#arma3_feedback_tracker

1 messages · Page 3 of 1

dreamy bane
#

great

untold sky
#

yep it does, it says empty.p3d but I assume thats correct

azure crane
daring wagon
#

@untold sky You might wanna shorten that translation.

20221119173033_1.jpg
sinful kettle
#

command info errors:
ctrlAt: says returns nothing, should be ctrl
ctrlSetURLOverlayMode: prefer steam overlay is enabled -> if enabled
displayUniqueName:

  1. return value is nothing. should be string.
  2. ...dusplay... -> display
sinful kettle
#

also I seem to have missed something. why is the Arma icon in the main menu gold? think_turtle

alpine tulip
alpine tulip
azure crane
sinful kettle
desert trench
gray wharf
#

this is not feedback tracker related, more wiki/config

fluid meteor
azure crane
sinful kettle
gray wharf
untold sky
desert trench
untold sky
#

probably

desert trench
#

i see. well fingers crossed

alpine tulip
desert trench
#

Is it worth to report that Eden swaps certain attributes stored in the sqm upon each save? (makes diffing a lot more annoying)

devout wave
#

The HitPart EH has a lot of parameters that would be super useful to also have available in the HandleDamage EH. Speaking with the confidence of ignorance, I'm hopeful it might be relatively simple to copy them over. https://feedback.bistudio.com/T79970

west onyx
#

i get a lot of this spam when executing "doSuppressiveFire"

# 
21:58:40 Ref to nonnetwork object 283c208a080# 1698235: <no shape>
21:58:40 Ref to nonnetwork object 283c26c7580# 1698234: <no shape>
21:58:40 Ref to nonnetwork object 283c26c4b80# 1698233: <no shape>
21:58:40 Ref to nonnetwork object 283c26c7580# 1698234: <no shape>```
deep widget
restive bone
gray wharf
#

(anymore 😋)

restive bone
#

did you just book out the escorts lou?

gray wharf
#

all of them at once, so the ad is pointless now - removed

#

ooooooffffff

image.png
gray wharf
#

@deep widget / @untold sky when dealing a spam ticket, let's
1- close the ticket to resolved ← so the search for "open" documents doesn't find it
2- remove project
3- set visibility to administrator
4- disable the related user
thanks 🙂

west onyx
#

some more invisible targets would be cool

#

example, attaching an invisible Air target to a missile/bomb ... allowing anti-air assets to attack the missile

devout wave
real bronze
deft girder
#

did Bi changes anything related to shaders in the 2.10 update ?
am having this:

22:25:11 Error: 80004005 in D3DXCompileShader while compiling the PSSpecularAlpha shader
22:25:11 error X3501: 'PSSpecularAlpha': entrypoint not found
game is not working for me, though reverting to 2.08 profile binarized exes seems to make the game starts normally

desert trench
#

@deft girder any mods loaded?

lean mica
#

Maybe related to TI overhaul in 2.10

deft girder
deft girder
gray wharf
#

thermal imaging

deft girder
#

did other people reported similar issue ?

azure crane
#

you're probably the first one

regal nimbus
#

I recall issues with some Intel IGPs in the run-up to release but those were fixed.

#

What IGP is this?

rose cloak
deft girder
deft girder
#

20:06:49 Error: 80004005 in D3DXCompileShader while compiling the PSSpecularAlpha shader
20:06:49 error X3501: 'PSSpecularAlpha': entrypoint not found
20:06:49 ErrorMessage: Error compiling shader PSSpecularAlpha
20:06:49 Application terminated intentionally
ErrorMessage: Error compiling shader PSSpecularAlpha

Exception code: 0000DEAD at 2745CD29
cpu: Intel(R) Core(TM) i3-7100 CPU @ 3.90GHz
graphics: D3D11, Device: Intel(R) HD Graphics 630
resolution: 1366x768x32
Version 2.10.149818
Fault time: 2022/11/29 20:06:56
Fault address: 2745CD29 01:0002BD29ll C:\Windows\System32\KERNELBASE.dll
file:
world: Stratis
Prev. code bytes: BA BB 06 00 48 8D 4C 24 20 48 FF 15 17 9B 19 00
Fault code bytes: 0F 1F 44 00 00 48 8B 8C 24 C0 00 00 00 48 33 CC

#

does anyone knows the causes of 0x0000DEAD, its not listed in wiki

azure crane
#

it's not listed because it's "unknown software exception"

#

I've seen numerous solutions to the DEAD code, but none of them were about shader issues

#

have you tried reinstalling drivers or upgrading/downgrading them?

#

maybe your IGPU is too weak clueless

deft girder
#

i tried everything, but how my igpu works in v2.08 and not on v2.10 are there any major changes in shaders ?
right now am playing the latest version of the game using 2.08 .exes, despite having some eden rpt errors.
right now am suspecting either driver incompatibility or windows issues.
i will try and investigate further with this...

azure crane
outer frigate
#

iirc there were a handful of issues around intel gpus after 2.10

outer frigate
unborn acorn
alpine tulip
#

Things my brain thinks for our QoL:

  • If a player enables a Mod which requires an Optional DLC such as a CDLC on Launcher, it suggests to load the CDLC, or if the player doesn't own it, suggests to sub/load the compat file
  • The pulldown box for Platform (64-bit, 32-bit) option should have Diag exe option if is in Dev Branch
  • A game option to tilt camera depends on the vehicle's roll if is TPP and on plane/heli
  • Modded game settings like Modded Keybinds ploxxxx (https://feedback.bistudio.com/T159838)

(I think I forgot something else I wanted to suggest 🤔 )

#

Oh and also - if this is a good idea, please implement officially. Or, at least an EH for a tree control's “search has performed”

alpine tulip
#

Oh I just recalled what I wanted. What about this idea: a way to lock/make it undeletable for EH commands to prevent from a Mod conflict when some Mod maker uses removeAllMissionEventHandlers or something unreasonably

#

(IKR the answer could be “just slap the faulty Mod author and never use the Mod”)

alpine tulip
livid sleet
alpine tulip
#

Mmmmaybe

desert trench
#

is there a chance to see this fixed? (aka worth an A3 FT)
or is it more of a limitation/too complex to render properly

#

would be handy as unit icons on 2d map would be easier to make out if the map background has similar colors

desert trench
#

KK if you are up to turn your scripted solution into engine 🙏

untold sky
untold sky
untold sky
devout wave
#

"I don't see why that can't just be rotated"
click
game deletes itself

daring wagon
untold sky
#

Ye already known :x they have been there for quite a while

daring wagon
#

ok

real bronze
alpine tulip
#

Hmm, maybe I'm crazy enough but... weaponInfoType display doesn't get destroyed properly when I Team Switch?

#

Context: I have unit A with a weapon with weaponInfoType X. I switch my player to unit B with a weapon with weaponInfoType Y. The display X doesn't get destroyed, it doesn't return displayNull

gray wharf
#

isn't the display simply reused?

alpine tulip
#

Nop

#

Even if I switch to my pistol which has display Z, will be destroyed

gray wharf
#

welp, memleak

alpine tulip
#

If I hop into a vehicle, will be destroyed as well

#

So I could assume that this is a bug

#

Only tried in SP tho...

alpine tulip
#

Looks it fails to destroy display when switch binocular > rifle

alpine tulip
#

(the reason why I report this is I'm making HUD related script)

limpid rune
#

Yeah if you drop the weapon on the ground it stays too

#

its only when you replace it Im pretty sure

#

like if you have a rifle with a custom weaponinfotype it will only go null when you take another rifle

alpine tulip
#

Sorry for bump and ping Ded, but... is it that expensive for a GPU/rendering? Does retextureable mag proxy have such a bad performance?

#

(I'm comparing it with mag proxy since I think both are really similar feature, correct me if both are completely different system/method)

untold sky
#

I don't know but I remember all the extra handling that is there for magazines and I don't want to copy that for the other parts

alpine tulip
#

But me wants Understandable yeh

untold sky
#

You just need to get me to actually look into it someday instead of guessing about it, currently I don't wanna

gray wharf
#

*whipcrack*

dreamy bane
#

what troubles me is that when there are loose files in arma main folder and I turn the filepatching off, those files still get read. I'm using symbolic links btw

#

is that normal?

#

I have to delete the folder/symlink to prevent those files from being read

untold sky
#

what u mean read? accessed by game or actually being applied in-game?

#

then I guess filePatching is not actually off

dreamy bane
#

applied...

#

its off...

#

well maybe not off from server, I will need to check again

untold sky
#

isFilePatching command in debug console returns false?

dreamy bane
untold sky
#

ah isFilePatchingEnabled that is it

dreamy bane
#

well I cant reproduce the problem now, maybe i just did something stupid

dreamy bane
#

my apologies 🙂

real bronze
desert trench
#

ThingX 20a61497580# 1976630: dummyweapon.p3d was falling under terrain! Position corrected to ground [4993.69, 46.59, 1626.79]. Disable simulation to keep object under terrain.

#

is there a FT ticket about that?

