#arma3_feedback_tracker
1 messages · Page 15 of 1
could you add this to the ticket too please will help to look in the right place when i start it
Done,
C_Hatchback_01_F:
["B Alpha 1-2:2 (B_Soldier_F)","",0.494967,"1e36be98080# 1793921: hatchback_01_f.p3d (C_Hatchback_01_F)","",-1,"<NULL-object> ()","",false,0]
Land_ToiletBox_F:
["B Alpha 1-2:2 (B_Soldier_F)","",0.494967,"B Alpha 1-2:2 (B_Soldier_F)","",-1,"<NULL-object> ()","",false,0]
with same repro
same amount of damage interestingly
New version probably fixes our busted code tbh :P
Amazing 😄 thank you for investigating! It does sound cursed lol
…from OFP era?
isnt the artillery computer OA?
that or A3 yeah, definitely not OFP
and the first vanilla artillery pieces were Arma 1 afaik
a2:oa
https://feedback.bistudio.com/T79971
found a task in 2014 about multiplier = 3 in CfgWeapons, the original answer is
This is as designed, Your note has been taken into consideration, thanks a lot.
so after 10 years, is that considered as a bug now in 2024?
That's probably an optimisation thing.
The notional rate of fire for miniguns is higher than what the game can consistently support, so instead of trying to simulate the actual rate of fire, they fire "3-for-1" bullets that consume more ammo and have a small amount of splash damage.
is there any chance that one of these could be swapped so theyre the same way around for both? i know its really minor it just really bugs me
I remember looking at some command that was added in OFP. Probably mixed it up with the dialog command...
currentWeaponTurret was only added in 1.12, that would have made getArtilleryComputerSettings obsolete.
😾
So annoying that "Links" was put on the Rechts side
i love you
Thanks for explaining that ❤️
Any chance there's been progress on this suggestion? It's been a while since the internal ticket was created, and there're a few things I'm looking to accomplish that this will allow
https://feedback.bistudio.com/T177742
(There are a few that this one was tied to at some point:
https://feedback.bistudio.com/T151696
https://feedback.bistudio.com/T151698
https://feedback.bistudio.com/T178384)
ENTITY = createShot [CLASS, POSITION, VELOCITY, VECTORDIR, VEHICLE, INSTIGATOR, REAL, TRACER]
If you create an entity and delete it right away, EntityCreated still fires and if you add Deleted EH there, it won't fire
Makes me wish for BOOL = isDeleted ENTITY so I can skip doing stuff with the entity in EntityCreated if its gonna be deleted this frame as I can't do that with Deleted EH
Can't not delete in a same frame, action DisAssemble does that
EntityDeleted is 2.18 away and way too broad
Any chance to get this annoying bug fixed (2nd one mentioned in the original report)?
https://feedback.bistudio.com/T74244
You could solve it with
https://community.bistudio.com/wiki/selectThrowable
That's workaround, not a fix.
There are other workarounds that don't require 2.18, however it would be neat not needing to handle throwables specially because of said bug.
Yeah I guess its not really a fix
Tail sound glitch if Mk20 and Katiba shoots at the same time, seems due strange volumeFactor and frequencyRandomizer values relative to other classes for rifles tail sounds
https://feedback.bistudio.com/T178138#2598736
nice catch!
There's a small gap in our ability to reverse-engineer attachTos. https://feedback.bistudio.com/T180786
Mic setting slider seems won't work, change value does not affect to volume, if set to 0 player still can be heard by others
purged!
yäy
(and some others while I was at it)
yea, i bet there have been more but that was the only one that catched my attention ^^
idk where to ask this the best, but is there a known issue regarding a3 film grain, AMD GPU's and black pixels?
I would say update your drivers etc, then if it keeps happening search the Feedback Tracker, and if it doesn't exist you can create a ticket w/ repro
yea.. its not on my pc but we've been able to reproduce it on my AMD GPU's friends machine. im making the ticket rn.
is it possible for something like a lod switch factor setting/command? would be very nice for screenshots as well as lowering lods in air etc
ft ticket https://feedback.bistudio.com/T180798
test on hosted MP and dedi 2 clients, no voice change if mic sensitivity set to any value
BI has LOD forcing in their internal binary. Would be nice, at least for diag exe
yeah would be really nice to have
cant use some mimallocs when taking screenshots because of how aggressively they assign lods etc
Tried UAVs on a live server:
21:30:46 Observer B Bravo 3-1:1 (chris) REMOTE in cargo of B Bravo 3-1:1 (chris) REMOTE; message was repeated in last 60 sec: 1238
```frequent spam
Some locality mess
Taking control of drone pilot ends up not transfering drone locality itself
Thanks Arma
Showing a classnames in this error message would've been helpful
as that B Bravo 3-1:1 (chris) can mean: player, vehicle, uav controlled by player
Somehow people end up with both "To gunner seat" and "Take UAV turret controls"
Which locks the game up (no more scroll wheel menu to leave it)
wait what, mimalloc affects lodding? 
yeah
using the jbgold one or whatever its called makes screenshot taking impossible
🤷
i just know i cant use it when im taking screenshots because i usually go back and zoom in
swapping alloc makes it fine
Amazing
LOL you indeed can disassemble other vehicles as drones
Matrix is not valid as physx transform! Object: 1ef76ad8b80# 1724290: dummyweapon.p3d REMOTE, WeaponHolderSimulated, Remote
Get matrix of physx actor is not valid! Object: 1ef76ad8b80# 1724290: dummyweapon.p3d REMOTE, WeaponHolderSimulated, Remote
it this new/something of concern?
be worried when they start assembling each other.
@solid marten Can I please get some engine insight for exact conditions for UAV Waypoint editing to work? The menu that appears when you right click on the waypoint in AV Terminal
I think it could be https://community.bistudio.com/wiki/isAutonomous check but probably certain checkAIFeature checks?
The whole system is really strange, if you disable only some AI features like targeting and shooting but leave movement and pathing, you can add waypoints but can't edit them
Probably not fixable through script, but maybe if I knew exactly what are the requirements, I could make a crutch?
Oh I found a hack how to make the engine let you use this menu with drone as not autonomous
findDisplay 160 displayCtrl 51 ctrlAddEventHandler ["MouseButtonUp", {
uav setAutonomous true;
onEachFrame {uav setAutonomous false; onEachFrame {}};
}];
Imagine doing anything properly in Arma 
But its a solution to this issue/bug
if you do ```sqf
#define TEST_DEF true
#if TEST_DEF
systemchat "TEST";
#endif
because of the "true"
I'm pretty sure #if wants 1 and not true
yes I just wanted to point out that the error message it self is bizarre
thats arma standard
Please ask me again towards the end of the week
Ownership transfer of the vehicle takes unreasonbly long time, why can't we force it with setOwner once group transfer is complete? Took 13 seconds in this clip. First you control the driver, then later group transfers, then later the UAV itself.
Huh, looks like ownership order is undefined as well, this attempt I got vehicle first and the rest of the group 10 seconds later
Hint shows locality of group, vehicle and each UAV unit in the video
Its like there is regular ownership transfer check each X seconds or something
Sometimes it transfers quicker, sometimes slower
If we can't force switch this locality, maybe somehow speed up this engine transfer?
uav setOwnerTransferPriority true or something?
Or maybe the engine should prioritize player controlled ownership transfer?
I think it does for player driver already?
Some mess with UAVs on perf build, UAVControl is very desyncronized
We really need a way to hide engine actions
Having UAV terminal pollutes action menu with so much crap
Open UAV terminal, disconnect terminal, Take UAV controls, Take UAV gunner controls
When all this should be through the terminal instead
👆
player disableEngineAction "UAVTerminalReleaseConnection";
player disableEngineAction "SwitchToUAVDriver";
player disableEngineAction "SwitchToUAVGunner";
I know actions are contextual and it can be very complicated to add checks everywhere, but perhaps it can be simple flip to show/hide the action from UI. Leave it for AI command even.
So simplest form could be something like
"UAVTerminalReleaseConnection" inGameUIEnableAction false;
"SwitchToUAVDriver" inGameUIEnableAction false;
"SwitchToUAVGunner" inGameUIEnableAction false;
OR redirect it into an event handler:
addMissionEventHandler ["HandleEngineActionShow", {
params ["_action", "_unit", "_target"];
_action in ["UAVTerminalReleaseConnection", "SwitchToUAVDriver", "SwitchToUAVGunner"]; // true for handled
}];
Or in inGameUISetEventHandler but that thing sucks
I would say just tell the encountered issue, don't explain a potential solution - you may keep hair this way 😄
Sorry for being annoying, I'll move this into a ticket comment.
made a ticket about the collision bug we found earlier: https://feedback.bistudio.com/T180863
there is a bug if you create Post Process Effects then go to video options and change the PIP settings to anything after setting the Post Process Effects it will destroy Post Process Effects
example:
"ColorInversion" ppEffectAdjust [0.5, 0.5, 0.5];
"ColorInversion" ppEffectEnable true;
"ColorInversion" ppEffectCommit 0;
make a tikit if it doesn't exist already!
Does this event handler trigger? https://community.bistudio.com/wiki/Arma_3:_Event_Handlers#PostReset
@gray wharf btw why does it say obsolete?
tl;dr:
- it's an object EH instead of a mission EH for some reason
- it doesn't work properly
- it was only implemented in DX9
I had it working long time ago pretty sure of that
I wrote the Obsolete part, but I have no idea/memory why
https://community.bistudio.com/wiki?title=Arma_3:_Event_Handlers&diff=next&oldid=337328
so yes #community_wiki message
thanks for the dig o7
nukeded
The error formatting is wrong yeah.
It should be more like
"Preprocessor failed on file C:\Users\user\Documents\Arma 3\missions\garbage.VR\init.sqf - error 11 (source (empty filename), line 3)."
I'll fix it.
Btw do we document these error codes anywhere?
NoError,
StreamOpenError,
IncludeError,
IncludeMaxRecursion,
DefineError,
DefineParamError,
ParseExit,
InvalidPreprocessorCommand,
UnexceptedEndOfFile,
ToManyParameters,
ToFewParameters,
UnexceptedSymbol = 11,
EndIfExcepted,
Might be smart if we would print them as string, but I don't wanna do that now, maybe I'll do it some other day.
actually even the empty filename is wrong, we have filename.
