken mods or your GPU overheats
same with 1050 ti here
The 13 latest posts on the A3 section of the FT are all spam
Would be nice if we could block buttons via the MouseButtonDown EH like we can with the KeyDown EH 
@azure crane are you suggesting overheating GPUs cause out of memory errors? π
Has adding UDP hole punching been ruled out? I'm sick of trying to help people who "just want to play with friends".
For A3 there are no plans for that
does no plans mean not happening or just not on the list right now
I wonder how Steam Networking would work for arma
internet I'd assume
Not on the list
years ago I would've told you that setObjectScale and terrain editing were not happening
Can't say stuff like that anymore
suffering from success
DJ Khaled
the d stands for dedmen
I have found an issue with getUnitLoadout and a specific handgun from SOG:
player addWeapon "vn_m79_p";
player addWeaponItem ["vn_m79_p", ["1Rnd_HE_Grenade_shell", 1, "vn_m79_p_shot_muzzle"], true];
getUnitLoadout player select 2 // Returns `["vn_m79_p","","","",[],[],""]`, but magazine was added
The weapon has 2 muzzles, both support at least several of the same magazines.
I'm on the 2.12.150787 profiling branch.
You are adding a vanilla mag to a sog weapon. Can you check that this happens on a vanilla weapon and a vanilla mag?
I don't think there are any vanilla weapons that have two muzzles which share magazine types
Yeah. Funny enough, The Ace Arsenal will show that round, but the vanilla Arsenal will not
What NikkoJT said
The compatibleMagazines command shows that "1Rnd_HE_Grenade_shell" is compatible with the secondary muzzle of "vn_m79_p".
I was running CBA, ACE, ACE SOG compat and SOG. As far as I can tell, CBA nor ACE nor ACE SOG compat change anything related to "vn_m79_p", but I will test with SOG only tonight.
The ACE arsenal can't show secondary muzzle magazines, it can only show primary muzzle magazines for handguns and launchers
CBA adds a grenade compatibility in this case, though
by default the second muzzle only takes buckshot rounds
Found an interesting bug:
a addEventHandler ["Deleted", {
params ["_obj"];
systemChat format ["removed %1", getObjectID (_obj)]
}];
addMissionEventHandler ["EntityCreated", {
params ["_obj"];
systemChat format ["added %1", getObjectID (_obj)]
}];
so let's say a was a unit in vehicle. when you delete the vehicle (I did it thru Zeus), first the deleted EH triggers, then the entityCreated one (with the same unit) 
what if you add deleted inside created 
what's the difference? 
also I don't understand why every time I delete a unit (on foot) something else is created immediately after 
one possible explanation would be game trying to create the unit outside of the vehicle when it's gone, i suppose?
I wonder if it will recurse and the game will die
well no it won't
actually I noticed this due to a leak in my "engine" which registers units using the EntityCreated EH and adds the deleted EH to them there
I noticed crew units are never deleted 
at first I though deleted doesn't trigger for units in vehicles
turns out it does get triggered but then EntityCreated adds them back in...
i mean, it equally fires "EntityCreated" when i tell a unit to dismount the vehicle 
well, is the crew member inside the vehicle a real "entity", or just a partially controlled proxy that's still a part of the vehicle?
Same still happens with just SOG loaded:
player addWeapon "vn_m79_p";
player addWeaponItem ["vn_m79_p", ["vn_40mm_m576_buck_mag", 1, "vn_m79_p_shot_muzzle"], true];
getUnitLoadout player select 2 // Returns `["vn_m79_p","","","",[],[],""]`, but magazine was added
So the loadouts array doesn't support secondary muzzle at all?
But it works with rifles that have grenade launcher, and the promet right?
I haven't had any issues with primary (or even launcher) weapons with secondary muzzles. It's only handguns it seems.
I thought at first it might have to do with the fact that both muzzles have multiple common magazines, but if only SOG is loaded, "vn_m79_p" does not share any magazines between both muzzles and the problem persists.
Unsure about binoculars having this issue though, haven't come across any that have a secondary muzzle.
Ahhh pistols! That makes sense thanks
west setFriend [west, 0]; -> Cant open Vehicle Inventory from inside anymore, cant get inside a vehicle with someone else sitting inside.Makes sense because of the Side not friendly but then please make a workaround for the suppressed event handler
"anymore"? thought this was a thing from the very beginning. And I don't see how is this an issue tbh since you make a side hostile to its own side. Wouldn't using fired near eh be more convenient?