#

from what i understand the problem is that weaponholder simulated dont get simulated when no player is close enough - as such they are floating/their position gets reset once a player gets close
this rpt message should be connected to that

#

aka he floating weapons getting dropped to the ground should be due to this

desert trench
#

Tried to create group while group limit is reached on side Game Logic
can this please looked into early next year

people already identified the bad coding design in the function library along with a possible fix in

fn_initModules creates a group for each cfgfactionclasses
https://feedback.bistudio.com/T167804

its bad if you do have to create many logics or game modes do (ie BI combat patrol and escape)

real bronze
west onyx
#

I wonder is it possible to toggle an entity between agent and unit state

#

liek createUnit, then "setagent", or createagent and then setunit/setagent false

#

_unit setagent [_group,false]

#

the entity remains the same but the "brains" turned on/off

#

would allow some pretty big perf improvements

sleek scaffold
#

Also would be great if when creating a unit on the eden editor it has a tick to make them agents instead of regular units for getting NPCs with better performance that do not consume groups slots.

gray wharf
#

my guess would be "hard no" for both (although the latter might be more realistic)

sleek scaffold
#

Placing NPC units as agents can be quite difficult for most mission makers because currently it can only be done via scripts. With some EDEN Editor functionality setting the correct direction, uniform and placement of the NPC units would be far easier, specially inside buildings and also adding Addactions to them via the EDEN editor would be far easier. Identifying NPCs via script alone can also be quite a headache that is why most mission makers use regular units as NPCs and Civilians while the agent option is far better performance wise 🙂

gray wharf
#

I never said it would be unhelpful.

sleek scaffold
sleek scaffold
# west onyx the entity remains the same but the "brains" turned on/off

Having a script command to seamlessly turn agents into regular units and the other way around would be quite awesome too! When my NPC agent guards are under attack I have to do a pretty complex script to turn them into regular units so they can respond to the agression properly... Such command would make that far easier indeed

gray wharf
#

yeah, well, it is still not a suggestions channel
I get it that we can ask if something something can be sneaked in the next update, but let's not abuse that here

real bronze
sinful kettle
devout wave
sinful kettle
#

especially considering that the mod came out a few days ago

untold sky
#

Close, not a bug

#

Don't even know what it's about, but i know we won't do whatever it is

sinful kettle
#

oh it's a DayZ mod meowsweats (and a DayZ ticket)

west onyx
#

another easy feature id bribe a dev for: invisible air target entity. we could attach it to cruise missiles, allowing people/turrets to shoot down missiles and bombs

limpid rune
#

...Has anyone noticed that mines don't draw shadows? They seem to when you place them in the editor but ingame/placing them yourself they don't have a shadow

#

I can't seem to find a ticket mentioning it

image.png
untold sky
#

Won't fix anyways

fading tulip
#

is the 'water visualization' glitch when using setTerrainHeight known? If I sink an entire area, eventually there are grids of water that don't render unless up close, while other ones render from afar just fine.

example: https://imgur.com/a/AJinPFM
you can see the shore water rvmats and everything shows fine, just not the water itself

fading tulip
gray wharf
#

I sense a "won't fix" but you never know

untold sky
#

Won't fix. As the wiki page for the command says

fading tulip
#

the wiki mentions that for "extreme changes", is 2-6m difference considered extreme? also, everything in those pictures is being affected by the same amount, yet that visual glitch doesn't happen across the board to the entirety of the affected terrain, it only happens to a whole terrain grid itself it seems, not by cells, it may be something else altogether

deep widget
#

is 2-6m difference considered extreme?
i guess that depends on how big of an area your sinking by 2-6m

fading tulip
# deep widget > is 2-6m difference considered extreme? i guess that depends on how big of an a...

that seems like it indeed may be the problem. I did keep testing my script and the same glitch would happen even when the height difference was totally minimal.
https://imgur.com/a/6qD7KpK

I will keep testing. I wonder if this happens due to 'total terrain' affected or maybe if dividing the script into parts would keep it from happening. I can confirm it still happens with smaller affected areas, it's always the same exact 'grids' that have the issue, also happens in Livonia aswell. Maybe it happens if the terrain 'grid' didn't originally have water on it?

west onyx
#

can anyone confirm that the PCML launcher alternatve fire mode (overfly attack) seems to be not working? Presumably the overfly mode is for attacking armor, but the damage vs armor is minimal in overfly mode

desert trench
#

14:28:39 No alive in 10000 ms, exceeded timeout of 10000 ms

#

is there a good reason to have this logged multiple times per frame or per second?

#

shouldnt this restart the count - so only happen every 10s?

desert trench
#

also is it intended those freeze logs are not cleaned up?

#

for the average guy not manually deleting those, the 30-60 MB each can take up quite some space over time

regal nimbus
#

@west onyx That's my experience. Flight path looks plausible but it doesn't do much damage.

devout wave
#

I haven't exhaustively tested it but I've got a feeling the penetrator submunition tends to get fired at a weird angle and miss/mostly miss the target

west onyx
devout wave
#

I spent half of last week having a very bad time with missile trajectories for my APS script, I don't wish to do it again :U

west onyx
#

hard to get it working very well in MP, since usually either the target or the projectile are not local

#

i wish id done area intersection rather than surface intersection

devout wave
real bronze
devout wave
#

It's the wrong channel to get into too much detail but that is kind of how it ended up working; everyone does local checks on the list of APS vehicles and whichever one ends up being local to the shooter goes ahead with the interception. I haven't got it into a full-scale mission yet but early MP test results indicate good reliability.

west onyx
#

"getOrDefaultCall" cant come soon enough!

untold sky
harsh dragon
#

AI using the CAR-95-1 appear to only use semi auto instead of automatic fire, I assume this is not intended behavior since pacific CSAT autoriflemen have this weapon.
https://feedback.bistudio.com/T169426

desert trench
#

is it a known bug that when you tab out and back in while in Eden, you may perform an action (move an unit/WP/etc) by mistake, but UNDO doesnt pick that up and thus can only be reverted by reloading the mission (if you had not made other unsaved changes)

#

(it seems the tabbing in/out can make you select something which wasnt the case before tabbing out)

#

the only way to stop this if you try to cancel the action (by clicking somewhere that isnt a valid target or RMB to abort)

sinful kettle
#

You can also press Alt again

#

When you Alt+Tab the Alt key gets stuck

dreamy bane
desert trench
#

is it a bug that forgetTarget doesnt remove entities/enemies from targets [true, 300]; // enemy targets in 300m?

#

knowsAbout is set to 0

desert trench
#

(player targetsQuery [objNull, west, "", [], 10]) apply {[player distance (_x#1),_x]}

#

[[41.5477,[0.0675484,B Alpha 4-1:1,WEST,"LIB_GER_SquadLead",[7546.53,11031.1],148.042]],[1e+010,[0.000311428,<NULL-object>,GUER,"LIB_P47",[7503.7,10780],321.101]],[1314.45,[6.63556e-005,B Bravo 4-6:4,UNKNOWN,"StaticMortar",[8656.68,10349.4],1507.03]

#
  1. reports also non west
  2. with target age >10
west onyx
#

yea dont use targetsquery

#

they added "targets" after i complained about targetsquery 😄

desert trench
#

so there is actually no sqf command to return "known enemies/targets"?
you always need to filter on-top via knowsAbout, targetKnowledge, etc

west onyx
#

hmm whats wrong with "targets"?

desert trench
#

at last once a group has known about an unit/group once

west onyx
#

yea dunno

#

seems more like a forgettarget bug

#

it has never worked well, IMO its only real use is singleplayer in a eachframe loop to simulate stealth

desert trench
#

hm now forgetTarget worked to remove the enemy from the targets list

#

underlying issue i am trying to solve is to get an AI lead group out of AUTOCOMBAT before the 5 minutes hardcoded timeout since last known target/engagement(?)

#

it just slows down AI way too much in case they dont eliminate all known threats

#

asked a couple of times for a way to manipulate that way too long timeout

#

so atm you can either disable AC completely, periodically, or try to determine yourself via sqf the last known threat seen/engagement (very heavy to compute..)

west onyx
#

for that id just do a new group for them

#

(units _grp) joinsilent (creategroup [side _grp,true])

#

wipes their memory

desert trench
#

another "funny issue" - hideObject[Global] seems to keep collision and pathfinding for AI at times

#

even locally (SP/hosted)

dreamy bane
west onyx
#

@untold sky with user action event handlers, is it possible to block expected result with a return?