And looking at that code I find even more bugs.
We added a hacky trick to not force-kill the game if a include in description.ext was not found.
But, that code makes the same mistake as the one that prints the bogus "empty filename" message.
That makes the fix only apply directly to description.ext.
If you include a file and that file includes another file that was not found, we still force-kill the game. Even if it started in description.ext. That doesn't seem like that was the intention
The message is not new.
But I have not seen that message until a couple months ago.
That can happen when that weapon holder has invalid position/orientation.
Did you manage to produce that yourself?
https://feedback.bistudio.com/T180885
It seems configuration for weapon of AH-9 Pawnee helicopter was written and lost it sound sets for shots and tails
thanks for the info.
happened in our last play-test last Sunday with 2.16 (mix of stable and perf branch).
nothing really unusual happened that we noticed
will check logs again this Sunday
Alt. syntax for moveInAny to force move a unit into a seat regardless of vehicle being alive, locked or upside down: https://feedback.bistudio.com/T180897
Right now you can't move say a driver into vehicle if its upside down. Turning it back up within same frame doesn't help.
Hmm. Did I ever report that bug with moveInAny...
(if you use it for a gunner seat, AI vehicles will not drive)
"Smoke out" radio message from units work ok but "Frag out" and "Incoming frag" do not work at all
https://feedback.bistudio.com/T180926
https://feedback.bistudio.com/T180930 missing define in spectator UI script
https://feedback.bistudio.com/T180929 Deleting backpack entity AFTER WeaponAssembled event handler could be useful to pass some properties from the backpack to vehicle it assembles to, for example in case where backpack wasn't assembled once yet and there is no entity attached to it, you can add variables to backpack and pass to assembled vehicle later on the event.
The backpack is already deleted after the EH runs, what is not working?
Hmm, I only checked on stable
dont think the code was touched unless dedmen touched it
Yeah its the same on perf
Backpack is deleted before the EH, backpackContainer returns null inside EH
so how one reproes?
Start game, run code bit:
if(isNil{player getVariable "WeaponAssembled"}) then {
player setVariable ["WeaponAssembled", player addEventHandler ["WeaponAssembled", {call WeaponAssembledFunc}]];
};
WeaponAssembledFunc = {
params ["_unit", "_vehicle"];
systemChat str ["backpackContainer", backpackContainer player];
};
player addBackpack "B_UAV_01_backpack_F";
player action ["Assemble"];
Observe chat
Vehicle exists, backpack doesn't
I want to ask to swap the order so both vehicle and backpack still exist during EH run
So you can transfer variables from backpack entity to vehicle entity
in case vehicle wasn't unpacked yet and non-existent
If done might as well expand the EH so it provides backpack there
so people are more aware of such feature
And I'm telling you the code deletes backpack after event ran
so if it is comes a null then it is something else
Do you see backpack entity in the chat?
havent tried the repro yet i am looking at code
I am not sure how the stringtables work internally, but would it be possible to have an alt syntax for the localize command to force a certain language?
I'm going to add primary, secondary bag to the EH, as I said they do exist at the time EH runs
Great, that will be very useful
enough to retreive variables unless you spawn the code
Hmm, maybe WeaponDisassembled could get same treatment?
whats wrong there
Have the vehicle there too
lemmie check
much easier too, as vehicle is simply moved out
"Disassembled" EH returns weapon, bag1, bag2
disassembled -> weapon, bag1, bag2, unit
weapondisassembled-> unit, bag1, bag2, weapon
Nice!
Gonna add Assembled EH as well since disassembled vehicle is not deleted
assembled -> weapon, bag1, bag2, unit
Not entirely related, but since we're talking about eventHandlers, has there been any notable progress on a solution to these?
Still need the backpack in WeaponAssembled because this is most crucial part, right now if vehicle doesn't exist yet, its already too late to access backpacks as they're deleted by the time of the EH
what vehicle
Oh wait I mistyped it, meant the backpack
weaponassembled -> unit, weapon, bag1, bag2
actually assembled -> weapon, bag1, bag2, unit woll be assembled -> weapon, unit, bag1, bag2 for consistency with weaponxxxx variant
Nice!
Next dev
Thanks!
nope, the stringtable is loaded at game start for the language it is started with, so other languages exist only in files but not loaded
Havent started on those yet
No problem, just wanted to see! Thanks for the update!
KK, grenade impact sound is default for any surface due missing soundHit param in CfgSurfaces
Can game take into account surfaceType to hit correct sound so no need change configuration, idk maybe less work?
https://feedback.bistudio.com/T165776
Someone was working on that not too long ago, I think they basically had it working
messages in this vicinity #arma3_feedback_tracker message
havent looked at those, but throwing grenade at shooting range wooden partitions produces correct hit sound now
impact sound, the collisions are under impactxxxx
wait for the next dev then we take it from there
@ KK: you marked this ticket (https://feedback.bistudio.com/T72838) as feedback in the middle of updating a bunch of tickets about grenades breaking glass, with the same rev number. I just want to check if that ticket's (substantially different) issue was in fact fixed, or if it got accidentally included in the ticket-updating spree because the title has the relevant keywords in it.
mistake
https://feedback.bistudio.com/T180137#2600104
Well-disguised spam
https://feedback.bistudio.com/p/Elizabeth27/
Spam account 😦
nukeded
didnt happen again this time - we played other scenarios tho
Would it be non-backwards-compatible to change canMove to check the fuel level of ground vehicles as well as aircraft? It's kind of bizarre that it doesn't already.
(https://community.bistudio.com/wiki/canMove)
BOOL = ENTITY canMoveInfo INDEX 
Though its probably very few conditions for a new command
- what is the reason Eden isnt able to create a consistent mission.sqm? (different rounding or precision for certain values/parameters, swapping the order of attributes - even if you dont change anything!)
- any chance of this still getting fixed?
its super annoying for diffs/verification/to identify what has been changed/what has been changed/committed by mistake) 🤯
-0 / 0, I get ya
true. there is actually a few more even 🙈
https://feedback.bistudio.com/T180988 Killed and EntityKilled events aren't firing if you crash heli drone while HandleDamage tries to stop the damage
Is there a chance of seeing an improvement to: https://community.bistudio.com/wiki/sendSimpleCommand
or a new command?
At the moment the turning commands are utterly useless as there's some massive delay (even without regard for the radio message) that makes it virtually impossible to do anything but make a vehicle spin around.
Something like
unit CommandTurn direction
or an altSyntax for sendSimpleCommand
vehicle sendSimpleCommand ["TURN",direction];
as the current ways to get an AI driver to face a particular direction are varying levels of suck/broken
Usually by ordering them to move somewhere in the distance in the direction you want
that or removing the delay so it works like player direct command does
My chat gone after press Esc key to open 'pause' menu, no (player messages, system texts, killfeed, VoN names) more, tested on stable and prof v5 in official warlords
Can anyone confirm please? seems this started a few weeks ago but my local arma files are ok idk why if happens
there was some BS regarding ESC menu hiding chat
if(!isStreamFriendlyUIEnabled && (uiNamespace getVariable ["BIS_shownChat", true])) then
{
showChat true;
};
``` on `unLoad` of `RscDisplayMPInterrupt`
So either you have stream friendly on or that BIS_shownChat wasn't set
StreamFriendlyUI, yeah thx 
[Bug] Saving in Eden causes inconsistent mission.sqm
https://feedback.bistudio.com/T181002
Same, I had it accidentally on (not sure how) and I was loosing my mind why things were not displaying. But it is what it is.
do you have any evidence that 1.23e-6 and 1.23e-06 are different value? I dont see any bug, the editor can write file anyway it wants as long as it loads it back with the same values
I assume the issue is that if you're using source control, SQM files are super-spammy with tiny changes.
thats a bug?
I wouldn't call it a bug.
It's quite an annoyance
you can see it as "reproducible builds"
for collaboration the inconsistency is super bad
you want only actual changes to happen
not some "bit flipping" back and forth
understandable annoyance. Is there an option in diff software to not detect moves or some other options? If the container for attributes is some unordered map then I dont know what one can do without changing container and then bunch of other things.
No diff software will be able to figure out that nothing changed is things reorder in such big text file, but quite a lot of annoyances also come from simple numbers.
Open mission, do nothing, save, reopen, save. It's different again :|
I guess it's floating point errors
to me looks like it should be rounded
https://feedback.bistudio.com/T142540 Anybody had this happen to them?
People suddenly started noticing mags vanishing during reload in KotH so I went to investiage, long going issue apparently
Yes, but given I play modded, I can't say it's due to mods or basegame
I've heard reports of it happening very occasionally to other people, but haven't noticed it happening to me personally
right now playing official combat patrol on hosted mp and every reload -1 full mag
Ticket has a video of it happening on empty VR mission 🤔
Try to drag mag from weapon to vest, my disappear after this action
upd: host cannot repro this only clients
We reproduced this on a vanilla server within a few minutes xD
In this video you see me reloading and it shows 7 for the magazine count for a short time and then 6 again.
https://www.youtube.com/watch?v=ucpJ-SA4dBc
We also reproduced the bug where you change a vest/uniform/backpack and single mag also disappears in the same session as the video above.
The other guy uploaded on streamable and the video is already offline sadly.
I still got the same bug from my modded server on a GIF tho https://gyazo.com/18f902af3a44e62b1f731b70ae0000d7
Huh, interesting. I think this bug could use some attention as it affects pretty much all gameplay 🤔
gestures (like throwing grenades) can break reloads and eat the magazine
might be related to that
Any gestures can eat mag another example is mod animations for TFAR
Had this issue for a long time now, but lately people complain about it even more that magazines even vanish while moving them from one container to another...
I made a little fix for magazines disappearing after reloading and it works 9/10 times, it basically checks your mag count before you started reloading and 2 sec after it's done and if it dissepears it regives you one, but for that you have to disable some thinga otherwise people could dupe the mag.
So yea mags dissappear when moving them between containers and sometimes 2-8 seconds after reloading
The netcode for the inventory system is completely buggy, it's crazy
People say that if they drag the magazine out of the gun to their inventory they loose it as well
Quick note: skipTime doesn't respect overcast limits and goes under 0 and over 1.