FiredNear is limited to ~69m max, so not very usefull (distance in meters between the unit and firer (max. distance ~69m))
oh nvm, the biki says "Can be made to trigger for the same side if the side is set as enemy to itself" so it's a bug indeed
Not really seeing it.
Even if you set the radius larger than 69m, does suppression ever trigger beyond that?
I figured that'd be an engine maximum.
Is game should trigger different sounds of a grenade hitting a surface at all?
Maybe it's were WIP but not completed and issue from ticket https://feedback.bistudio.com/T165776 is "as designed"?
Debris sound set of explosions not audible due config issue
https://feedback.bistudio.com/T173899
Btw can someone please take a look regarding the magazine deletion bug ? Its still a thing these days and it happens 0.5 seconds after the reload is done.
Here is a video from 4 years ago : https://www.youtube.com/watch?v=edpCLwFvtdI
2:40 video for a reload bug
plz provide vanilla repro
From what I know is that it happens after you shot and then reload, probably happens more often when you reload immediately after you shot.
Also I noticed that the magazine which came out of the weapon and disappeared , could come back into your inventory 8 seconds after .
This bug probably happens only on a multiplayer environment.
plz provide vanilla repro
Yea it also happens on vanilla multiplayer server. Shoot and then immediately reload, sometimes it gets removed and sometimes not.
You could atleast say where in the video the bug appears.. I'm not gonna watch 3 minutes very closely to try to find where something might be wrong
You are right, it happens the first two times he is reloading
It also happens while looting, also had in different game modes (Exile, KotH etc).
https://gyazo.com/18f902af3a44e62b1f731b70ae0000d7
If you look at the ammo coutner lower right corner you see the mag disappearing when i switch my backpack.
So now you want me to watch the video and wait till he reloads
Still, vanilla repro please
Need repro, cannot diagnose a random rare event
According to the magazine count on the top right, the maggazine appears for half a second and then vanishes. Happens at 0:27 and 0:40 in the video
Are you sure this is not a mission issue?
Im 100% sure that its not.
I'm 100% sure it's mission/mod related, because both the inventory system and ammo counter are not vanilla...
So the problem is in there, and not in Arma itself
It happens in every scenario / mod and for sure also in arma vinalla multiplayer without any scripts. There are just no people playing vanilla arma multiplayer without any scripts running π
Just gather 5 people or more on a vanilla arma server without any scripts running and then try it out.
I dont have the people who could do that
135.181.74.185:2312 - I set up a server, no mods, no scripts running, just 10 units placed. Hop in and test it with me
Ok tested it with one more guy, which probably dosent matter at this point. And I repro it first try. Will upload video
Here we go the video
I really can't reproduce that π€·ββοΈ
Completely vanilla Arma 3, single unit, both in SP and MP
Do you have a .rpt from both client and server?
Also got it on video, at the start 7 mags, i shoot once and then reload and i only got 6 mags.
https://www.youtube.com/watch?v=ucpJ-SA4dBc
dedi or player-hosted?
I'd think this could happen on dedi only
currently player hosted, will try dedi as well
https://www.file-upload.net/download-15167204/Arma3_x64_2023-07-13_14-29-08.rpt.html - Client rpt (the <lineIntersectsWith [ATLtoASL(player mode> error was due to opening the debug console)
https://www.file-upload.net/download-15167205/arma3server_x64_2023-07-13_14-23-14.rpt.html - Server rpt
Because (looking at their videos) it seems that non-full magazines are removed on reload
Yes, its the one which comes out of the weapong after reloading and shooting
can't download from there; got a big red screen in my browser due to possible malware
lol
ok let me upload it on google drive
https://pastebin.com/Dw0ZLwVt - Client Rpt
https://pastebin.com/qrsY86Mx - Server rpt
https://drive.google.com/file/d/1C-go4xqnqw13ofqoA-WijKMY_o3LINpS/view?usp=drive_link here is also the mission I had loading
Why your client rpt has a lot of scripting error spam?
Like I said , it was due to the debug console
Ah I've missed your last
Still can't reproduce on a dedi server (running on local server with FASTER), my mags will just reappear in my inventory, just as unique items due to the fact that they have less ammo in them
135.181.74.185:2312
try to connect here and lets see if you can repro it there
or give me the ip of your server and I connect there
on your server I do get the issue, but very inconsistent...