#

sort of like how we can block in "keydown" event with true return

#

user action events much cleaner than ui keydown event, however without ability to block, ui keydown still needed

desert trench
sinful kettle
west onyx
#

there is a bug which causes serious performance degradation, should be easy fix and repro. the S-300 SAM launchers when fired up into the sky, the missile keeps going forever

#

causes serious stuttering

gray wharf
#

ticket

solid marten
#

time to live not set in config?

dreamy bane
west onyx
#

_cursorObject isKindOfAny ['LandVehicle','Ship','Air'] would be nice

#

instead of conditions like this

# 
(_cursorObject isKindOf 'Ship') || 
(_cursorObject isKindOf 'Air'))```
west onyx
#

just put down the CSAT S-750 and fire a few times straight up. in the editor i guess the stuttering goes away after the 55 seconds ... in multiplayer it seems to persist

dreamy bane
rose cloak
#

sqf loop is strictly worse than in-engine check

dreamy bane
#

yeah slower , but less lines like this

fluid meteor
# 
['LandVehicle','Ship','Air'] findIf {cursorObject isKindOf _x } != -1
fluid meteor
untold sky
#

"less lines"
SQF doesn't need any linebreaks, so just remove them and put everything into one line

dreamy bane
#

yeah i just meant instead of many ORs and stuff you can do what was shown above

solid marten
west onyx
#

yep

#

so i guess its 55 seconds of serious stuttering when its being used, esp in MP

west onyx
desert trench
#

why is this MP and playableUnits only? doesnt it just grab this from the config/cfgRoles?

alpine tulip
#

Speaking of, commands like sizeOf, getDLCInfo (IIRC) or such that require actual object, if it can read not the entire mesh of the given p3d but only headers of them, it is nice

desert trench
#

it seems there is no way to get the vehicle position name via sqf directly (assignedVehicleRole is close, but only returns the "technical" term and isnt actual position but assigned as it says)
one has to query that from gunnerName from config via turret index. also cargo positions with FFV is quite nasty

#

any chance for a sqf command? the engine does have that info as it can get displayed

#

(alternatively a function to make it available for everyone easily)

solid marten
solid marten
desert trench
# solid marten fullCrew?

what about it? i am using it, but it doesnt provide the vehicle position names - just the technical terms and turret path to grab from config space as said

solid marten
desert trench
#

as said its the technical terms

#

gunnerName from config space terms what is actually shown

alpine tulip
#

Haaaaaaaa

#

I've just spent hours just to realize that diag exe doesn't update texture resolution properly

#

I've make a change a texture 4k to 2k on-the-fly, it glitches so couldn't show it properly, until rename the texture or restart game

solid marten
gray wharf
#

maybe the position's "displayName" (e.g copilot, cargo gunner, etc)

solid marten
#

left door gunner?

#

so add displayName to fullCrew? but it is localised most likely

gray wharf
#

I wonder if a separate command would not be better? e.g command ["role", index] returns string

solid marten
#

Hwy?

gray wharf
#

(I'll assume "why" ^^) aren't there other commands using role+index? if not then I'm not against

desert trench
#

any chance for a sqf command? the engine does have that info as it can get displayed
(alternatively a function to make it available for everyone easily)

I brought it up as i searched BIF, reddit, etc and there was no solution around. Presumably as people struggle to understand working with turrets, or let alone know about gunnerName to define the display name. Hence I asked if a SQF command can be added, or a function [can be provided] (as i am not sure if this is still within A3 updates possibilities - so far its been essentially only fixes to A3 data AFAIK).

#

[in A3 data only arsenal uses/resolves gunnerName]

solid marten
#

so is it gunnerName only you are after?

desert trench
# 
        case "driver":    {localize "$STR_POSITION_DRIVER"};
        case "turret":
        {
            getText ([typeOf _vehicle,_turretPath] call BIS_fnc_turretConfig >> "gunnerName")
        };
        default {localize "$STR_GETIN_POS_PASSENGER"};
#

i have no problem to write my own. i brought it up for general utility/to make it more accessible

solid marten
# desert trench ```sqf case "driver": {localize "$STR_POSITION_DRIVER"}; case...

Something like this?

[B Alpha 1-1:2,"gunner",-1,[0],false,B Alpha 1-1:2,"$STR_POSITION_GUNNER"],
[B Alpha 1-1:3,"turret",-1,[1],false,B Alpha 1-1:3,"$STR_POSITION_COMMANDER"],
[<NULL-object>,"turret",0,[2],true,<NULL-object>,"$STR_A3_TURRETS_CARGOTURRET_L2"],
[<NULL-object>,"turret",1,[3],true,<NULL-object>,"$STR_A3_TURRETS_CARGOTURRET_R2"]]```
west onyx
#

the manual action is unnecessary, or maybe legacy

#

if you are already receiving a service, you have a few moments to select something manually

#

but when you get into the radius something will start

solid marten
#

there is auto supply that is always on

west onyx
#

you know more than i do 😄

solid marten
#

now i do

west onyx
#

the "broken gameplay" stuff occurs when player can sit beside an ammo truck and receive infinite ammo. with rearm detection we can script our own resources/depletion/etc

solid marten
#

I added allowService command to deal with it

#

doesn't happen with manual supply only auto

west onyx
#

yea nice

solid marten
west onyx
#

so the manual was kind of redundant?

#

i imagine they created it as manual and then to patch an issue (AI related maybe) they applied auto

solid marten
#

tbh i'd prefer manual and disabling auto

west onyx
#

yes, but people who play the game for years expect things to work in a certain way.

#

maybe check to see that disabling Auto doesn't affect the AI auto ability

#

we still need AI to be able to apply services

#

also what about terrain services like fuel pumps. will the allowService command work on them (are they streamed entities?)

west onyx
#

odd bug (VTOL throttle)

real bronze
#

yep, been around for ages, and only when you're not the group leader

west onyx
#

to block the bug, simply "dostop player" when in vtol pilot seat

#

so group leader AI seems to retain some control over player. occurs with player group leader too

#

my guess is theres just an internal "isplayer" check missing somewhere

solid marten
solid marten
west onyx
#

I believe only VTOL. if it happens in others it is not noticeable

gray wharf
#

the Y-32 Xi'an is VTOL
VTOL means Vertical Take-Off/Landing

west onyx
#

yea sorry all VTOL incl xian

desert trench
real bronze
solid marten
# west onyx thoughts? https://feedback.bistudio.com/T169481

I need some examples to test. Plus this will not be identical check as you do with objects because it takes into account distance, surface, mass (which you can change from default model mass), the sling load can be disabled on the object just as well

west onyx
#

ok, if not practical dont worry about it 🙂

west onyx
#

if there is focus on SQF at the moment, is an AI "AttackTargetChanged" event possible? allows mission makers to do some cool stuff based on what the AI is attacking (for example switching from APFSDS to HE), or modulating aiming accuracy based on distance to target

#

dedmen says no 😦

#

we have the "getAttackTarget" command which is great

untold sky
#

I most likely said no because of the effort, not of the possible

west onyx
#

ill settle for something like this 😄

#

Actually a suite of invisible target vehicles would be good, with varying AI ammo usage flags

#

There is one invisible vehicle target but it has collision so it actually blocks projectiles fired at it

#

but the invisible air target would be the most useful of them

#

the best we can do at the moment is attach a darter drone to things

solid marten
alpine tulip
#

Nice, it would make some DLC detection scripts faster and work without reading the entire mesh as well

solid marten
alpine tulip
solid marten
#

yes

west onyx
# 
"B_Heli_Light_01_F" canSlingLoad "B_Heli_Light_01_F"```
# 
"" canSlingLoad "C_Kart_01_F"```
# 
"B_Heli_Light_01_F" canSlingLoad ""```
#

@alpine tulip could it be useful to support empty "" as well? a fast check to see if an entity is able to sling load at all, or if an entity can be sling loaded at all

#

"" canSlingLoad "B_MBT_01_F" // false bc it cant be sling loaded at all by anything

alpine tulip
#

Maybe, if either left or right argument is "" will simply return false I guess, though

west onyx
#

"" canSlingLoad "C_Kart_01_F" // true since it is technically able to be sling loaded

alpine tulip
#

Ah, you would see that if is a slingloadable object?

west onyx
alpine tulip
#

I see, sounds reasonable but not sure what KK would think

west onyx
#

yea we arent the ones that have to do the work/testing 😄

devout wave
desert trench
desert trench
desert trench
#

either setUnloadInCombat [false,false] is not always working or canUnloadInCombat is bugged

#

testing in hosted MP from Eden (so locality doesnt matter)

#

AI is stuck in a get-out-get-in loop at a halftrack

#

canUnloadInCombat always returns true no matter what i set with setUnloadInCombat

#

[cursorTarget,simulationEnabled cursorTarget,isObjectHidden cursorTarget,unitReady cursorTarget,stopped cursorTarget,canUnloadInCombat cursorTarget,getAttackTarget cursorTarget,assignedTarget cursorTarget]

[westtruck2,true,false,false,false,true,<NULL-object>,<NULL-object>]

#

tried also with driver/effectiveCommander in case its not the vehicle

#

pretty sure the command works in general tho

#

so either its something asset specific (ie the config parameter overwrites the sqf command) or something situation specific

#

the given vehicle has unloadInCombat = 1;

solid marten
#

how is that settter has 2 args and getter only one value return?

solid marten
desert trench
#

i tried with just bool for the setter, but it doesnt like that

#

had actually made a ticket on that a while back

#

will do some more testing tomorrow

#

pretty sure its either related to the config definition or "combat state", or a mix of both