Ex: skipTime (random 24);
Yes - almost every time, server only and new since Tuesday
We also recently had a report of a player experiencing this.. he had two navid mags, one loaded at 148 shots and one full in the backpack. After reloading the one with 148 shots just goes poof.
Radio protocols when throw frag grenade won't work
https://feedback.bistudio.com/T180926
this is likely an easy fix. simulation change in A3 with the ammo probably broke this
Yeah while testing around i could a unit to lose the mag on each reload...
What are you talking about? From that ticket it is not obvious at all what the problem is. If you have more info, please add to the ticket
Bug in inAreaArrayIndexes:
private _area = [positionCameraToWorld [0,0,0], 20, 20, 0, true, 20];
[
[player] inAreaArrayIndexes _area
,[player] inAreaArray _area
]
``` => `[[],[B Alpha 1-2:1 (Sa-Matra)]]`
```sqf
private _area = [positionCameraToWorld [0,0,0], 20, 20, 0, true, -20];
[
[player] inAreaArrayIndexes _area
,[player] inAreaArray _area
]
``` => `[[0],[B Alpha 1-2:1 (Sa-Matra)]]`
Simulation type of CfgAmmo shotGrenade doesn't trigger radio protocoll
Yeah there's a voice line for 'frag out!' etc that's unused, while the smoke one gets used
there's also an 'incoming grenade' one that's unused
😮
-1 but usually it checks <0 so yeah not a bug
doesnt mean the simulation is wrong as kju seems to be implying
I changed the simulation for the frag grenade to the simulation of the smoke grenade, after when throw frag grenade the radio protocol trigger "Smoke out" message
this reason I assumed that the problem is in the shotGrenade simulation
I don't think that's what kju meant. I think they were just saying that something changed sim type in the A2>A3 change, and the voice protocol thing was never updated to account for that. So it was broken by a change in simulation, but the simulation isn't broken (and kju didn't say it was).
I've got a feeling it's something like ShotGrenade didn't exist in A2 and frags were simulated as bullets.
they were shotShell iirc
well its an assumption - from what i recall both grenade and smoke grenade did work in A2:OA when it was introduced.
what changed in A3: 1) the RP system was revised to some extent 2) smokeX simulation was introduced
so yeah could still be something else. however there is a good lead and the test by @unborn acorn tells its related to the simulation
Tested it, works in 0.50, doesn't work in 1.10
Config differences:
1.10 reduced damage and typicalspeed, added more sounds, added deflecting = 15; and shadow = 1;
class CfgAmmo {
class Grenade;
class GrenadeHand:Grenade {
hit = 20;
indirectHit = 18;
indirectHitRange = 7;
typicalspeed = 20;
};
};
OA:
hit = 20;
indirectHit = 18;
indirectHitRange = 7;
typicalspeed = 60;
class CfgAmmo {
class Grenade;
class GrenadeHand:Grenade {
hit = 10.00001;
};
};
Posted in the ticket
Huh, turns out there is absolutely no scripted control over countermeasures
No way to get current mode, no way to select mode
selectWeaponTurret, action UseWeapon, non seem to work with vehicle CMs 🤔
ARRAY = VEHICLE currentCM TURRET -> [WEAPON, MUZZLE, MODE, MAGAZINE, [magazineId, creatorId]]
BOOL = VEHICLE selectCM [TURRET, WEAPON, [MUZZLE, [MODE, [MAGAZINE]]]] -> BOOL = found
Personally I need it to check if player selected single use aircraft countermeasure and revert it back to burst only to forbid single drops
Anything else to include in request ticket?
for anyone else interested in the particular issue when AI stops following the convoy leader: https://feedback.bistudio.com/T181098
I have been able to reproduce the problem. which means it should be easier for a developer to pinpoint the issue. hopefully it can still get fixed because it can happen so often that it makes it impossible to create cool convoys in Arma 3.
How feasible it would be to look over LODing, it's bit ridiculous with vegetation when you zoom a tiny bit and models turn 2D 
it feels like there's something really wrong in how it works when objects in center of your FOV are so low res while ones on the edge are fine.
(using extreme object quality here but it does not matter)
it feels like there's something really wrong in how it works when objects in center of your FOV are so low res while ones on the edge are fine.
yes, iirc there was a wrong calculation of things very close to screen centre, I saw it reported years ago (Arrowhead perhaps?) but I don't think it got fixed
at least you have very fast 2D billboards!
and also bit gameplay disrupting too as it's harder to see through the 2d ones as they're bit more dense usually.
I feel it got worse, don't remember having this problem some years ago.
private _area = [positionCameraToWorld [0,0,0], 20, 20, 0, true, 20];
[
[player] inAreaArrayIndexes _area
,[player] inAreaArray _area
]
``` => `[[],[B Alpha 1-2:1 (Sa-Matra)]]` this is still bugged though
Posted it in the original ticket: https://feedback.bistudio.com/T170535
maybe @gaunt depot could try comparing some trees through strong scopes in the old build.
I think it got this bad after Apex update with LOD blending
I remember noticing it back then but didn't both complaining
[[0],[B Alpha 1-1:1 (KK_DEV)]]
dunno why you have empty array
🤔
can you be specific with repro plz?
AGL/ASL issue?
i just ran your code
Yep, it is
Try on a mountain/large hill
inAreaArrayIndexes uses ASL, inAreaArray uses AGL
yep
inareaarray - center: Array format Position2D or PositionAGL, Object or Group
inareaarrayIndexes -
positions: Array of Objects and/or Positions and/or Locations and/or Strings and/or Groups, where:
Objects - entities
Positions - 2D or 3D positions
Locations - locations
Strings - markers
Groups - AI groups, the position is the leader's position
dunno what can be done it is not the same command that just returns indexes
I'd make it uniform and change it to treat position as AGL
yeah too late for that
I'm fine with leaving it as is, just mention on wiki, but I'd prefer to do the right thing and fix it now
The chances for it to break something are very slim, you have to use the command and care about Z to affect you
@gray wharf ^^^
it is 4 versions away, no chance
2.14 -> 2.18
since it lasted that long I guess it was never this big of a deal
I'm pretty sure not much people use it
i'm not
it was built for speed AFAIR so the position aquisition is the fastest as well as it take another bunch of positions
only antistasi uses it on GH and they assume it's ATL 💀
strange, it says what kind of position it takes right there on wiki
if you pass object for position all is good
in fact I would argue it is more convenient inAreaArray to have world position rather than agl
so all good?
inAreaArrayIndexes uses PositionASL for center, unlike inAreaArray that uses PositionAGL
Had a look at it, its just as shitty in both 0.50 and 1.10
The threshold in 0.50 seems to be somewhat different but its hard to say exactly as binoculars have different magnification
Same issue, focusing on a tree makes it go into lower LOD
i also notice this during debriefs after ops; people at edges of my screen will be res 0 but center will be res 1 or 2 regardless of object quality
@untold sky did some digging in some 3d calculations recently in 3den, maybe this one won't be too hard too 
inshallah
This ASL-AGL difference is only about area position, I'd change wording in warning a bit
fixed
Actually the assumption is that it uses the same defaults as inAreaArray, which is infinite Z anyway.
The inAreaArrayIndexes documentation doesn't mention the defaults.
I haven't used Z with the command before so I'm not affected but I'd vote to fix it and make it uniform
Anybody would find something like BOOL = isDeleted ENTITY useful?
A command that would check if deleted flag was set or if entity is already deleted (null)
private _obj = createVehicle [typeOf player, getPos player, [], 0, ""];
isDeleted _obj; // false
deleteVehicle _obj;
isDeleted _obj; // true
isNull _obj; // false
Perhaps as an extension tohttps://community.bistudio.com/wiki/getEntityInfo
Good idea!
On another thought, if object is already null, it would return nil
While dedicated command should return true 🤔
What I mean:
deleteVehicle _obj;
isNull _obj; // false
_obj getEntityInfo X; // true (isDeleted)
...later...
isNull _obj; // true
_obj getEntityInfo X; // nil (isDeleted)
unless the scripting command will return true that property for deleted entities
object becoming null after it set to be deleted is very quick, what is the use case?
I dont like this
Don't have a dev build right now to test, but doesn't ENTTIY getEntityInfo INDEX return nil for null object, according to wiki?
Meant that I don't like that, if that's the case
probably returns default value
If so then having it as part of getEntityInfo should be fine
true for when flag is set false otherwise, but what do you need it for?
Personally I have two use cases:
- I have a complex branching conditions for deleting projectile inside
Firedevent handler and some branches end up with projectile deletion. Checking if projectile was deleted later would be useful to avoid other code that expects it to be there. Sure, this is solve-able by restructuring code, adding deleted flag, etc., but this is second time when I needed to know if entity was marked for deletion - If you add
DeletedEH from withinEntityCreatedand delete the entity right away, the event wont fire. I have some logic that relies on this EH, but I could skip it if I was able to check if entity is already marked for deletion.
Didn't explain full extent at #2 but I have cases where Deleted doesn't fire on deletion and having that flag would let me skip code that assumes the entity won't be deleted right away
🤔
are we talking about ammo projectiles?
they have terminated flag that is set when they need deletion
Yes, ammo
if(!isNil"eceh") then {removeMissionEventHandler ["EntityCreated", eceh]};
eceh = addMissionEventHandler ["EntityCreated", {
diag_log [diag_frameno toFixed 0, "EntityCreated", _this];
deh2 = u addEventHandler ["Deleted", {diag_log [diag_frameno toFixed 0, "Deleted 2", _this]}];
diag_log ["Deleted info 1", _this getEventHandlerInfo ["Deleted", deh1]];
diag_log ["Deleted info 2", _this getEventHandlerInfo ["Deleted", deh2]];
}];
u = group player createUnit [typeOf player, getPos player, [], 0, ""];
deh1 = u addEventHandler ["Deleted", {diag_log [diag_frameno toFixed 0, "Deleted 1", _this]}];
deleteVehicle u;
```=>```
19:10:08 ["1467748","Deleted 1",[238d4b79240# 166165: c_poloshirt.p3d]]
19:10:08 ["1467749","EntityCreated",238d4b79240# 166165: c_poloshirt.p3d]
19:10:08 ["Deleted info 1",[true,false,2]]
19:10:08 ["Deleted info 2",[true,true,2]]
Both EHs exist, only one fires
so entitycreated keeps reference to the unit which is why it is not set to null but deletion eh simulation had run already, so you can add eh as it is not null but it will never run.