Im pretty sure it has something to do with the netcode since it gives you back a item
I know there was an issue with it 10-11 years ago (happened in SP as well), but haven't seen it since then even on modded servers.
But yes I can confirm the issue, but is hard to replicate consistently
Similar/same behavior to this bug i recorded on Exile.
https://gyazo.com/18f902af3a44e62b1f731b70ae0000d7
So should be clear it's not a mod/mission problem π°
So, I think I've just realized that there is no getter for
- Mouse is shown (display/dialog)
- Actual controlling unit (I know Zeus' remote control has a getter but more generic way)
- The game is controlling an unit right now (slightly overwrapping with Display idea but let's say WASD does character movement)
What do you think? "Ticket please"?
Also if a camera object is taking main screen, I think there's no getter too
cameraOn doesn't do it IIRC
I need this
If we get such commands, would be good to get a "BIS_fnc_currentPlayer" function too
Or "currentControl"?
I thought we already had a ticket for multi-mag addMagazineTurret, but apparently not π€·
https://feedback.bistudio.com/T174007
BTW: I do think this is a known bug, but sometimes it fails to remove a part of fleets when it supposed to be removed. Do we know why it happens?
Is that a different "object", ship ladders act like this
What do you mean? I'm aware those USSes are made of multiple objects
Did you move the ship in Zues?
Nope. The above pic is just done by placing multiple USS Freedoms and create a new scenario (Ctrl+N)
It should be very easy to reproduce
It seems it is not even an object
It looks like a relative child object is being lost in the entity management system
allMissionObjects ""```doesn't detect it
can you walk on it?
I do not think so. I haven't tested it yet but Editor camera doesn't consider it a solid object
does it disappear when you look away from it and look back?
it might just be an errant graphics bug
Nope as far as I tested
CursorObject and or TargetObject?
objNull. I've just noticed that there is no shadow too
I've had the same issue before too, never found a fix for it
or even a consistent reproduction 
yeah it's just the model
Bad lod?
I think the game removes the "object" from the quadtree but doesn't remove the actual object
so the object reference is probably not deleted 
what happens if you move to another part of the map then come back? in that case the object should not render anymore
Ah, that makes sense
@sinful kettle
it still renders
Same
I guess the game might have some kind of leak then... 
I do feel there is some kind of iffy with the model itself
for "_x" from 0 to 15 do {
for "_y" from 0 to 15 do {
createVehicle ["Land_Carrier_01_hull_02_F",[_x*30,_y*30,0],[],0,"CAN_COLLIDE"] ;
} ;
} ;```
Wait wtf I've just Alt+Tab back and it disappears
I remember Dedmen saying the game clears graphics cache on alt-tab, so this smells like the model not being cleaned up on deletion, for this specific object for some reason
i didnt have it disappear on alt+tab iirc 
Yeah, It's keeping the draw calls
then again im suffering from pretty bad memory loss at the moment so i could just be wrong if killer has the same behaviour as polpox
for "_x" from 0 to 25 do {
for "_y" from 0 to 25 do {
createSimpleObject ["\A3\Boat_F_Jets\Carrier_01\Carrier_01_hull_02_F.p3d",[_x*30,_y*30,0]] ;
} ;
} ;```It doesn't happen
that's a super simple
Yeah
can you try with just simple?
Oh boi
Just in case let me relaunch A3
also, try. Win + Ctrl + Shift + B
Restarts the Graphics Driver, that should remove ever draw call
what simulation does that object use? house?
configFile >> "CfgVehicles" >> "Land_Carrier_01_hull_base_F" >> "simulation" is "house"
Holy Geez I've just found out the crucial thing
featureType = 2; is the faulty I guess
for "_x" from 0 to 15 do {
for "_y" from 0 to 15 do {
createVehicle ["Land_spp_Tower_F",[_x*30,_y*30,0],[],0,"CAN_COLLIDE"] ;
} ;
} ;```
This could mean... let me check
for "_x" from 0 to 15 do {
for "_y" from 0 to 15 do {
createVehicle ["B_MRAP_01_F",[_x*10,_y*10,0],[],0,"CAN_COLLIDE"] setFeatureType 2
} ;
} ;```Okay this is 100% repro
I mean not 100% but you know
« 50% of the time, it works everytime! »
Very strange π
featureType 1 doesn't do it it seems?