#

might be "sqf" fighting "config" and thus you get the get-out-get-in loop

#

another element could be the driver and the cargo being separate groups - shouldnt affect the sqf commands/the getter, but may break thee functionality

solid marten
#

canunloadincombat returns config value, nothing to do with setUnloadInCombat

#

So here how it works, if config unloadInCombat == true then the engine checks if setUnloadInCombat is set to true as well, and if so it issues orderGetIn false. Thats all it does. If unloadInCombat == false, the check is ignored and if setUnloadInCombat false, then it is also ignored

#

@desert trench now explain to me what doesn't work

outer frigate
#

any way to stop magazines being loaded into guns upon them being added from an engine pov? i.e being able to do

player addWeapon "myweapon";
player addMagazine "myweaponmagazine";``` to load `"myweaponmagazine"` into `"myweapon"`? Currently if player's inventory is full `"myweaponmagazine"` would never be added with current method where magazine must precede gun
rose cloak
#

you can link the magazine into the weapon if you want to add it after the weapon, iirc

outer frigate
#

replaces the magazine does it not?

rose cloak
#

yes. Am i misreading the question?

outer frigate
#

wait no that command doesnt work for partial magazines like addMagazine iirc

rose cloak
outer frigate
# rose cloak yes. Am i misreading the question?

believe so. atm addMagazine must precede addWeapon to ensure that the weapon added has the magazine added loaded. however if there's no room for said magazine it will never be added to the inventory

rose cloak
#

well addWeapon to add the weapon (with default magazine), then addWeaponItem (or whatever) to shove the needed magazine into the gun (which automatically gets rid of the original one)

outer frigate
#

oh wait youre saying i can just take the original one and readd it to the inventory after replacement?

rose cloak
#

probably

outer frigate
#

i mean, would work -- though would certainly be annoying vs just swapping the magazine addition around

rose cloak
#

i seem to remember something like that in "Fired" EH to just replace the fired mag with fresh one instantly. With hilarious enough results (and without my inventory getting overflown by old mags) 🤣

outer frigate
#

i'll shift to #arma3_scripting as to not flood this channel but is still annoying nevertheless

west onyx
solid marten
alpine tulip
#

Huh... is this a GUI element tank?

outer frigate
#

its 2035 tech for morale boosting

solid marten
#

CT_OBJECT

image.png
alpine tulip
#

I've been tried to import a gun model into the game, and found out that Shadow Stencil LOD doesn't calculated properly if there is a hidden sections via hiddenSelection and more than one same models in the screen?
This is actually a part of experimental project so this AK-74M has a lot of components like Grenade Launcher or AKS-74 variant in it into one p3d, but hidden - of course shadows are supposed to be hidden as well but somehow it doesn't work properly like the screenshots suggest

20230101133028_1.jpg 20230101133034_1.jpg
alpine tulip
#

Or, is it even reasonable to use section and hiddenSelection to hide a part of a shadow LOD?

solid marten
#

@sterile nova ^^^

desert trench
# solid marten So here how it works, if config unloadInCombat == true then the engine checks if...

thanks a lot - very good info and thanks for the new getter too 👍

from my notes (https://community.bistudio.com/wiki/AI_Group_Vehicle_Management) orderGetIn false will not unassign the units from the vehicle - which in general make sense as AI should board the the transport again after the threat has been taken care of.

This leaves the questions what causes the get-in-get-out loop. Is there some internal flag to makes AI get out of vehicles? ie when stuck/not moving for a certain time?

#

The specific situation i am seeing it atm is a Halftrack running into fences/stone walls (crushing them, but slowing down or stopping) - the AI cargo jumps out, only 1-2, and almost immediately back in - this repeats indefinitely and the transport waits (from what i can tell) until the whole groups is back in.

solid marten
#

We can guess all day long, but if you give me vanilla repro I can get to the bottom of it

desert trench
#

@solid marten tested a bunch of things but different terrain and vanilla is quite a change. from what i can tell the eject requires the vehicle to get damaged and a known threat - can you confirm that?

#

(if its related to unloadInCombat in the first place)

solid marten
#

Can you provide a repro?

devout wave
#

Repros are for cowards. Real coders just change things that look like they might be related until the problem is either solved or superseded by a worse one

rose cloak
#

yaaay, i'm a real coder

desert trench
#

is there some objects AI in cars/trucks from vanilla gets "stuck" on on Stratis/Altis/Malden? (aside from stones and low walls)

west onyx
#

bridges

regal nimbus
#

From vanilla, only the quadbike routinely sticks on bridges.

#

Units hopping in and out of trucks when they're the sole group in the vehicle and there's some shooting going on is pretty common though.

#

I worked around that by putting the driver in a different group.

#

I don't recall them doing it out of combat.

desert trench
#

meh no success getting a reliable repro with vanilla so far

#

Placing a single CYCLE waypoint will do nothing, at least one MOVE waypoint has to be placed to start with. The CYCLE waypoint can be placed anywhere on the map, the unit will ignore its position. When the unit reaches the MOVE waypoint preceding the CYCLE waypoint. it will return back to its starting position then repeat the cycle endlessly. If additional waypoints are added after CYCLE waypoint, they are ignored. In short, the CYCLE waypoint is just an instruction to the unit to go back to the starting position.

is this true only for MOVE WP, or CYCLE always restarts the WP string if the last one is reached and finished (aka also with SAD/HOLD/GUARD/etc) ?

vale wasp
#

I think CYCLE can start from starting pos or any WP when placed in editor, it shows a black line to which position is nearest. AI will repeat all or part of WP string. But it needs to finish the previous WPs, don't know if e.g. HOLD finishes by itself

desert trench
#

most of these WPs have a timeout - not sure about HOLD atm either tho

#

bit of a shame cycle always restarts the whole WP chain - would have been nice to sync/set how far it goes back
as alternatively one probably can just delete any WPs before, or do a scripted cycle behavior with your own timeout and reset behavior

gray wharf
desert trench
#

hm above comment was just added by KK a few days ago in the BIKI

#

either "starting point" needs to be clarified, or the Eden visualization with closest WP isnt correct

gray wharf
#

that and/or behaviour changed at one point

west onyx
#

@solid marten is "nearSimpleObjects" feasible?

#

we have "nearEntities" , "nearestObjects" , "nearestTerrainObjects"

#

but getting nearby simple objects requires an SQF loop over nearestobjects, or an "inAreaArray" loop

#

(allSimpleObjects []) inAreaArray [_position,_radius,_radius,0,FALSE] is the current method

#

an idea form would be similar to "nearestObjects" but which also accepts models in search, so it effectively captures 'super simple objects' too

# 
nearestSimpleObjects [
    player,
    [
        "B_APC_Wheeled_01_cannon_F",
        "apc_wheeled_01_cannon_f.p3d"
    ],
    100
];```
#

and this one might be nice for some modders

#

"incase" and "findcase" or something ...

west onyx
#

and a getter for "settowparent"

devout wave
west onyx
#

@devout wave have you done much with the settowparent stuff

#

?

#

fiddling with it the past couple days, naturally the players arent satisfied with 1 combo and want to make conga lines of vehicles

devout wave
west onyx
#

one Q i have, how did you manage the ability for helicopters to sling load the "tow parent"?

#

like clearly i want to prevent the lead vehicle from being sling loaded

#

but i also dont want a cumbersome monitor thread running on server

#

and pain in the ass that "ropebreak"/"ropeattach" events dont trigger for the child vehicle

devout wave
#

I didn't consider it at all tbh. In my usecase it's very unlikely for the mission to include both towing and slingloading. If someone did try to do that I'm not sure what would happen - if they were able to pick up the lead vehicle and it resulted in physics disaster, I'd probably just write it off as getting what they deserved for trying it

west onyx
#

i put my stuff on a live server and within an hour there was satanic behavior occurring 😂

devout wave
#

Yeah, I work in a fairly controlled mission environment where players are restricted in terms of what kind of BS they can get up to. If I don't want to deal with tow/slingload interactions I just don't put them both in the same mission.
In the absence of getTowParent, I save some variables on the involved vehicles with info about their towing status. I would probably approach it by putting a userAction EH on everyone to detect when they slingload something, that EH checks their getSlingLoad, and if the slingloaded vehicle has a tow status, then immediately run setSlingLoad objNull to force them to drop it.

west onyx
#

did you support conga lines/trains?

#

might take to DM, dont want to spam this channel

west onyx
#

@untold sky what happens when we execute vehicle settowparent objnull is that the same intended effect as something like heli setslingload objnull to cancel the relationship

gray wharf
#

… try?

alpine tulip
#

catch!

west onyx
# gray wharf … try?

ive encountered situations where the towed vehicle becomes permanently uncontrollable by the player who executed "settowparent", but perfectly controllable by others afterward

desert trench
# 
        _x addEventHandler ["Local",
        {
            diag_log ["Local change",_this,_entity,_isLocal];

            params [["_entity", objNull,[objNull]], ["_isLocal", false, [false, objNull]]];```

> 10:12:31 ["Local change",[charlie_u5,"amovpercmstpsraswrfldnon"],any,any]
> 10:12:31 Error in expression <Local change",_this,_entity,_isLocal];
> 
> params [["_entity", objNull,[objNull]], >
> 10:12:31   Error position: <params [["_entity", objNull,[objNull]], >
> 10:12:31   Error Params: Type String, expected Bool,Object
#

can anyone explain how _this can have an animation for the local EH?