I suppose it makes sense yo have getter for set for deletion option similar to isDeadSet
Fixed: Any surface impact for grenades was using the ground impact sound
This fix not for surfaces i.e. not for surfaces where characters can move
see 1st picture (metal wall) - metal impact sound, 2nd picture (metal floor) - still default ground impact sound
This is due issue from this ticket https://feedback.bistudio.com/T165776
17:26:15 Attempt to override final function - bis_fnc_spaceship_removeparachute_meta
Has it always been the case that whenever I preview a mission in editor I get 5k lines of this?
Tested on stable, vanilla, no startup parameters.
I find it weird that initFunctions.sqf tries to overwrite functions itself compiled final
Nevermind, that "Recompile Functions" checkbox in Eden Editor was checked....and suddenly previewing is like twice as fast.
oh you have to enable recompile...
if(!isNil"eceh") then {removeMissionEventHandler ["EntityCreated", eceh]};
eceh = addMissionEventHandler ["EntityCreated", {
diag_log [diag_frameno toFixed 0, "EntityCreated4", _this, u getEntityInfo 14];
deh2 = u addEventHandler ["Deleted", {diag_log [diag_frameno toFixed 0, "Deleted 2", _this]}];
diag_log ["Deleted info 1", _this getEventHandlerInfo ["Deleted", deh1], _this getEntityInfo 14];
diag_log ["Deleted info 2", _this getEventHandlerInfo ["Deleted", deh2], _this getEntityInfo 14];
}];
u = group player createUnit [typeOf player, getPos player, [], 0, ""];
diag_log [diag_frameno toFixed 0, "EntityCreated1", _this, u getEntityInfo 14];
deh1 = u addEventHandler ["Deleted", {diag_log [diag_frameno toFixed 0, "Deleted 1", _this]}];
diag_log [diag_frameno toFixed 0, "EntityCreated2", _this, u getEntityInfo 14];
deleteVehicle u;
diag_log [diag_frameno toFixed 0, "EntityCreated3", _this, u getEntityInfo 14];
9:13:55 ["110865","EntityCreated2",[],false]
9:13:55 ["110865","Deleted 1",[2518fcc8280# 10: b_soldier_01.p3d]]
9:13:55 ["110865","EntityCreated3",[],true]
9:13:55 ["110866","EntityCreated4",2518fcc8280# 10: b_soldier_01.p3d,true]
9:13:55 ["Deleted info 1",[true,false,2],true]
9:13:55 ["Deleted info 2",[true,true,2],true]
Nice!
Does it also work for projectiles or it will be in getShotInfo?
Does objNull getEntityInfo 14 return true?
gimmie repro and i will tell you
player addEventHandler ["Fired", {
private _shot = _this select 6;
systemChat str [_shot, _shot getEntityInfo 14];
deleteVehicle _shot;
systemChat str [_shot, _shot getEntityInfo 14];
}];
getEntityInfo is for EntityAI , shots are Entity, so it will have to go to getShotInfo, so it will have to be ```sqf
player addEventHandler ["Fired", {
private _shot = _this select 6;
systemChat str [_shot, _shot getShotInfo 7];
deleteVehicle _shot;
systemChat str [_shot, _shot getShotInfo 7];
}];```
What do these flags return on null entities?
default
My original idea for isDeleted assumed it to return true if its null entity
So you can check for both flag and actual deletion
you can always do the isnull check
Alright
the flag is just that whether or not the flag is set, it can only be set on existing entity so null entity is false
Yeah, I guess doing additional check is alright
isNull _entity || {_entity getEntityInfo 14}
you should do it anyway for other getxxxinfo
otherwise you wont know if you are getting a default value or real value
Deletion is kind of on another level than just entity property, so I thought it would be logical for it to also cover nulls
But since its part of entity info and not a dedicated command, it wont make much sense for it to return true
if (isAboutNullButNotQuiteYetButSoonWillBe _entity) then {...}
isUnitDeadButTheirBrainIsStillAlive ENTITY
private _isUnitDeadButTheirBrainIsStillAlive = not alive _unit && isNull group _unit;
if (_isUnitDeadButTheirBrainIsStillAlive) then { hint format ["%1 is dead", name _unit] };
no, when it reads the stringtable it skips other languages. They aren't loaded or stored
and is there a trolling possibility to setLanguage "Greek"? 😄
Yeah doable. "VTMAbs, _dirWanted RadioMessageVAzimut" into ticket
would need to reload all stringtables. probably not
setLanguage "Pirate" 🥲
cool do you want me to make a ticket or did you make an internal one?
you
rgr
just press ARRRRRR
@untold sky https://feedback.bistudio.com/T181170
Ticket requesting to grenades make foliage impact sounds not soft ground sound when hitting vegetation
https://feedback.bistudio.com/T181237
Any chance to get isNil's Syntax 3 (in dev) to work with locations?
what do you mean?
Have the namespace isNil variableName syntax support locations
Because locations support setVariable/getVariable
and ctrl and everything that can have a namespace actually
https://community.bistudio.com/wiki/createSimpleObject supports both shapeName and className
it seems to first try to generate via className and only after with shapeName (model path)
this leads to
Calling 'CfgVehicles >> ""' (empty parameter name)
testheli = createVehicle ["O_Heli_Light_02_unarmed_F", player modelToWorld [0,200,100], [], 0, "FLY"];
testheli setVectorUp [0,0,-1];
testheli setVelocity [0,0,-50];
testheli addEventHandler ["Killed", {systemChat str ["Killed", _this]}];
````Killed` event handler doesn't fire 
and if you add setvelocity with delay?
or add EH first thing after vehicle creation?
no difference
with delay it happens sometimes
Looks like there is no event with some crash angles
is the helli definitely killed as in alive false?
alive testheli => false
could you make a ticket plz
unable to get word:'IAmReady' in protocol:'radioprotocolbase'
anyone knows how this can happen? it seems an unit without a proper identity (speaker) tries to use the RP
however the engine should default to a valid speaker - at least i assume this is what this is about:
No speaker given for 'Ryan Nelson'
Speaker male03ger not found, patched to default Male01ENG
https://feedback.bistudio.com/T178415
requesting 2 commands for set time to reload magazine and time between every weapon shot
- getters
is there any interest/ability to expand rebreathers to be able to be facewears/helmets?
a3 has working rebreathers? can units go underwater?
Yes...?
I know no one really uses the underwater stuff but I didn't realise people didn't even know it existed
yeah, i think most people dont know about underwater operations in arma 3 unless they're looking for it or join a unit that does it. it seems like its pretty niche and sort of a underrated feature
I don't know if it's underrated, seems very niche
Pun detected
If you'd just played the showcases you'd have seen it 
I think it was the 2nd or 3rd showcase
oh okay, i forgot
Switching to a playable unit and then switching back will make the 2nd unit frozen and wont fall in formation
If I'm an SL and switch to the sniper for example and switch back to the SL, the sniper unit will stand there and not fall back into formation unless given a command
STEPS TO REPRODUCE
-Place a group that is all playable
-Switch from one to the other and then back to the first
-The unit you switched to the first time will stand frozen and not fall in formation
post it in https://feedback.bistudio.com
Anybody would find useful a scripting flag to force close the parachute?
I assume that current implementation initiates closure when parachute is touching ground which ends up with parachute starting and stopping closing animation making it looks terrible, it looks even worse with remote parachutes. The whole thing is ancient, but if we could control that flag through scripting we could improve it in our missions
ENTITY forceParachuteClosing BOOL
Maybe even 3 states: Default (closes when touches ground by the engine), Always closing, Never closing?
an eventhandler for when its closing would be good too
so you can see when the parachute landed
iirc it was just isTouchingGround check?
But yeah, firing an event when the engine would've liked it to start closing could be useful too
at least on terrain its the same yeah
Less per frame checks
I think isTouchingGround also works with touching objects?
But I also remember it being unreliable for remote vehicles
just delete the chute
these details dont even register in players mind
tree falling in the woods stuff
I do notice it all the time. I know Arma is made out of jank but I still try to reduce it as much as possible
But I do 
There are sometimes cases where parachutes aren't deletable too, not sure why this even happens
your weakness has always been budgeting time to stuff the end user doesnt feeel or care about xD
precious dev time
I eject player when parachute have landed (touching ground for X time), but then deletion doesn't seem to work randomly
Half a year spent developing accurate damage tracking, just to show grenade icons in a kill feed and magazine name on "killed by" screen 
i wasnt joking 😄
nice
do you track when laser target is near the killed entity?
and give the laser target owner an assist?
i thought of doing something like that to reward people who lase targets for CAS in coop, but never tested anything
Its even more complicated, shooter checks if there are laser targets on HitPart hit and informs the server that this damage combo should be attributed with laser assist
ah wow so laser logs as a hit?
So you can lase a helicopter, it gets damaged, crashes 10 seconds later and lasing player is still attributed the kill
No, HitPart from the shooting damage
Checks for lasers near hit target
ah
how about like, player with laser, adds variable to the target. killer checks for recent laser variables
I thought that looking for lasers on hits would be cheaper than PF for laser shooters
Plus its more reliable so you know target was lased exactly when it got hit
No assumptions
thank hell i never had to get into damage/kill stuff like that
I felt like these perfectionist movie/cartoon makes that spend years on one project never finishing it or finishing it and its all for nothing as nobody cares
it is disappointing tho when you spend ages doing something cool and no one notices (despite good traffic/playtime) 😄
I know almost everything about Arma damage now
Just in time for RV to become irrelevant 👏
still i think we were both quite lucky with being able to create things and see thousands of players having fun with the creations shortly after ... obviously fewer players now, but thinking 2014-2019
Yeah, that was always #1 thing for me, players having fun
yep i did most of my dev stuff in debug console while quietly observing with zeus
I wonder if we ever could get lineIntersects* that works only within a single entity (several entities), to avoid hacks of spawning simple model outside of the map and all the mess associated with it? Is there even such functionality in the engine?
no one ever told you... you did a good job, you can rest. no need to keep arma dev. you can ride off into the sunset, take up fishing or something
that goes for @solid marten too
take up a quiet hobby, far from RV4
Start a chicken farm?
nice way to ditch the competition
but please keep it about FT / suggs 😄
what does?
ofc there is. that's how it works in the first place (it transforms the world pos to model space and performs the intersection)
Would've been useful to have proper version of https://community.bistudio.com/wiki/intersect
That one only works for named selection or some mess like that
Yeah
https://community.bistudio.com/wiki/ropeEndPosition
are simulation scope positions, right?