Cannot repro with 1
for "_x" from 0 to 15 do {
for "_y" from 0 to 15 do {
deleteVehicle (createVehicle ["Land_spp_Tower_F",[_x*30,_y*30,0],[],0,"CAN_COLLIDE"]) ;
} ;
} ;```This doesn't make any artifacts. So I'd assume deleting an object with some special case would cause it
https://feedback.bistudio.com/T174009
Okay ticket
It was very satisfying to bughunt π
https://feedback.bistudio.com/T174010
https://feedback.bistudio.com/T174011
https://feedback.bistudio.com/T174012
https://feedback.bistudio.com/T174013
Unrelated: I've spammed the feature requests π
burger(ja)pan POLPOX
panic
PS. flush cheat also removes the artifact
I said file cache
Features are actually rendered twice. And stored in two render lists.
It'll be failing to remove it from the secondary list
But if the entity gets flagged for deletion, it should delete itself from the list :/
Maybe only in eden?
// 3DEN can keep deleted objects in history, therefor Link is still valid even object is not in world
Though eden shouldn't do that with dynamic spawned objects
Visual and sound effects (fn_planeEjectionFX.sqf) in Ejection System functions work not correct in MP
https://feedback.bistudio.com/T174022
bumpity bump for an EntityDeleted mission EH
https://feedback.bistudio.com/T172847
#arma3_scripting message
And EntityCreated bug when you disembark π
So there's a thing with missile AI units that they tend to spam rockets or missiles very quickly. Be this RPG units or vehicles shooting missiles. Infantry AI for example is able to target, shoot, reload, re-acquire and shoot AT rockets quicker than a human player could.
I imagine different people, servers and communities will have different preferences when it comes to this. So could we get class or weapon family specific AI skill params? This would allow people to say, make rockets and missiles less spammy or more inaccurate, etc, according to their preferences.
(any other solutions welcome)
There is https://community.bistudio.com/wiki/setSkill, have you played with that yet?
Oh, yeah, have
Guess the class filtering can be done manually
Although with vehicles it tends to be somewhat...weird
Also if a unit has both missiles and guns, and you want to modify skill with just the missile/rocket launcher, then setSkill doesn't help
But maybe it's okay if the odd AT guy is just a bad shot
hello, when loading the altis map, the game crashes and gives an error Warning: prenlod format in object a3\structures_f\ind\windpowerplant\wpp_turbine_v2_f.p3d
Please help me
steam verify Arma 3
I installed the game 1 hour ago
What, and you think that makes it less likely that you have corrupted files?
Maybe
It does make it a bit less likely overall, but not impossible - and that's the most likely cause of that error
Thx ,it help
Depends if it worked fine an hour ago :P
If you just downloaded the thing then the absolute peak failure time is the first execution.
Files "rotting" later is extremely unlikely, and tends to break in a different way (CRC errors)
Hence the second most likely time to break is after an update. Well, maybe more likely if steam is bugged.
Oh yeah, wrong channel. This stuff should go in #arma3_troubleshooting
This channel is not #arma3_troubleshooting
For troubleshooting go to there
Oh I see
!issuewarning 544967184766730240 crossposting, please re-read our #rules
Done.
I suggest to do just that
No sounds and particles on V-44 X Blackfish (Armed) pilot Manual fire for other players
https://feedback.bistudio.com/T174076
Terribly sorry for necroposting. Regarded to the retexturable weapon accessories. Besides the code, does it really add a big rendering cost? I really, really and really doubt a 2020 ish computer can suffer because of it when there is a lot of 2k to 4k textures and retexturable vehicles all around the place
Its a small cost, on every weapon
Probably not too bad
@daring wagon when you have time, can you confirm that this bug is gone? https://feedback.bistudio.com/T119171
@solid marten please also mark spam as resolved, so we have this link's visual difference
People/modders will prefer to have retexturable assets/accessories. Since I am no developer or CG geek, I really don't know how big it is but it definitely is worth it, if is affordable
I closed and disabled about a hundred of them it already took too long, no time for fiddling with the tickets beyond that
I would need a FT ticket, including a repro mod that should work so i can test it with
https://feedback.bistudio.com/T168410
Let me make a small Mod, I think there is a bipod that have camo selection but never does it
Okay \A3\Weapons_F\Acc\accu_bipod_01_snd_F.p3d does have camo_1 selection that covers the entire model. This should do
Do I need to link this to FT too?