#

i had also faulty data passed on to HD a few times, but people said it cannot be

#

} forEach playableUnits;

#

the local bugging out happens with hosted at the start

#

the locality change apparently happens from moving units into vehicles

#

nvm happens for all playable

#

maybe its somehow connected to dynamic simulation/enableSimulation-disableSimulation

#

[XXX",acrgpknlmstpsnonwnondnon_amovpercmstpsraswrfldnon_getouthighzamak"]
[XXX",acrgpknlmstpsnonwnondnon_amovpercmstpsraswrfldnon_getoutmedium"]
[XXX",amovpercmstpsraswrfldnon"]
[XXX",amovpercmstpsraswrfldnon_turnl"]
[XXX",amovpercmstpsraswrfldnon_turnr"]
[XXX",amovpercmwlksraswrfldf"]

solid marten
#

Also you have already posted something similar and AFAIK I have suggested to print out _thisEvent variable, but you never did, so here it is again, print it and come back

gray wharf
desert trench
desert trench
#

["Local change",[charlie_u3,"amovpercmstpsraswrfldnon"],"AnimStateChanged",any,any]

# 
private _LIB_XXX_AnimStateChangedEH = _unit addEventHandler ["AnimStateChanged",
{
    scriptName "LIB_XXX_AnimStateChanged";

    _this call LIB_fnc_XXX_handleAnimChanged;
}];```
# 
if !(missionNamespace getVariable ["LIB_XXX_globalEnabled",true]) exitWith //perma disable, incase it got set late
{
    if !(isNil {(_unit getVariable ["LIB_XXX_eventHandlers",nil])}) then
    {
        (_unit getVariable ["LIB_XXX_eventHandlers",nil]) params ["_GetInManEH","_AnimStateChanged","_WeaponDeployed","_XXX_GestureChanged","_XXX_GestureDone"];
        _unit removeEventHandler ["GetInMan", _GetInManEH];
        _unit removeEventHandler ["AnimStateChanged", _AnimStateChanged];
        _unit removeEventHandler ["WeaponDeployed", _WeaponDeployed];
        /* _unit removeEventHandler ["WeaponRested", _WeaponRested]; */
        _unit removeEventHandler ["GestureChanged", _XXX_GestureChanged];
        _unit removeEventHandler ["GestureDone", _XXX_GestureDone];
    };
};```
#

_x addEventHandler ["Local", happens it a completely different function and both systems are initialized completely separately

#

so my best guess would be the internal EH reference management can get confused by the adding (and removal?)

#

from what i recall wasnt there an issue the EH index getting changed when removed? i think this was patched a few years ago. maybe some concern case still present to botch it up

solid marten
#

So what thisEvent returns?

winter moth
#

are BI devs aware that you can jsut bump vehicles into tanks and vehicles always win or nah?

#

dont think it will ever be fixed in arma 3 right? but will it happen in reforger?

sleek scaffold
sleek scaffold
vocal abyss
#

Would it be possible to patch (or extend) setCruiseControl so that the speed limit also applies when driving backwards?
Right now this is not the case, i.e. MyVehicle setCruiseControl [20, false] only limits the speed when driving forwards, but the backwards driving speed of MyVehicle remains unlimited.

desert trench
solid marten
rose cloak
#

but it can have equivalent effect on players

solid marten
solid marten
#

Perhaps optional bool to only use speed limiter @untold sky

rose cloak
#

oof, misread by me (again)

west onyx
rose cloak
#

current logic seems consistent with what i drive IRL 🤣

west onyx
#

or just split off the speed limiter to separate function

#

augment existing “limitSpeed” function to work as an independent alias for cruisecontrol speed limiter

#

this way i can use speed limiter to help sinulate vehicle mass when towing, while player can also use cruise control

#

very not essential tho, just a nice-to-have

vocal abyss
west onyx
#

well it would ideally check velocity not speed

#

but i also dont know how it works internally

#

with limited dev time id prefer a getter for “setTowParent” and a getter “rope” command to get rope from rope segment. _ropesegments = (player nearobjects 5) select {_x isKindOf 'RopeSegment'}; _rope = rope (_ropesegments # 0);

vocal abyss
# vocal abyss Huh? It already has that (by setting the `autoThrust` parameter to `false`) <:me...

My use case here is that I want to limit the speed of player-driven vehicles when they go offroad, and the best way to limit the speed of player-driven vehicles is setCruiseControl. However, it wouldn't take long for players to discover that they can just circumvent the scripted offroad speed penalty by driving backwards, so I'd have to use some whacky SQF loop instead of setCruiseControl.

solid marten
rose cloak
#

i'd read that as "if the speed is over 120% of the set limit and user doesn't press throttle - the vehicle will apply 100% of braking" based on irl logic. Needs testing in-game, though

#

in-game testing seems to match. Hunter with cruise set to 10 goes to 6090+ with the accelerator pressed, then brakes after accelerator is released

devout wave
#

i.e.:

accelerating beyond set speed limit is restricted by the PID controller

#

I just tested it and that mode behaves as described by Ansin in forward travel; the Hunter does not exceed 10kph when restricted by vehicle player setCruiseControl [10,false]

desert trench
vocal abyss
solid marten
desert trench
#

@solid marten not me. the engine

solid marten
#

no u

desert trench
# 
    {
        _x addEventHandler ["Local",
        {
            diag_log ["Local change",_this,_thisEvent,_entity,_isLocal];

            params [["_entity", objNull,[objNull]], ["_isLocal", false, [false, objNull]]];

        }];
    } forEach playableUnits;```
#

so where is the coding mistake here?

solid marten
#

are you saying if I execute this I will have the same problem?

desert trench
#

as said the engine before hand problems with the event index when you removed them

#

it was said to be fixed

#

but apparently there is still situations where it bugs out

solid marten
#

make a repro

#

I bet 99.99% mistake in your code

#

index problem was for the same kind of event

#

you are claiming the events leak into each other

uncut briar
#

Aren't every event type handlers in different container?

solid marten
#

yes

uncut briar
#

It would be really hard for it to leak then. As you said.

solid marten
#

the event handler is executed upon event, so when Local change happens in the engine and EH exists for it it executes code for all Local EHs

#

there have to be some kind of out of bounds access to mess this up, but as far as I can see the system is solid

regal nimbus
#

Aren't the diag_log and params backwards there...

solid marten
#

instead all sorts of weird things can happen with scripts especially in scheduled

#

they are imo, but kju thinks otherwise

gray wharf
regal nimbus
#

oh right, the params is just testing the types.

desert trench
#

i am not ruling out a coding mistake on our end - hence why i am posting here in the first place
however we are not the only ones to experience this either.

i will try to get a vanilla repo but obviously the layers of complexity in between can be at fault.

for me i cannot see another reason for the rpt logging output as its defined only within the local EH scope than the EH index getting mixed up. either by us or the engine

#

the most likely mistake i could see on our end would be a removeEventHandler to use the wrong index

#

but that would require different EH types to be influenced too as local we only add twice and never removed

west onyx
#

An issue which is causing group-related bugs, is that Group Leadership changes when the Group Leader becomes incapacitated ... this is fine for AI, but is glitchy when applied to players

#

I think group leadership should not change when the player group leader goes incapacitated

#

It causes a ping-pong glitch where group leadership changes every second, from the incap leader to next-in-line and back

#

I can provide repro steps if needed

solid marten
#

yeah please do @west onyx

desert trench
west onyx
#

group leadership will change

#

it changes back when no longer unconscious. if possible IMO it should not change in the first place when player is group leader. i will try to repro the ping-pong issue still

solid marten
#

but you said it keeps changing back and forth while leader is unconscious, it is expected it would go back to old leader, he is not dead

uncut briar
#

There was a bug when it was changing back and forth constantly.
(5. Taking command, 5. Taking command...)