Am I the only one who missed FPS cheat setting 5 FPS?
Very useful to reduce network updates rate
Rope network sync is so bad I wish I could just create local ropes or something
Is setGestureSpeedCoef possible? Met a problem where animations get speed up due to arma mess but setAnimSpeedCoef is not enough to fix them, unit ends up with x4 gesture animation 🤔
TF2 Soldier be like
@solid marten regarding https://community.bistudio.com/wiki/diag_remainsCollector , it seems strange to have meta data in the same array as the entities
[disposalPeriod, managerMode, removeLimit, removeMinTime, removeMaxTime, minPlayerDistance, unit1, unit2....]
any chance to split
[
disposalPeriod,
managerMode,
removeLimit,
removeMinTime,
removeMaxTime,
minPlayerDistance
],
[
unit1, unit2....
]
]```
Agreed. Or move those info into a separate command
its diag command for diagnostic, added so we can identify if it doesnt work as expected
Ropes attached to units update unit position not frequently enough:
https://feedback.bistudio.com/T181442 (Repro included)
Units get rope simulation scope position or something?
Yeah, exactly that, unit uses rope's simulation scope position (shown with drawIcon3D here) instead of rope's visual scope position
Yup
Updated ticket
But when you control rope attached unit it uses render scope position instead
Just arma things
Anyway, plox fix
With mess like this sometimes I wish there was a way to stop network position updates or something
ENTITY ignorePositionUpdates BOOL
Though I'd still need orientation and velocity, so its not as simple as just one flag
Wish you could provide a timeout for moveIn* commands
So when you try to moveIn* into remote entity that has their game frozen/network down, it would get aborted if request taken too long
Seen cases where moveIn* seemingly failed, player went on with their game the other way, then remote client finally went alive and player got teleported into a vehicle
player moveInCargo [vehicle, cargoIndex, canReassign, 3]; If request was handled after serverTime + 3, abort it
something like this
This way you could setup loading screens more reliably
https://feedback.bistudio.com/T181456
Hey please let me synchronize everything
Good suggestion, would make 3DEN much more useful
Unless these connections present only during mission read 🤔
Seriously it sometimes simply confuse me why it doesn't work and take a while to realize
https://feedback.bistudio.com/T181382
AI unit turns unresponsive when you switch from them to another playable unit (eg. from the SL to the sniper and back to SL, sniper unit will not fall back to formation and just stand there even as the group has moved off)
Some concrete examples and repro would help to push the ticket
Will do after I'm back
If you guys could take care about this, it would be greatly appreciated. Moreover, i don't think it will be difficult.
https://feedback.bistudio.com/T180432
This game has everything required for naval battles, there just need to fix some details
I just posted this task to explain the issue with MFDs as well https://feedback.bistudio.com/T181499
Updated
Is it technically possible to adjust attach point of rope after ropeCreate?
Looks like you can sort of do it with https://community.bistudio.com/wiki/ropeAttachTo but you have to re-attach first plus rope loses its tension resulting in it springing up and down
I wonder if there could be easier way
ROPE ropeSetAttachPoint [POSITION, VECTOR] (toPoint, ropeEndDownVector)
No change of attached entity, just update attach position
Attach rope to helper object and reattach that instead?
Sure, but this loses some of rope's properties, like unit/players rotating back to rope's anchor orientation
Gonna have to simulate that too
it looks like if you try to hide setFeatureType 2 objects they still render the proxies ;d
advanced stealth capabilities
In the upcoming 2.18 update, can we please modify the MagazineUnloaded event handler so that it fires when the reloading sequence starts? Or add a Reloading event handler?
KK, is there any news fixing MP work of Jets DLC scripts?
https://feedback.bistudio.com/T174022
http://feedback.arma3.com/view.php?id=21028 <-- Was this infeasible? It's always nice to know why a certain feature won't make it into the game, even if it's "we don't think it's worth the effort" :)
That is the case.
The connections get resolved at mission start, and are not kept.
I don't think that will change.
Technically yes, but probably not.
Please make sure to have it fire at the start of the reloading sequence as well.
The magazine is not being unloaded at the start of the reloading sequence so no to that.
Reloading, probably, maybe not 2.18
Done.
Same as Muzzle event.
"MagazineReloading"
[B Alpha 1-1:1 (dedmen),"arifle_MX_ACO_pointer_F","arifle_MX_ACO_pointer_F",["30Rnd_65x39_caseless_mag",30,1e+07,0],["30Rnd_65x39_caseless_mag",30,1e+07,0]]
Ugh what a fun mess.
"Reloaded" eventhandler on soldier fires after the reload is done.
On a vehicle, like static 50 cal turret, it fires at start of reload.
Good luck documenting that 😄
the main issue is that when you start the reloading sequence, the player loses a magazine in their inventory (even though the magazine in the weapon slot still is there) and it is re-added when the player finishes reloading. This effects any scripts that modify inventory space. So that's why I was asking for something to trigger when the unit starts reloading, so I can detect if this is going to happen.
ace beats it with this mutex https://github.com/acemod/ACE3/blob/be77ef233ee87e00cca6e71a2ae307c6b073c6ca/addons/common/XEH_postInit.sqf#L439-L495
and I beat it with my own listen handler: <#arma3_scripting message>
yep the animation sources are backwards too
its so interesting that bots and spammers target the bug tracker
It's not specifically because it's the BI FT, it's just that it's a form you can submit text to, and can be found by Google and by links from other popular sites, and the account creation process isn't very bot-proof.
Same thing happens to all sorts of forums and stuff. (You can tell when a forum no longer has active moderation when the bot posts stay up...)
lol
https://community.bistudio.com/wiki/Arma_3:_CfgWorlds_Config_Reference#class_RainParticles
When snow param is true the following changes occur in the engine:
thunder and lightnings effects are disabled
This ain't working 👆
I still have thunder sound and lightnings. Tested in CUP though. I can test in Altis if you want.
BothsnowanddropColorStrongvalues show up as strings (I don't know if related).
setRainseems to work fine with both thunders and lightnings.
vanilla repro and make ft ticket
Altis is even worse, there's also rain sounds whereas CUP does not. Snow shows up.
make ft ticket
yep, i'm trying to use my forum account but it doesn't let me 😦
(cleaned the config)
iirc the accounts are separate on the phab instance?
I guess 😦
they're separate.
should i report the random freezes i have with the game on windows 7 with very high memory usage? i have had it like inconsistently hanging and never responding again after selecting server browser on startup (it shouldnt be happening and i use world=empty) and also the strange ocassional access violation on shutdown/exit?
No, you should change your faulty RAM sticks as we discussed last time #perf_prof_branch message
And probably also upgrade from Windows 7 😅 but that's secondary (could leave your PC exposed to malware though, since Win 7 hasn't been receiving security updates for 4 years now iirc)
But yeah you're seeing the software glitch out possibly due to your hardware issues, and it's not right to send devs on a wild goose chase when you know that your system's memory has problems
So RMA/replace your RAM and see if those issues persist
I don't know what BI's policy on supporting Windows 7 and 8 systems, but min requirements for Arma 3 still include Windows 7 so i guess they're supporting it. Although Steam only supports Windows 10 and later, so I wonder if you have an updated (or even legal?) version of Arma?
workshop build runs on win7
prof doesnt i dont think
tho that may be 32 bit not win7
okay thanks
wdym? arma 64 and 32 runs on 7, perf/prof is 64 only, dev should still fully work on 7
steam client versions older than the support end date are still fully functional, the end of support claim only applies to versions after the date
this has been the same way with xp
Ah okay
Yeah that's just 32 bit that was dropped
Not an official answer, but I'm assuming that all the work it would involve to solve certain issues make it a non-priority.
First problem, what do you do with the hand reload anims. You can't tie them to the mags, since different weapons can use the same mag. You can't leave it on the weapon as that means a single hand anim will be used for different mags which will result in clipping and whatnot.
If you solve that problem, next comes the actual creation of new anims for each weapon/magazine combos that are in the base game.
Then the modification of the existing weapon models so they use a proxy insead of a modeled mag.
Additionally, if the hand anims change to vary by mag, what about the mag itself, it's an anim on the model synced with the character. How to deal with that. :P
etc
Some file paths to mechanism sounds of Promet and MX assault rifles are gone in configuration, I don't think this is by design
https://feedback.bistudio.com/T181573
any chance for a config parameter to define the "seat priority" if you order AI (or AI itself, or via sqf - aka moveInAny) to enter a vehicle? (or a scripting command if thats easier)
if it would be at least driver-commander-gunner - yet with multi turret setup, you end up often the main gunner not even be 3rd 😐
The EH was already added by the time you replied 😄
When creating FT tickets, make sure to provide enough info.
Don't be the guy that makes me write these
that new feedback tracker assistant AI really asks a lot of questions 
how is the situation with game crash on exit? still need more crash logs on that?
oh yeah good question, i wonder the same
Since there seems to be a bunch of awesome additions coming for 2.18... I wanted to throw in a couple of my older FT tickets for consideration again
https://feedback.bistudio.com/T174951
https://feedback.bistudio.com/T174936
crash logs are useless, need a repro. And I can't find one
I propose a "compartment" parameter for moveInAny commandd. It will be really usefull for large vehicles with 100+ proxies
https://feedback.bistudio.com/T181607
Is there actually any distinction between "compartments"? Never heard the term re Arma.
yes compartments are what stops like people in the cargo bed of a truck from moving to the drivers seat
hmm, that reminds me though. I should really write up that moveInAny bug :/
could extra logging help to track it down?