I will try and see if we can have a button that does it all
yes put all on FT ticket, that's what I will see when I get to it
Done
Will do
Are Eden editor module related requests still possible?
For 2.14?
You probably aren't gonna get it for 2.14 if it's a very big request. You can always ask and then wait and see, if it's possible.
No I meant in general, 2.14, 2.16 whatever
You can suggest whatsoever
dont onow how feasible but diag command for ping to current server would be nice
elaborate more please?
normally you'd use
getUserInfo
That's a server exec command, which makes it a bit awkward to use for clientside monitoring
Yeah just didn't want it to be met with "were not developing Eden anymore sorry"
https://feedback.bistudio.com/T174149
Swivel target animations work not correct in MP mission, but if object has enabled "local only" checkbox the animations will work correct, but object will local
ChatGPT spambot? https://feedback.bistudio.com/T123627#2475523
I was going to create a ticket asking for the minefield module to support the Area widget, but seems there's already a ticket for it https://feedback.bistudio.com/T155463 -- it suggests other modules too
yep - deleted, thonks
So my two eden requests are:
- Add area scaling widget to minefield module https://feedback.bistudio.com/T155463 (2020)
- Add a vanilla garrison module https://feedback.bistudio.com/T85067 (2015)
unsure whether it was from the very beginning or not but when directSay is used in vehicle on AI and player-passenger is in 1st person, the sound immediately starts fading out to "outside" of the vehicle. If you switch to 3rd person you can hear it again and again it fades out to "inside". Why does it have to be like that?
Is there any possibility of angular velocity commands this late in development?
if so, would like to point towards the several requests: https://feedback.bistudio.com/T73570
https://feedback.bistudio.com/T160321
https://feedback.bistudio.com/T169008
π
Probably addForce/addTorque?
as far as i can see no getter for torque? π€
Maybe π€
Getter/setter for torque at the least would be useful rather than having to bodge together addtorque etc π€·
tho would prefer dedicated angular velocity commands for engine interpolation, addtorque has no interpolation
Could anyone from the development team please take a look at my post on the Forum with details regarding the issues I've been having launching the Arma 3 Publisher, I raised a ticket on the FT last week but so far hasn't been addressed. Raised a Steam Support ticket that include the logs they asked for and their reply was basically "not our problem, contact BI"
System.OutOfMemoryException: Insufficient memory to continue the execution of the program.
π€The whole crash callstack is completely microsofts code. Don't know what I shall do about that
The SteamInitializationFailed part is weird but.. It doesn't say why it failed, and I don't understand why it would considering that's the first time i see this problem
There were issues if you had too many mods, inside Arma Launcher.
Steam recently changed their rate limits without telling anyone. But I think that limit was higher than just 125. And that also shouldn't fail the Steam connection
That SteamAPI_Init failed error is straight from steam.
It's literally
bool success = SteamAPI_Init();
if (!success)
Print("SteamAPI_Init failed. Unable to initialize SteamWorksSDK.")
That is quite a bit before we even start processing your mods.
There is literally none of our code inbetween there, we call a standard Steam SDK method, and it returns false.
That must be on steam side
there is a way to launch steam in debug mode with a console
https://partner.steamgames.com/doc/sdk/api/debugging
You'd have to do that, and then look in steams console.
use the set_spew_level to up the log level until something pops up
Thanks for taking the time to look into this man, much appreciated, I'll follow your lead. π
These are the logs from the steam console using set_spew_level, I launch the Tools, then try and open the Publisher and it just doesn't open.
These are the latest Tools log from the same session.
Unable to retrieve all process info: Access is denied
Well that's weird...
Steam log is basically empty, maybe higher spew_level? 4 4 is max
I'd expect it to be spamming.
But in case where Steam API really cannot connect to Steam client, it would probably indeed stay empty. But then its Steam's fault.
But if they say its not their problem 
I don't know if there is also steam debugging on the publisher side
Ah you also have to run log_ipc publisher to see all the communications between steam and publisher
Then a publisher start looks like this
steam saves that in Program Files/Steam/logs/ipc_SteamClient.log and console.log
Spam comments: https://feedback.bistudio.com/T172048
https://community.bistudio.com/wiki/addPrimaryWeaponItem
only returns false if item cannot be added but is compatible, else fails silently, also returns nothing if is added -- true for success (or expanding false for icompatible) would b useful, any interest in ticket?