#

Can't recall the repro. Last time I had it it was like 4 years ago when I was still working on Liberation hide

devout wave
solid marten
#

I'll have a look if it is something simple

gray wharf
#

from the look of it, it seems to "only" be a sound config - I hope it's just that 🙂

unborn acorn
oak mauve
#

I wanted to ask if the developers noticed this bug that was reported here, because there were no responses again.
https://feedback.bistudio.com/T167742
This is a major pathfinding bug with amphibious AI vehicles.

untold sky
# west onyx and this one might be nice for some modders

ACE team requested that years ago 😄 I created findCI and inCI in intercept back then.
But I don't know if the performance difference compared to toLower+find would be worth actually adding new commands for that. Aka I don't think the difference will be large

untold sky
untold sky
untold sky
untold sky
west onyx
untold sky
# desert trench update: with this disabled, it no longer bugs out

So something on deletion.
But I cannot see an error in the code there that could mix it up. Every entity has an array of EH containers, and each container knows its type. I don't see a way for a containers type to be changed after its creation.
And that changing is the only idea I have how that could happen, this was all newly written last year

With a repro we may be able to trace it out, but without repro I can't seem to find a problem.

west onyx
#

for instance if i have player towing a tank, using setmass to simulate weight is … meh, not ideal result … but if i can just apply a top speed then that is nice. under that top speed they should still be allowed to use cruise control tho

#

not sure if thats alot of work to inplement tho

untold sky
desert trench
solid marten
desert trench
desert trench
#

no luck with vanilla. might be init EH from config vs just in init.sqf

# 
["AEH_AnimStateChanged","AnimStateChanged",0,[UNIT,"amovpercmstpsraswrfldnon"]]

["handleAnimChanged - TEST_eventHandlers - removed",UNIT,[0,0,0,0,0]]
["REH_GetInMan",0,UNIT,"TEST_SquadLead"]
["REH_AnimStateChanged",0,UNIT,"TEST_SquadLead"]
["REH_WeaponDeployed",0,UNIT,"TEST_SquadLead"]
["REH_GestureChanged",0,UNIT,"TEST_SquadLead"]
["REH_GestureDone",0,UNIT,"TEST_SquadLead"]

["AEH_Local-Player","AnimStateChanged",1,[UNIT,"amovpercmstpsraswrfldnon"]]
["Local change",[UNIT,"amovpercmstpsraswrfldnon"],"AnimStateChanged",any,any]

Error in expression <e",_this,_thisEvent,_entity,_isLocal];

params [["_entity", objNull,[objNull]], >
  Error position: <params [["_entity", objNull,[objNull]], >
  Error Params: Type String, expected Bool,Object
File xxx.sqf, line 35```
#

fairly sure the problem stems from the AnimStateChanged EH being removing, as well as local EH called in the same frame

#

["AEH_Local-Player","AnimStateChanged",1,[UNIT,"amovpercmstpsraswrfldnon"]]
has index 1

["AEH_Local-Index-Server",UNIT,"TEST_SquadLead",0]
["AEH_Local-Index-Player",UNIT,"TEST_SquadLead",1]

#

while AnimStateChanged just has 0, then gets removed

#

["InfantryInit - TEST_eventHandlers - added",Agent 0x9299e100,<NULL-group>,[0,0,0,0,0]]

#

maybe this is relevant. seems the config init EH can get initialized before the units group exists [edit: seems rather agents created for AI vehicle tank driving]

sinful kettle
desert trench
#

this is my attempt for a vanilla repo, but the code execution order is different

#

either due to init EH from config being earlier (with above null group outcome), or the animChanged gets triggered differently

#

or something with the local event - it doesnt execute the server one. only once the mission gets aborted both get

#

["AEH_Local-Server","Local",0,[UNIT,false]]
["AEH_Local-Player","Local",1,[UNIT,false]]
["Local change",[UNIT,false],"Local",any,any]

#

so for comparison

#
  1. our mod
# 
["AEH_Local-Index-Server",UNIT,"TEST_SquadLead",0]
["AEH_Local-Index-Player",UNIT,"TEST_SquadLead",1]
["InfantryInit - TEST_XXX_eventHandlers - added",Agent 0x9299e100,<NULL-group>,[0,0,0,0,0]]

["AEH_AnimStateChanged","AnimStateChanged",0,[UNIT,"amovpercmstpsraswrfldnon"]]

["handleAnimChanged - TEST_eventHandlers - removed",UNIT,[0,0,0,0,0]]
["REH_GetInMan",0,UNIT,"TEST_SquadLead"]
["REH_AnimStateChanged",0,UNIT,"TEST_SquadLead"]
["REH_WeaponDeployed",0,UNIT,"TEST_SquadLead"]
["REH_GestureChanged",0,UNIT,"TEST_SquadLead"]
["REH_GestureDone",0,UNIT,"TEST_SquadLead"]

["AEH_Local-Player","AnimStateChanged",1,[UNIT,"amovpercmstpsraswrfldnon"]]
["Local change",[UNIT,"amovpercmstpsraswrfldnon"],"AnimStateChanged",any,any]

Error in expression <e",_this,_thisEvent,_entity,_isLocal];

params [["_entity", objNull,[objNull]], >
  Error position: <params [["_entity", objNull,[objNull]], >
  Error Params: Type String, expected Bool,Object
File xxx.sqf, line 35```
#
  1. the vanilla repro attempt
["AEH_Local-Index-Server",UNIT,"TEST_SquadLead",0]
["AEH_Local-Index-Player",UNIT,"TEST_SquadLead",1]
["InfantryInit - TEST_XXX_eventHandlers - added",UNIT,GROUP,[0,0,0,0,0]]

["AEH_AnimStateChanged","AnimStateChanged",0,[UNIT,"aidlpercmstpsraswrfldnon_ai"]]

["handleAnimChanged - TEST_eventHandlers - removed",UNIT,[0,0,0,0,0]]
["REH_GetInMan",0,UNIT,"TEST_SquadLead"]
["REH_AnimStateChanged",0,UNIT,"TEST_SquadLead"]
["REH_WeaponDeployed",0,UNIT,"TEST_SquadLead"]
["REH_GestureChanged",0,UNIT,"TEST_SquadLead"]
["REH_GestureDone",0,UNIT,"TEST_SquadLead"]

["AEH_Local-Server","Local",0,[UNIT REMOTE,false]]
["AEH_Local-Player","Local",1,[UNIT REMOTE,false]]
["Local change",[UNIT REMOTE,false],"Local",any,any]
#

["InfantryInit - TEST_XXX_eventHandlers - added",Agent 0x9299e100,<NULL-group>,[0,0,0,0,0]]
vs
["InfantryInit - TEST_XXX_eventHandlers - added",UNIT,GROUP,[0,0,0,0,0]]
[edit: above should be rather agents created for AI vehicle tank driving]

#

and

["AEH_Local-Player","AnimStateChanged",1,[UNIT,"amovpercmstpsraswrfldnon"]]
["Local change",[UNIT,"amovpercmstpsraswrfldnon"],"AnimStateChanged",any,any]
vs
["AEH_Local-Server","Local",0,[UNIT REMOTE,false]]
["AEH_Local-Player","Local",1,[UNIT REMOTE,false]]
["Local change",[UNIT REMOTE,false],"Local",any,any]

#

however the local EH is meant to trigger only at mission abort (and it does also in the mod case). however the supposed earlier local EH with the AnimstateChanged parameters is the one out of line after all

#

so in other words it probably should trigger another AnimStateChanged event, but for some reason it raises a local event but still with the AnimStateChanged parameters

#

maybe the use of agents is actually involved (we use them for tank drivers to ignore AI threats)

#

this happens in briefing phase

["InfantryInit - TEST_XXX_eventHandlers - added",Agent 0x91bf0100,<NULL-group>,[0,0,0,0,0]]
["InfantryInit - TEST_XXX_eventHandlers - added",Agent 0x950d0180,<NULL-group>,[0,0,0,0,0]]
["InfantryInit - TEST_XXX_eventHandlers - added",Agent 0x94bd0180,<NULL-group>,[0,0,0,0,0]]
["AEH_AnimStateChanged","AnimStateChanged",0,[Agent 0x91bf0100,"amovpercmstpsnonwnondnon"]]
["InfantryInit - TEST_XXX_eventHandlers - added",Agent 0x9477c080,<NULL-group>,[0,0,0,0,0]]
["AEH_AnimStateChanged","AnimStateChanged",0,[Agent 0x950d0180,"amovpercmstpsnonwnondnon"]]
["InfantryInit - TEST_XXX_eventHandlers - added",Agent 0x7c41c0c0,<NULL-group>,[0,0,0,0,0]]
["AEH_AnimStateChanged","AnimStateChanged",0,[Agent 0x94bd0180,"amovpercmstpsnonwnondnon"]]
["InfantryInit - TEST_XXX_eventHandlers - added",Agent 0x9299e100,<NULL-group>,[0,0,0,0,0]]
["AEH_GetInMan","GetInMan",0,[Agent 0x91bf0100,"driver",Agent 0x91bf0100,[]]]
["AEH_GetInMan","GetInMan",0,[Agent 0x950d0180,"driver",Agent 0x950d0180,[]]]
["AEH_GetInMan","GetInMan",0,[Agent 0x94bd0180,"driver",Agent 0x94bd0180,[]]]
["AEH_GetInMan","GetInMan",0,[Agent 0x9477c080,"driver",Agent 0x9477c080,[]]]
["AEH_GetInMan","GetInMan",0,[Agent 0x7c41c0c0,"driver",Agent 0x7c41c0c0,[]]]
["AEH_GetInMan","GetInMan",0,[Agent 0x9299e100,"driver",Agent 0x9299e100,[]]]
["AEH_PathCalculated","PathCalculated",0,[Agent 0x91bf0100,[[9510.75,2424.75,78.1482],...]]