I don't know what to log
array.hpp(1411) : Assertion failed 'base::_n == 0'
could this be related or another crash type?
I don't believe we have a way to detect a change in a turret optic's zoom/FOV
Proposing: "TurretOpticsModeChanged" Event Handler
https://feedback.bistudio.com/T181643
I wonder who will win
https://github.com/CBATeam/CBA_A3/pull/1665
is this:
_unit, _weapon, _muzzle, _newMagazine, _oldMagazine?
edit: yeah I think I'm interpretting it correctly
Yes definitely. That says "Someone inserted into this array, while we were destructing it"
I also think that is the crash on exit thing.
Sadly this assert triggers after someone else messed it up.
its the same as the old reload handler things
Not the same thing. That's for nvg and thermals
That's true, my bad
Is there a chance for this to be fixed for good in the engine?
Otherwise somehow let us script fix it?
Fast response time by Wulf
Wasn't there a fix to damage state synchronization that should fix the edit terrain object module?
There was engine and script fixes for module work, works ok on dev now
I just updated this post because i forgot to say daylight property of compartmentLights is not working underwater
https://feedback.bistudio.com/T181499
Not entirely true. I have a boat with a compartment below the water line and daylight compartment lights work fine.
Can you give me the config of your compartmentlight ?
Maybe this glitch only happen on type="submarinex" boats. I have a 75m long submarine (reglementary size) and light doesn't work during day
I made an ocean map (3000m deep) and i don't see the light in the depth
Yes, just pointing out that your observations are not sufficient to result in a summary that "daylight compartment lights don't work underwater".
What should i add ? On every map i tested i don't see the light during day. There is a video joined to my post so you can see what i'm taling about
But indeed, i should have tested it on more vehicle type before saying compartmentLights doesn't work during day. But anyway that's a fact, it doesn't work underwater on submarines
Sound FX transition from building_interior to building_exterior instantly, but to building_interior this goes smoothly
https://feedback.bistudio.com/T181711
@gray wharf your note on getMissionPath states:
Before the introduction of getMissionPath, absolute mission directory could be obtained as follow in config:
does that mean you can use getMissionPath in the config? cause it sounds like you could. maybe clarify?
I mean it as "before that, you could get the mission's path this way"
so not related to the config then? purely sqf?
before, you had to use "get mission config value" as a hacky workaround
now you can just use getMissionPath
I'd like to precise, that i only tested the compartmentLight in ViewPilot LOD
Seems the OnMissionNew event handler for 3DEN has two different behaviors. If selecting "New" and choosing same map or loading from main menu, you can set variables into missionNamespace during event and it will carry through to editor. If you select "New" and choose different world, you can see it runs the event and then , I suspect after the event, it loads the new world which wipes out missionNamespace. Is this intended? It seems inconsistent to me.
Changing the terrain probably loads a new mission thus resetting the namespace.
Save the variables in Eden display to have them persist between loads
Shouldn't that be when the onNewMission event handler fires? Not prior to? Functions are persisted but any onNewMission event is discarded. So eden display is Uinamespace? Trying to track it down. Just seems weird it is different than other behavior.
Eden Display is its own namespace which is persistent between terrain changes and preview iirc.
Not sure about functions . Maybe they get recompiled in the background.
I'm posting a new function inspired from BIS_fnc_findInPairs. This function is better than BIS_fnc_finInPairs andd i've been using it in pretty much all my mods.
https://feedback.bistudio.com/T181813
for "_i" from 0 to 10 do {}; syntax is better than your syntax, but otherwise looks nice
Wish you could do ARRAY param NUMBER as out of bounds safe select
{if(_x select _column isEqualTo _value) exitWith {_forEachIndex}; -1;} forEach _array;
Wont work if _array is empty though
maybe can be done with isNil check, also gets rid of -1;
You basically can though, can't you?
_array param [5, "myDefaultValue"];```
[5, nil], but its an extra array!
muh microseconds
Actually just [5]
Not very serious though, but I just hate seeing single-value array there
I still need to know when _column is out of bounds. With my syntax, it will throw an error
Scripting error?
Yes, but indeed, it can be coded in a better way
It wont trigger an error because you can have single out of bounds index with select
[1] select 1 => nil
[1] select 2 => error
what is safe select? you can have default value returned with param already, when do you not need that?
Sure its already works, it was more about getting rid of extra array around index
hence my question
[1,2,3] param 5 instead [1,2,3] param [5]
see my question
Having nil is sometimes better than some value
When command or function returns nil and its used in condition, it skips conditioned code blocks, which is very useful
thats not intended functionality just byproduct
But its a useful tool
It sort of lets you have 3 values in bool
if(true) then {123} else {321} => 123
if(false) then {123} else {321} => 321
if(nil) then {123} else {321} => nil
not about value but code executed or not executed here
statements skipping when nil is such a WTF moment, for a scripting language meant to be used by people with not much experience
Another reason could be using nil to initialize something heavy only when you need
As in:
private _value = _array param 5;
if(isNil"_value") then {_value = call someHeavyFunction};
vs
private _value = _array param [5, call someHeavyFunction];
Its already there with _array param [5] but it was more of a my OCD kind of thing with seeing single-value array when it could've been just the value
private _index = {if(_x param [_column] isEqualTo _value) exitWith {_forEachIndex};} forEach _array;
if(isNil"_index") then {-1} else {_index};
Very interesting way to code it. I didn't know about _forEachIndex at the time i coded this function. I think it deserves some speed tests
Very nice job, your way is faster
Can you post your code in my topic ?
Your code is about 25% faster
NeverMind, i'll do it
there is a bug with the useAudioTimeForMoves setting it to true will disable object collision and make the player go through any object similiar to hideObject 😭 😭 😭
should i create a ticket?
Fastest way is:
params ["_array", "_value", ["_column", 0]];
private _index = {if(_x param [_column] isEqualTo _value) exitWith {_forEachIndex};} forEach _array;
[_index, -1] select (isNil"_index");```
No, fastest way is using findIf
Oh yeah, findIf would be even faster
I appreciate the help. But my dumb brain just discovered that onTerrainNew exists. Solves the issue. I am caching this data in UI namespace already - just needed a time/event to happen so I can dump them into missionnamespace so they are accessible in editor and debug console.
Is there a chance that this gets a fix? I'm asking because idk what has more priority and if it is worth a ticket.
if you want something to be fixed then make a ticket for it
If you like a thing you better put a ticket on it 🎵
Kinda pointless and TOO specific suggestion... but, a return value for drop which returns the spawned object?
It would be great if we could play the sound of a video with playSound3D. There currently is no way to play a video with located sound
There just need to extract the ogg sound from the ogv and play it with XAudio (after converting it) or something else
Can be easily done in c#
but it's a particle, not an object per se
It is
...Like, you can deleteVehicle it
then make a custom dll/extension that implements it to show that it is easy to be done in A3
Weally yes
Here some configuration problems clear to climb
upd: list too big for discord, all I found config issues added to pastebin https://pastelink.net/a3configIssues, plz fix it)
Just investigated, 63 allObjects 3 or -1 allObjects 3 is the only way to get particles, which is... heavy enough
Idea: Introduce ARRAY = NUMBER drop ARRAY which returns particle entities so you don't have to search for them
NUMBER being number of particles to drop
But how can you randomize drop if you do this
Hmm, good point 🤔
How about:
ENTITY = drop ENTITY;
``` to trigger particle source to create a particle through script and also return it
With all associated particle source randomness
Except setDropInterval?
Yes, regardless of interval
So you can setup a source and drop particles when you want it yourself
Was instructed this belongs here as opposed to its initial channel.
Will repost files here when I get home
Would love some Insight if anyone has an idea
sorry for asking but, is there support for actual 3D audio or hrtf?
hope you are aware no one will test this with so many mods you are requiring. If this is an actual game bug rather than one of your mods being shit, then replicate the same behavior in vanilla.
not really for this channel no? There are hrtf DLLs out there and they change how audio in the game behaves but it's not perfect, so I'd say support is to the same extent as for VR.
Most of the mods I'm using are from people in this very discord, for what its worth. And the script im calling wouldnt really have any interaction with.. any of the mods that I can think of. But I'm not an expert, so maybe I'm wrong on that front.
If anything I believe the issue may come from my own scripts currently running on the init, and OnPlayRespawn if anything.
not worth much.
im in here, after all
Considering the deployment after this one uses some of yours that's great timing
you will be certain of it if you replicate the issue in vanilla. If it happens in vanilla game, you'll know it's game's issue not some awful piece of code included in one of the mods or made by you.
I'm not certain it is an arma issue. It may be my scripts
when you do #monitor 2, for example, to monitor FPS every 2 seconds and then you log out, without doing #monitor 0, before logging out, it will still continue to monitor FPS every 2 secs, basically showing stuff to a "non admin"
also, please, concider fixing objects distance windows/field in video settings, because any time you type anything there, it always does value 5.
so you always need to delete 5 and retype again the value you need
so no slider, juste typing needed value inside the objects distance window/field
@solid marten Any chance to look at https://feedback.bistudio.com/T171670 ? Its a pretty significant MP bug and I'd love to have it fixed before 2.18
Fix is useful to everyone as often players spend up to a minute during loading (downloading the mission, receiving, etc.), a large chance to end up with missing vehicle
ok, no probs
Guys, I haven't been following dev as much lately (ergo why I don't annoys the devs anymore) but anyone knows if these were already fixed during my absence?:
1- Players not able to pick up items from the inventory of dead bodies in MP. (I remember work being done there some time ago but don't know if it was released already)
2- Command to disable the native add actions such as the climb ladder, heal self etc. (I vaguely remember someone starting work on something related but that was in recent times so perhaps it is still in the oven)
1- IDK
2- no
1 was reported fixed in 2.14 (https://feedback.bistudio.com/T126030, https://dev.arma3.com/post/spotrep-00111)
Found the problem, the fix may fix a few other similar cases
is support for arma tools still ongoing? ive managed to get it to throw me a series of updating base class errors in a different folder when building with a p3d axis missing verts
The game has different footstep sounds for indoor/outdoor surfaces, for now in both cases the outdoor sound will only work
This is due configuration (seems) incomplete during implemented this feature
ticket → https://feedback.bistudio.com/T181949
I have a situation in which I have a weapon that has three muzzles and is fed by three magazines. I understand that rifles only technically allow you to load two muzzles worth of ammunition, but aside from the third magazine that is loaded into the third muzzle disappearing when the weapon is dropped on the ground, it works as expected if you load the third muzzle using scripts.