I love spam
i saw that too, went and opened a bottle
https://feedback.bistudio.com/T174259
With Spearhead also approuching custom menus. The CDLCs now start to conflict with each other when all enabled. This includes duplicate menu items, and not being able to access the mission editor when all are loaded.
A launcher and/or startup param option to force default menu might be a nice solution
Agreed, or maybe all the CDLCs just decide who gets prioritized when all loading in thier menus.
Or maybe they should all just not modify the main menu...
This is probably the better solution though, retains player choice and creator choice
Which CDLCs conflict?
Load all of them
You will get something like this.
The Spearhead will have a CSLA label, and go to the Spearhead's credits. There will be a blank entry for were possibly the editor is, etc.
This can also happen with reducing to CSLA, SOG, and Spearhead. As they are the CDLCS that have custom menus
@plush brook This is with vanilla menu enabled?
Ok. If I disable the custom menu I get the same
the guy legit wrote "Hello there" as new message with the second account (3 days apart)
shift+t Tactical Ping sound effect on Radio channel (not for Effects channel), this is as designed?
Fake account - https://feedback.bistudio.com/p/whatsappaero/
@untold sky Requested FT for getOrDefaultCall: https://feedback.bistudio.com/T174379
did you mean to ask in #arma3_scripting ?
Yep, I'll move it.
found a bug that has been around since last patch.
placing a player down and then a group down in editor with a placement radius for the group causes game to crash with this error message and no mods when you play the mission from editor, sp or mp
can confirm, instant crash with any type of group and any radius given other than 0
Is this on profiling? There was a change related to placement radius iirc
Yup known, already fixed on profiling branch
Btw Placements Radius on groups never worked
it did work for me for spawning in groups
On units/objects it works.
But on groups it only works on the last placed group
I don't know if you guys like this idea, but i think it would help especially people new to arma multiplayer scripting to make their mission work in MP: https://feedback.bistudio.com/T174485
or would that be too much work if you have to add new code to every command?
That would probably result in a lot of script errors from inefficient-but-fully-functional global remoteExecs
Not super backwards-compatible
It's a good idea, just not for this late in games life imo
no, but passing invalid args means the code is already broken
so i guess it wouldnt cause harm
If an LA command is remoteExec'd to all machines, it is "broken" on the machines where the argument is not local, but still works overall because the machine where it is local also received it.
That's not ideal for efficiency, but the code will work - and it will also cause a script error under your proposal.
I wouldn't be surprised to find a lot of existing code that does things like that.
ok wasnt thinking about remoteExec, not sure how all that works π
actually reread your msg and I see what you mean. but error/log message itself wont hurt (imo)
The problem is that at this point in the game's life cycle, there is a lot of code out there that's probably never going to be updated (for example, because the authors have gone away).
If that starts generating script errors (possibly a lot of script errors) despite actually working fine, the people who use it are going to get very annoyed.
true..
we dont have any stats how much loggin that would cause, would be interesting to know
not happening
too much work huh?
dangerous as it will cause rpt/error spam for no reason
- tonnes of work for microscopic benefits
so, beginners π
reading wiki can help beginners for sure
it would help, but its too late in the games life to be added as all the currently broken code that looks like it works wouldnt be fixed and you end up with rpt spam
if they find page about locality π
ive never read the wiki page about locality and understand it
you can chose to let any argument through though by executing globally, only one client will do it the rest silently fail. thatβs legit way of doing thing, no need for error spam on all clients but 1
i didnt even first know what the locality icons meant in wiki page, fortunately I was making SP mission back then ;D
executing code that silently fails isnt broken code? π€
of course not, if there is no error it is not broken
think we probably have differing opinions of what constitutes broken code
It's a "safe" break, and if it's caused by an untargeted broadcast then it's not truly broken - the overall objective will be achieved because it succeeded on the one client that did need it.
still executing code that doesnt have the desired outcome though 
the engine has many global broadcasts internally where only one client takes notice of
in arma, A lot of things "fail" silently
If I fire 12 MLRS rockets into an area and only one hits the enemy position, the other 11 were not broken
ive not written sqf that i can remember thats failed silently π€·
We could add local arg check only if you put -debug parameter. To test your own stuff.