#

this is the first AnimStateChanged post briefing on a non agent

["AEH_Local-Player","AnimStateChanged",1,[UNIT,"amovpercmstpsraswrfldnon"]]
["Local change",[UNIT,"amovpercmstpsraswrfldnon"],"AnimStateChanged",any,any]

Error in expression <e",_this,_thisEvent,_entity,_isLocal];

params [["_entity", objNull,[objNull]], >
Error position: <params [["_entity", objNull,[objNull]], >
Error Params: Type String, expected Bool,Object
File xxx.sqf, line 35

#

will add the agent element to the vanilla repro to see if this changes anything

#

could be also the HD was triggered by agents

desert trench
#

["InfantryInit - TEST_XXX_eventHandlers - added",Agent 0x91bf0100,<NULL-group>,[0,0,0,0,0]]
these seem actually when you directly load into Eden from the startup parameter path to mission.sqm

desert trench
#

no dice with vanilla repo 😐

#

posted the mod repro (100%) on the FT

#

pretty sure its connected to init EH from config (with maybe agents, maybe dynamic sim) and code execution order

solid marten
#

As Bob Marley used to sing
No repro no fix
Yeeee eeea aaah
No repro no fix

desert trench
#

100% repro with the mod

west onyx
#

that is usually the bottleneck for no repro

solid marten
#

I start dedicated server with n playable units and m clients, then join dedicated server from each client by altbtabbing. My pc is crap but can handle dedi+2 clients

#

have a bat file that launches all the exes at once

west onyx
#

can dm me an example bat file?

#

i use that for server and hc

solid marten
#

tomo, not on pc, but it is just 3 lines with maybe a small pause in between

west onyx
#

k

#

i still launch interface client thru steam, which doesnt like multiple instances

devout wave
#

The Launcher has no such reservations, you can just click the Play button as many times as you like for more clients

solid marten
regal nimbus
#

Steam only checks when you run the launcher, so you can start the DS after the launcher, or just start the launcher manually.

#

You need to set loopback=true in the DS server config btw, otherwise it won't let multiple clients with the same steam ID join.

west onyx
oak mauve
solid marten
west onyx
#

oh cool thank you

solid marten
#

obviously exe would be different for you

west onyx
#

i figured out slightly but not all in same bat like that

solid marten
#

cfg:

hostname = "INIT_TEST";
password = "";
serverCommandPassword = "serverCommand123";
allowedFilePatching = 2;
passwordAdmin = "abc123";
//reportingIP = "arma3pc.master.gamespy.com";
//logFile = "server_console.log";
maxPlayers = 3;
persistent = 1;
loopback = true;
//skipLobby = true;
BattlEye = 0;
voteThreshold = 2;
//allowProfileGlasses = 0;
voteMissionPlayers = 0;
verifySignatures    = 0;    // Prevent players with unknown mods from joining the server (best kept at 2 if you want to reduce the number of hackers)
// STEAM
//steamport=8766;
//steamqueryport=27016;
//allowedLoadFileExtensions[] = {0};
//allowedPreprocessFileExtensions[] = {"sqf"};
//disableChannels[] = { 0, 3 };
class Missions 
{
    class Clean
    {
        template = "SERVER-TEST.VR";
        difficulty = "regular";
    };
};
west onyx
#

awesome

#

ill see about that repro this evening

desert trench
solid marten
#

@desert trench insideBuilding man; 0...1 ok?

desert trench
#

@solid marten i guess so. anything else seems complicated/potentially unreliable as dependent on model setup

gray wharf
#

@solid marten plz also close the ticket as "resolved" (along visibility & userblock) so it doesn't appear in some searches 🙂 thonks!

solid marten
#

removed it from arma 3 category it should not pop up in searches

gray wharf
#

I closed it
it's very useful in Phabricator's context as one can search for open/closed documents

west onyx
# 
vehicle allowCrewInImmobile [brokenWheels, upsideDown]``` possible to extend this to include cars underwater?
west onyx
#

_rope = objectParent ropesegment is great

# 
// Effect of looking at a rope segment and detaching the rope from the attached object

_ropesegment = cursorobject;
_rope = objectParent _ropesegment;   // new possibility
_parent = objectParent _rope;   // not possible
_child = (ropeAttachedObjects _ropeParent) # 0;
_child ropeDetach _rope;```
west onyx
#

tried with and without dynamic groups active

west onyx
#

possible for getAudioOptionVolumes to be extended to include Map Factor?

#

getAudioOptionVolumes params ["_effects", "_music", "_radio", "_von", "_UI"];

#

Have searched profile namespace, cant find any kind of getter for this value

#

this would be a good value to use for "earplugs" scripts

solid marten
#

ticket?

west onyx
real bronze
west onyx
#

😂

daring wagon
#

Is it possible that ctrlMapAnimAdd is not fully compatible with Eden Editor's map?

_ctrlmap ctrlMapAnimAdd [0, ctrlMapScale _ctrlmap, _position3D];

Even though I input the current scale of the map, the command seems to reset the scale to a "default value" if the mapScale is smaller than a certain number.

I think I can provide a repro for it if wanted.

solid marten
#

you followed example 3?

daring wagon
#

@solid marten I didn't, since it says for dynamically created CT_MAP controls

#

Gonna give a shot

#

Edit: Using _map ctrlMapSetPosition []; before the map animation does nothing.

devout wave
#

I've heard [citation needed] that the options menu is getting a new option added soon. With that in mind, I'd like to briefly bring this up again, since the last report was that it was super easy, and that having to do stuff to the options menu was the only real blocker. Ever the optimist, I'm hoping it could piggyback on the upcoming work to slip in. https://feedback.bistudio.com/T156158

solid marten
#

If @untold sky adds delay variable I will add a slider

west onyx
#

@solid marten how do you manage perf with second connected client? FPS on the second for me drops quite low, like 2 fps. oddly enough getting into and out of a vehicle solved it

#

and i am running it with this start "" "G:\Steam\steamapps\common\Arma 3\arma3_x64.exe" -connect=192.xxxxx.113 -password=xxxxxx -port=2312 -loadMissionToMemory -limitFPS=40 -setThreadCharacteristics -cpuCount=2 -maxMem=6000 -maxVRAM=3200

solid marten
#

try default settings

untold sky
#

2.12 will close very soon though

devout wave
#

But wait, if you haven't seen it before, then who was post? 🤔

untold sky
#

Lol what

#

Uff

limpid rune
#

that was the previous dedmen

gray wharf
#

that's why it is written Dedmen, they are legion

desert trench
#

are you guys fixing faulty bool checks that got broken with the more rigorous checking?

# 
Error in expression <mespace getVariable "BIS_missionChosen" != -1;};


private _chosenMission = miss>
   Error position: <!= -1;};


private _chosenMission = miss>
   Error Undefined behavior: waitUntil returned nil. True or false expected.
File A3\Functions_F_Tacops\Systems\fn_missionSelector.sqf..., line 272
    [] L272 (A3\Functions_F_Tacops\Systems\fn_missionSelector.sqf)```
#

(aka getVariable without default value)

gray wharf
#

I believe so, KK fixed some functions recently

fickle root
#

can you, please, fix that for AI as well, because it's still a problem

gray wharf
#

and vanilla?

fickle root
#

?

#

it's vanilla

#

no mods used in the mission

gray wharf
#

oh wait, it is the revive script indeed
well, AI can't use the Revive system for a reason I would say

#

I would love to, don't get me wrong

fickle root
#
  • injure AI slightly
  • order him via "6" to heal himself
  • while AI is playing healing animation, call setUnconscious true
  • at some moment later, call setUnconscious false
  • as result, AI is stuck on his belly with weapon on his back - won't take his weapon back in his hands, stand up/move or do anything
fickle root
#

this also needs to be fixed - https://feedback.bistudio.com/T148411
List of vehicles with commander turret not returned back by AI to vehicle's direction, when player switches to another seat

fickle root
desert trench
#

i believe even the vanilla script has a fallback fix for that in, but its incomplete

gray wharf
#

yep

desert trench
# 
            //hotfix: revived while performing an action & playing animation
            _unit playAction "Stop";

            //hotfix: revived while having no weapon or binocular
            if ({currentWeapon player == _x} count ["",binocular player] > 0) then
            {
                [] spawn
                {
                    sleep 0.1;
                    if (({currentWeapon player == _x} count ["",binocular player] > 0) && {IS_ACTIVE(player)}) then {player playAction "Civil";};
                };
            };```
#

we have this

# 
        /* Check if the unit is stuck in the unconscious animation */
        private _unconsciousStates = ["unconscious"];

        while {((tolower (animationState _unit)) in _unconsciousStates) && (alive _unit)} do
        {
            sleep 0.1;

            if (((tolower (animationState _unit)) in _unconsciousStates) && (alive _unit)) then
            {
                [_unit, "UnconsciousOutProne"] remoteExec ["switchmove", 0];
            };
        };```
#

and

# 
_timeOut = time + 30;

//TODO: non hardcode if possible
_unconsciousStates = ["unconsciousrevivedefault","unconsciousrevivedefault_a","unconsciousrevivedefault_b","unconsciousrevivedefault_c","unconsciousfacedown","unconsciousfaceleft","unconsciousfaceright","unconsciousfaceup","unconscious"];

while {((toLower (animationState _unit)) in _unconsciousStates) && (alive _unit)} do
{
    sleep 0.1;

    if ((((toLower (animationState _unit)) in _unconsciousStates) && (alive _unit)) || (time > _timeOut)) then
    {
        //TODO: improve with actual transition?
        if (!(_instant)) then
        {
            _unit switchMove "UnconsciousOutProne";
        }
        else
        {
            _unit switchMove "PlayerStand";
        };
    };
};```
gray wharf
#

huuuuuuh ```sqf please meowsweats

#

or pastebin

desert trench
#

would be great to still see this for 2.12 🙏 probably a simpler getting in the engine and would allow us to avoid the ugly setVehicleVarName hack to determine the id

outer frigate
#

Noticed random 1 seems to have a fairly high chance of generating 0, is this a known issue?