Was curious as to whether this was intended behavior or whether it might be reasonable to file a FT ticket for a potential fix.
added video demonstration https://www.youtube.com/watch?v=Ma7SnmBd66E for metal surface
do you guys want more crash dumps? I have a lot of those
but don't want to spam the feedback tracker, if they are of no use
I propose a new LoadBytesFile function that would return file content in an array of number representing the byte array of the content of the file. This would be greatly appreciated to be used with DLLs. https://feedback.bistudio.com/T182005
Actually if not LoadBytesFile, a LoadFileBase64 function would be great
That probably won't get done exactly the way you want. For instance you can only have 10,000,000 elements in an array... that's less than 10Mb of data.
additional, that's gonna eat alot of memory up.
It can be a base64 string instead, but we definitely need a way to fully read a non text file
So the issue is also compounded by the fact that the internal data types will cause alot of memory to be allocated for it.
16 + (28 *10,000,000) + 48 = 280,000,064 Bytes!
That will work.
you don't want the array.
@sinful kettle can you double check my math, been a while
This poses a risk. It'll allow people to easily rip stuff out of EBOs
I think number was 32 bytes
You missed another 4 bytes of padding at the end
Is it possible to extend lightAttachObject to have the same syntax as attachTo?
If so ~~please do it ~~ I'll make a ticket
OnMissionNew runs after the variables were cleared and the terrain was switched.
As far as I can see both for same terrain or new terrain..
The only different handling is when you open Eden editor for the first time. That has a hardcoded 5 second delay till it triggers the new mission and new terrain events. Because technically on first eden open the "new mission" action isn't triggered.
Use ArmaScriptProfiler then and it'll optimize away the array creation
Ticket. Backwards compat is a concern.
But changing nil to object should be fine (boolean wouldn't because eventhandlers)
I don't see any blockers in there, its the same as for the sound commands we did recently
crashes need crashdump.
you can always just dump them in my DM instead of spamming the FT
This would be greatly appreciated to be used with DLLs.
Wha. But if you have a DLL you can just read directly from the PBO, I do that in a couple of mine
will do 🙂
I think the game crash at exit, is related to vehicle tracks.
A while ago there was a bug where when you delete the vehicle, the tracks would stay behind and never get deleted.
That caused some issues when the tracks/vehicle gets deleted at mission end. But I added a workaround for it that should be catching that issue.
But we're still seeing the crashes...
I had a repro for this crash, and with my fix I cannot repro it anymore, so it does work in that case. There must be some other edge case probably also related to the tracks.
duh 
I know how to fix the crash, and give back a indicator telling us what object is the buggy one.. Should've gotten to that earlier. Next profiling branch will have a bit better diagnostics
I'll probably intentionally crash it though, so people send us the reports :3
Indeed, i didn't think about that
Isn't the PBO locked when the game launches ?
But anyway, nevermind, i think it's better if i put my own base64 files in MY pbos, this way it won't pose any problem. Anyone who wants to use my audio engine to play audio file stored in pbos will have to encode them first, using provided encoding software
Write locked yes. But not read locked.
It'd also be better for performance if the extension can just fetch the file by itself. Instead of you having to encode it first, then Arma script having to read it, having to send it to extension, having to decode it there.
If you use C++, I have code ready to throw in-game filepath at, and get byte array out.
Hmm, that's true. I code in c# but i think i've already seen an updated nuget package to interact with pbos
Issue is gonna be to find games loaded mods, if you only load from your own mod PBO its easy to find where its at
To find game's loaded pbo's I only have C++ code, have not seen a C# version
Hmm, then in between parsing game files and encode the file first, for performances it will be better if i encode files first.
Can you also fetch EBO file contents?!
I don't think there is a library for this, and this format is not documented
Is there an option to not generate RPT File ? If there is not, I could also read the RPT file to get the list of all PBOs
-noLogs
Hmm, then i guess i have to encode first. My audio engine is not meant to replace game engine, it's just an additional engine
what file types do you want to read? all audio files? or in general terms what data/info specifically do you need for your project/system?
Only audio files with the audio engine
I was just thinking it would be great if we could also play ogg files from other pbos in the audio engine but it's not necessarily a good idea
how to you disable the native audio engine? (direct sound i believe)
i guess theoretically Dedmen could expose the audio stream (still has the EBO aspect to it, yet audio only may be less problematic)
just curious - does https://community.bistudio.com/wiki/loadFile return anything [useful] for ogg/wss/wav?
I don't need to, ijust need to adjust the volume using scripts. Basically, i made this audio engine in order to replace TFAR and TeamSpeak and also because i'm working on a multiserver experience so i needed somethink to communicate from one server to another. Now, i need to make the Acoustic Warfare Analist job, so i need to play audio files. This is how i came to the idea of allowing people to read sounds from all pbos, but well that's not much possible
I haven't tried, it would be a corrupted string anyway
thanks. you may elaborate more in #arma3_tools if you want to share more specifics or look for new ideas to overcome obstacles. its the channels extensions are usually talked about
yeah. not saying its useful for you. was more if sounds could be extracted that way already, the EBO argument would be less valid
that said i'd assume there is tools to do that already anyway for audio
I thought you guys were talking about reading the contents of a file in a PBO/EBO
Are you just trying to get the whole PBO and read its contents manually?
If so that's relatively easy
No. It encounters a null too early and returns garbage text
It can't give you the whole file via a SQF string
(I haven't checked if internally it loads the whole file in buffer tho)
http://feedback.arma3.com/view.php?id=26725 - from the 1.54 hotfix RC post. "Access Denied" :(
the problem with this window is that I tested it on two different GPUs
no?!
Yeah I thought you guys meant fetching stuff directly from the game...
Not sure, if this is a bug, but when a player joins a game with AI enabled players, selects a unit, and then exits the game, and the AI takes over the unit again, the variable name assigned to this unit in Eden (e.g. player2) ceases to exist until any other player selects this unit again and respawns. Until then, the var player2 returns nil, though the unit still exists in the world. With respawnOnStart=0 the variable can be empty long time before the unit dies and respawns. (Tested on Hosted Server dev build)
I'd like to report this spambot for plagiarism https://feedback.bistudio.com/T182000#2622951
is there a way to detect/determine when Eden(?) decides to set a dependency to a cfgPatches class?
like our case is that CBA adds cba_jam_spe (details #976228110078992456 message in ACE discord)
or some docu/insights on the system - or is it actually a bug for it to happen?
i do recall "mistakenly" set dependencies in missions/savegames in the past "years", but been a while since we/I encountered it.
might not be a problem of Eden but in the config parsing and how it determines links/dependencies
why does the crash dumps have more than one .mdpm file?
shouldnt do
but why 😐
hmm. Looks like they do sometimes.
might be from background threads or smth?
there's also more than one .rpt and it says at the bottom of those which world was run and sometimes it's VR sometimes Stratis.
so it seems different crash reports gets merged to one report
Not seen that. If there's more than one RPT then there's more than one Arma instance.
I do appear to have multiple crash dumps from one process though.
How hard would it be to make setVelocity and addForce to work with rope segments?
Lack of ways to influence them is really disappointing and being able to swing the rope through script (especially with nothing attached) is useful
first aid kit heal works incorrect if revive system enabled in MP https://feedback.bistudio.com/T182110
it might be by design?
normal game = 75%, need a medic for 100%
revive option = (if revive makes you 100%), make FAK 100% health too (prevents from requesting being shot down then revived)
I would consider it a bug if reviving does not make you 100% health
//make ANY heal a full heal
[] call bis_fnc_reviveDamageReset;```literally (`\a3\functions_f_mp_mark\revive\fn_reviveehhandleheal.sqf`, lines 18-19):
it by design!
"The necromedic. A surgeon who turned to necromancy when she lost her medical license. She can heal you, but she'll have to kill you first." (c)
I really wish there was a way to change that, perhaps so that both reviving and FAKs only go up to 75%, would there be a way? I don't know how to mod but I'd appreciate the help
maybe hooking on revive scripted events, IDK
okie
Where can I find this line of code by chance?
editor -> tools -> function viewer -> find reviveEhHandleHeal
Right
Welp
Either I'm gonna learn how to mod or ask someone experienced to help me with it
I guess nobody cares about ropes? Found lots of small issues and tweaks with them, wondering if its worth submitting
My experience is that they work as well as anything does in Arma :P
It's kinda funny that they pop up onto boxes that you're carrying, but I figured that's a consequence of the system.
Yeah, usual Arma jank and unreliability everywhere
The thing I'd like the most would be able to apply forces onto rope segments so I can script swing them, especially free hanging ones
Can't you put a dummy object on the end?
Also given the box-popping, I don't think they're really physics objects.
I can, but rope starts looking terrible, completely stretched no matter what object is attached, even if you apply velocity onto hanging object in the direction of the rope it only bends last segments instead of entire rope.
It's more like they have a start and end and then it kinda fills the rest in.
I abandoned attaching stuff to ropes, so much issues with it and went with free hanging ropes, there look much better but for some reason they're always much longer where remote
Hmm, could be, but I still wonder if there is a method in them to apply force on them to make rope swing or something, would make them look so much better with a bit of a scripting
They seem to be fairly solid entities, there should be something
I do propose one the function of my base mod that i use in pretty much all interfaces with listboxes that need to be updated https://feedback.bistudio.com/T182150
Other players cannot see visual effects (fire, light, smoke) on launcher shot in MP
https://feedback.bistudio.com/T182166
Would it be non-backwards-compatible to make serverTime return something useful (either time or diag_tickTime) in SP? Having to do ([time,serverTime] select isMultiplayer) to account for both situations isn't very fun.
See also netId.
serverTimeReal
netIdReal

It would be great if we could pass arguments to EachFrame event handler. Actually when you need to move something by yourself using setVelocity or setVelocityModelSpace you have to put it in a while loop. The problem is that it is well known that on bad performances periods spawned scripts are slower. It would be better if we could use EachFrame event.