That deletes the backwards compat concerns, as only developers turn that on while working on their own code.
But still alot of code we would have to update
sometimes it is faster and more efficient to spam all rather than trying to target one
That should've been done back with the waitUntil thing, that made returning nil into an error and F'ed up a ton of stuff.
But we didn't have -debug parameter back then
if you have like ```sqf
alive _man
no we canβt it means manually going through every command, because locality is not a property or type like argument type
going through every command is not impossible. But we won't do that.
We may also not need to do every command, maybe just a few important ones
for 3 people that even know or use -debug?
tbf most beginners arent using -debug to start with
launch the Community Improvement Project and make GC8 go through all the commands
great I get to see the engine code...
you canβt unsee it
im under the impression its a lot of spaghetti
the inventory systems code was described to me as trying to eat spaghetti through a straw but youre lactose intolerant
If you are not a developer, you don't care about developer warning messages. So.. yes.
cost <=> benefit
Some "local-argument" commands have a useful effect even on non-local objects, like setMass.
Although in that case, mass changes should really be transmitted a bit faster :P
I was told that this would be the place to ask if someone could look at potential implementation of a proposed script command. Request is about a year old, hence asking.
https://feedback.bistudio.com/T167029
@gaunt depot About uav terminal disconnect, should there be any restrictions on who can do it?
I guess it is unnecessary since you can remoteControl uav unit directly anyway
Since its a scripting command, I'd say no restrictions, the script will do needed condition checks itself before using the command.
Sounds doable.
That ticket was either not seen or ignored as it has no internal ticket linked.
I'll put it on the list for 2.16
Appreciated.
@gray wharf howβs that button coming? either that or we need more ppl to remove spam as it becomes quite a nuisance
I bumped, let's see
Made a bug report about plane auto landing to wrong airfield https://feedback.bistudio.com/T174528
Whoops! I lost my 2FA code for the Feedback Tracker when switching my mobile phone. Is there anything I can do? (Noob mistake, I know) π

Was going to report a bug about Reforger constantly switching to Windowed mode (from the Borderless option)
has anyone else had an issue where despite having an action on vehicle to get inside (any seat), clicking it would not move you inside? Selecting from menu or default button, the option is there but it just doesn't fire. I have noticed it 3-4 times during past two weeks so it's inconsistent and I wasn't able to reproduce it yet, but perhaps there are more people who encountered that
moveInCargo lets you inside such vehicle
this is taken from Copperhead mission of Green Draw campaign, it did not occur once I reverted this scenario. Thought that it may be caused by not being a leader/being in a squad but seems to not be a case here.
there is also similar issue where AI will not mount the vehicle if you, a squad leader, is inside so perhaps its somehow connected.
inb4 post in trembleshooting - this channel is blessed with presence of green people who will say "won't fix" and I am about to post a ticket that has no steps to reproduce, so at least I could have the angry mob on my side
running [player] allowGetIn true then [player] allowGetIn false seems to fix this 
Apparently there's no way to properly clear waypoints aside from changing group, for example a heli pilot will return to a deleted waypoint if it wasn't completed. Worth making a ticket or is that too "not touching the AI"?
I'd say worth making a ticket, the AI issues can be infuriating
and the main thing that has discouraged me from making missions in the past few years
Not even doStop?
Always worth a ticket, it's not exactly related to AI behavior so might get fixed. Might be something just not being deleted properly.
I'd guess doStop would make them ignore the waypoint until the next order, but it wouldn't clear them.
hmm, I wonder if it's not really waypoints but individual vs group target position.
doMove being a unit order, then returning to group target position after completion.
Ah yeah, it's basically this note:
Deleting a group's current waypoint will not stop the group on its tracks.
You have to give the group a new move order or waypoint to change the group order.
No you don't "have" to do that
What's the alternative?
Well yeah but I meant after deleting the waypoints
If you give them a doFollow they're still gonna head to the last waypoint, right?
If the group has more than 1 member
doStop _unit;
[_unit] joinSilent group _unit
I guess
Yeah that was my first "solution".
Not sure
Best I found was that giving a move order will skip the current waypoint on termination, so you at least don't have to give a waypoint to stop a waypoint.
still no sus link, buuut I'll keep the page opened and refresh from time to time π
although I removed it altogether
https://feedback.bistudio.com/T174468#2483341 another one, π₯ it
also spam: https://feedback.bistudio.com/T174604