You can.
Arguments parameter was added in 2.06:
https://community.bistudio.com/wiki/addMissionEventHandler
Ooh !! Thanks !! That's something i missed !
#arma3_troubleshooting message
Message link for more context. Some animations (Not sure if it could happen on every anims or not) does stuck sometimes if the FPS is too high (60 > ?) and on Altis (couldn't repro on VR)
this switchMove "Acts_BootKoreShootingRange_Adams"```This is the only script I run on the soldier
Could repro on VR. Same thing, but place a tree (as a Simple Object) does it
@solid marten What affects rate of ownership changes? When you take control over UAV crew unit who is both driver and group leader, it takes barely a few seconds on local server to change vehicle and group ownership to you, but sometimes up to half a minute on a dedicated server in live environment. I wonder if its some dedicated server network setting or something else?
Hint shows locality for group, vehicle, driver and gunner
This is an example of fairly quick locality change, ~10 seconds for group and ~15 seconds for vehicle
But I've seen much worse cases, half a minute and more
Tried to print getUnitsMoveInfo and it does count the moveTime very normal
What about switchMove + playMoveNow?
(or the new switchMove?)
Doesn't seem to have difference
it could be due to scheduled update, can't say until I look under the hood
Seems explosive damage against static weapons & their crew is strongly affected by setCenterOfMass. This is a bug, right?
Antistasi was using a 1m center of mass offset forever to make static weapons fall over less, and it turns out this makes them invulnerable to explosives.
Might be specific to placing the things on buildings. Not sure yet.
Will appreciate if you will, this is a major MP issue, when you take control of a remote drone and you can't do anything for 30 seconds watching it fly straight line, hopefully not into the ground.
Sadly you can't properly setOwner the vehicle to speed this up anymore, I wonder if there are other tricks to make it work quicker
Though if you ask me, its a bug and vehicle locality should change instantly just like it does when you get into driver as player.
Dedmen are you also interested in crash dumps from BI tools - specifically binarize?
12:15:06 String STRING not found
possible to provide the context for this if you have -debug active?
it seems one/a few cases arent covered yet
<language> localize <key> is this possible or does the game parse+cache only the active user language at game start(?)
Only active language is availabe
Might be a good idea to perhaps use a string table verifier script such as the ace one.
but if set fast scripted fire the effects will work, randomly 
https://www.youtube.com/watch?v=0ODkMmmLLRI
Workthought actually. Having model anim and RTM anim same time, is it possible anyhow? (I know if Ded struggle yes) Thinking of somekind of usage like RTM'ing weapon when reload anim is playing
How likely is it to reduce the frequency of key checking and battleye ? there is a big problem that due to problems with the router, it often throws players off the server
wrong channel?
looks like the router needs to get replaced instead
all the players?
This problem is 100 years old
no, the server's router and stuff
nah, it's likely going to happen in other games that aren't very tolerant to unstable internet
you have no idea what is the gravity of the situation, do you?
the problem is the users environments not on BI to fix
Do you know how to distinguish a question from a statement?
are you projecting yourself by chance?
I asked a couple of weeks ago about the BE Query Timeout issue if it is possible to add a config. Either Dwarden or Dedmen, maybe KK, replied that BE needs to adjust their API for that. I assume other things running through their API too, so I assume that other BE related things can't be changed/adjusted too.
the funny thing is that a certain player can be constantly thrown off the server, but once you restart the server, the problem is solved immediately
unfortunately its seems not happening
quite a severe issue affecting various circumstances but seem BI has little sway on BE at this stage (or costs too much money to get things done)
the origin is Arma
BE "only" handles the timeout
aka A3 is busy
BE kicks you off
workaround - extend timeout
actual fix - solve A3 being busy/unresponsive
This is the one where clients get kicked during the connection process, right?
BattlEye Query Timeout? yes
and this can't be done just with configs as mentioned in https://feedback.bistudio.com/T180455, also see answer from Dedmen
What I'm not sure about is why it's going two minutes without responding to some basic network query?
I haven't checked getRespawnVehicleInfo yet, but does it work for player units from server side?
I'd love to have a command to check if unit is respawnable to know how to handle its variables and events as they're re-added by the engine after respawn
Can happen also when you are already on the server - ie can be also caused by mission rotation via server.cfg
basically from what i understand if the client with server handshake doesnt happen in the given timeframe, "BE boots you off" (BE probably tells the server to do so)
kinda like loosing connection for too long
UAV locality is all over the place
How can a group be remote but the unit local? 
That would mean the AI itself is remote 
It feels completely random, sometimes you get drone first, sometime turret, sometimes group
Takes a lot of time on live dedi for some reason
Not sure who to bother but I'd love if this could get looked at before 2.18 as its pretty major issue
Can make all needed repros
More locality bullshit
Everything is local and controllable yet you can't shoot and just spend ammo
2024/06/23, 17:12:25 No player found for channel 2068943551488 - message ignored
10s of 1000s of such messages with server being frozen (unfreezes after 5-10 mins), I wonder what happens 🤔
Build: Stable
Version: 2.16.151618
Wish there was more info on what kind of message is was
Could give a hint what was spammed
Turrets take the longest to switch locality on live server
Does this also interact with vehicle locality where there's a driver, commander, gunner?
There's that old bug in which vehicle commanders lose control of smoke screens if they leave and re-enter the vehicle in a certain order with respect to the driver
Water explosion debris sound request, config work, sounds exist in data
https://feedback.bistudio.com/T182345
seems not due locality see https://feedback.bistudio.com/T148482#2021130
Can dev add event handler 'getIn' in base class LandVehicle with setEffectiveCommander execution as workaround?
it should be global
Ooh interesting 🤔 we've seen this behaviour when the commander was the leader though... But will try
@solid marten
Hey, could you check why isSetForDeletion flag (obj getEntityInfo 14) doesn't work for simple objects? It always returns false
will do
Thanks 👍
it could be, havent checked yet, because it is not entityAI?
i can cast to entity this particular flag I suppose
Yeah entity makes more sense
BTW for some reason when I run A3 2 of my threads are always at 100% (i7 13700), even in a completely empty mission 
I'm not sure if I'm just noticing this or if it is new tho
Latest dev
dedmen is mining bitcoins probably
2 "P" cores threads or 1 "P" core plus its hypethreaded "sumulated" thread?
I think usually it's only one thread at 100%?
Maybe closer to two if you're running with unlocked frame rate and your GPU drivers are using multithread rendering.
Dunno but iirc it was threads 8 and 9, so 1 P core and a thread of another P core I guess?! 😅
I'll check tomorrow
Running at 60 FPS (vsync) 
Checked with 3700X on stable, one thread at 100% and a few with bits and pieces.
If I disable vsync (60) and fall back to GPU driver max frame rate (120), main thread utilization drops
For me it's 2 cores (CPUs 8 and 10), both P cores 
Disabling vsync doesn't help in my case
Actually disabling and re-enabling it did help 
Now only CPU 8 is at 50%
Looks like it was a driver issue. Updating my GPU driver fixed it 
Another issue that I spotted a few days ago (I reproed it again today): during the Scuba showcase mission, I noticed that when you get close to the mines, the player starts to vibrate and it keeps going on forever until you get out of the water (you have to be in 1st person to see it; it's very noticeable when you aim). I think this is a byproduct of the compiler update. Can someone else repro this on dev or profiling branches? Also does it happen in stable?
it's just the VibroMines ™️ that the ayyys left behind after Contact
Lou gets it 😄
ahh i want to listen to that song again now
i digress
Any chance anyone has had the opportunity to look into this since December?
https://feedback.bistudio.com/T177593
I know that it was mentioned it might be some sort of shader influence that was causing this... but I'd love to get some feedback as to whether it was determined to be impossible without causing other issues. I was wondering if some of it may stem from the fact that the horizont object tends to use the _sky suffix, and if that's the primary culprit here. If that was the case, is there some chance that it could be added that a horizont object could support _ca suffix files to effectively ignore the shader changes made to _sky files?
While browsing the Wiki I found that there is a kickPlayer function for Argo. Any chance for getting that as scripting function for A3? Maybe banPlayer too? If yes, I would open a FT ticket.
I know about them, but I thought about a slightly easier method as kickPlayer is using unit as parameter
let's keep talking in #arma3_scripting message 😉
Fitting to the talk, maybe this https://feedback.bistudio.com/T125565 could also be useful
next prof/dev
There was a fix last week for BE query timeout, its on prof now.
The BE messages could be dropped and wouldn't be retried, so with packet loss that goes bad. They will now be retried so there should not be any lost BE messages anymore, and thus less query timeout
Network channel being closed can make this happen.
Game killed or crashed on players side. You should see the player loose connection and disconnect shortly after.
Yeah I'll need the repros and a big ticket explaining how it is and how it should be.
Its unlikely for 2.18 though
Its an error, errors already have context. Also shouldn't this create a crashdump?
But showcases are part of QA tests that should've been noticed.
Mh you can request artillery in the scuba showcase? why 😄 I never played it before.
Is this close enough? I don't see vibration
its in 2 rpt from BI QA - no error context in rpt despite -debug used. can DM/link to rpt if useful
Original output filename: Arma3Retail_DX11_x64
Exe timestamp: 2024/03/21 14:45:00Type: Public
Build: Stable
Version: 2.16.151618
RPT isn't useful.
If there is no context, then the game just doesn't know. And we can't add context without knowing where to add it
Errors in "numberOfWindows" config value of Wooden House (Small, v1)
https://feedback.bistudio.com/T182436
Ah I got it in retail build
I was gonna post a video but I guess no need anymore 😅
bzzbzbzbz
So is it new?
This is fun.
When I debug step through, it doesn't happen.
At 200fps, its like in the video.
At 20fps...
Didn't test old version yet
Maybe you got a winter date set so water is cold 
This shouldn't have entire server halted for everyone though, isn't it? This ended up with whole server not sending any messages to players for like 10 minutes. Then it resumed and game continued fine 🤔
hitting the bounding box instead of the model?
Turns out it expects left operand positions to be ASL too
I think its better to just bite the bullet and fix the command to work with AGL just like normal inAreaArray