#

iirc I saw something along the lines of it in here before but can't find it again

outer frigate
#

However can't manage to create a repo with a simple loop of 1000 pushing back random 1 into an array -- 0 is never generated in any of them but in my code I frequently get 0 divisor issues when using random 1 meowsweats

regal nimbus
#

Well, it is allowed to generate zero.

#

It's not supposed to generate the maximum but this is not actually guaranteed.

outer frigate
#

I get it's supposed to generate 0, but it seems to be doing it much more than it should do -- versus using the while loop to generate 1000 and not one is 0 thonk

regal nimbus
#

Could you say approximately how often?

outer frigate
#

Couldn't have a clue as it was zero divisor error deep in my code nearly constantly

#

Going to keep trying to figure out how often it appears though

#

I could swear that someone else reported a similar issue before though

regal nimbus
#

Possible but hard to prove. Depends how the seeding works.

#

Maybe your code's just being called very frequently? :P

outer frigate
#

Possible but I would expect I'd be at around 1 fps when I'm at 130+ 🤔

#

Actually it's mostly just variable juggling so maybe not

#

Maybe I've scripted something optimised for once meowsweats
But the frequency being near constant is definitely not what I would expect and then getting no 0s in 10,000 generations of random 1 is odd, my code is certainly not generating random numbers more than 10,000 times however long the script error window appears for

solid marten
#

but in my code I frequently get 0 divisor
because you get float for example 0.12138 and if you use it as index it will be treated as 0

outer frigate
#

I tried 3 million and not one was 0 -- I was using the random number to divide

solid marten
#

post repro

outer frigate
#

Still trying to find one which is actually causing a 0 divisor repro as I can't actually generate 0

rose cloak
#

i remember random 1000 == 1000 being a thing after like 40M attempts on my machine couple of months ago. Do you have something like _array select (random count _array) in your code?

solid marten
#

comparing floats with == is

outer frigate
#

Nope, no indexing using any randomisation

rose cloak
#

inb4 the distance was 0 🤣

outer frigate
#

Already tried with that -- added 0.1 to the distance before using it; still no dice. The odd part is that the only random part is being used in a < comparison but still returns the zero divisor error (Code runs often but still have short times between script error messages) even whilst I'm stood entirely still thonk

#

Trying to find a repro still, but I'm 99% sure it's being caused by the random due to it not being consistent while stood still (and systemchats supporting that)

regal nimbus
#

Maybe figure out how often your code is being called first?

outer frigate
#

I don't seem to be getting the issue with my code any more now either which is... odd meowsweats

#

Certainly wasn't being called > 3 million times but possible it was just getting unlucky
Capped out at 100 times per second though

#

oh, found the issue... was comparing the fix for _range with == not isEqualTo facepalm

#

whoops, sorry for wasting time

gray wharf
#

agrrrrrr

outer frigate
#

i mustve got artemoz's random 1000 issue confused with random 1 and just thought it was that meowsweats

regal nimbus
#

The random 1000 issue is probably something like this: Low-level RNG generates integer value from 0 to 2^31-1, and then that's converted to a float, but floats only have 23/24-bit precision so a bunch of values just below 2^31 get rounded up.

#

The == with floats thing is misunderstood. We do it all the time in SQF, because all numbers are stored as floats. Integer values up to 16m or so have exact representations, so it works.

#

Issue here is that random 1000 will occasionally return the exact integer value of 1000.

real bronze
untold sky
untold sky
#

@alpine tulip tree search update eventhandler, fire before the search is processed, or after? I think after

alpine tulip
#

I would prefer after

desert trench
untold sky
#

yes, but command name fits

#

I'd just make alt syntax that takes unit and returns only the id number
I just don't want to make new command name 😄

desert trench
#

alright fine. return as string is fine too as thats what you usually want anyway (but can convert either way easily)

untold sky
#

Yeah groupId group player returns "Alpha 1-1"
and groupId player will then return 1
With side and name you can combine that together to B Alpha 1-1:1 (dedmen)

untold sky
west onyx
#

https://community.bistudio.com/wiki/getAudioOptionVolumes awesome upgrade thanks @solid marten

near cedar
alpine tulip
rose cloak
#

Yeah, i've encountered that while trying to apply a panel armor rvmat to geometry component for bullet penetration testing. Would be a nice addition :3

desert trench
#

there is https://community.bistudio.com/wiki/is3DENPreview and https://community.bistudio.com/wiki/is3DENMultiplayer
but not way to differentiate between Eden launched SP with or without briefing

we use (is3DENPreview && (isNull (findDisplay 37))) but this is not fully reliable as based on hardware speed, the user may not be in the briefing/have the briefing display generated in time, so the check fails even it was launched with briefing

why is it relevant: if you preview a mission x-th times while mission making, but need different behavior based on the option selected (ie disabling stuff to reduce the time waste)

in short worth to make a FT ticket?

daring wagon
#

https://feedback.bistudio.com/T169855

Anyone else having the issue that the marker of type "Empty" behaves different to other markers when dragged in Eden Editor? I don't remember it behaving like that in the past.

#

Tested on latest build without mods.

vocal abyss
daring wagon
#

@vocal abyss according to biki returns "NONE" in singleplayer

#

Edit: Tested, it does.

#

I wonder now. Why not have this return something in SP as well

real bronze
valid ice
real bronze
#

maybe, or just coincidence 😄

gray wharf
#

🔨'd anyway

gray wharf
untold sky
#

You may ask again on Monday, maybe I'll look

outer frigate
#

Is it possible to get object reference command for headgear or engine limitation on that?

gray wharf
#

I'd wager headgears are proxies, not objects

outer frigate
#

I'd figure as such but worth asking regardless; would help expand the possibilities of modding shit tons of variants without bloating load times if we could setObjectTextureGlobal onto headgear 🙂

gray wharf
#

uniform, vest & backpack are objects because containers yeah

outer frigate
#

well you can store things in a helmet if its upside down 😦

gray wharf
#

not in Arma

outer frigate
#

watch me

#

anyways, if not an object reference then some sort of ability to retexture helmets via script would be very appreciated

untold sky
#

🤡 💩

outer frigate
#

i assume this is dedmenese for either this is not possible in rv or i will not do it

west onyx
#

@untold sky im running out of SQF command feature requests 😄

#

almost all have been or soon to be implemented

#

❤️

#

can mark this one as Resolved since addition of "checkAIFeature" alt syntax

fierce vessel
# untold sky 🤡 💩

I've been experiencing a download issue with Steam for Arma 3 for several weeks now. There are 2 files that are trying to download but never do. My Arma 3 is stuck on "Validating" that keeps blinking oof and on rapidly, but never completes. Here is a image of what I'm looking at:

#

..... OK, I do not know how to post a .jpg image saved to a url link here. But I could use some help with this, if at all possible - Steam isn't very user friendly at all when something goes wrong.....

devout wave
#

Just rightclick on the image and "copy image link".
However, I'd suggest making an actual FT ticket rather than pinging people here. It may also turn out that this is a problem with Steam, rather than A3 itself, in which case you'll need to contact Steam support.

fierce vessel
#

I do not know how to make a FT ticket - I have registered with bitstudio before for a past issue, but somehow my password isn't the correct password anymore, and I need bitstudio to email me to change my passowrd, and it never emails me.....

regal nimbus
#

chkdsk, uninstall, reinstall.

fierce vessel
#

That worked well - thank you. You can look beside the Arma 3 where it has the blue "Validating" - that has been like that for literally weeks. Under the image of the Arma 3 picture at the top, you see the long rectagle with the Steam logo on it: that blinks to a green rectangle with a white triangle "play" icon, and the word "Play" beside it, off and on very quickly and has been doing so for weeks. You see the words "Starting Download" beside the rectangle, and where it says at the bottom "Downloading 1 of 2 files Complete". Again, this has been going on for weeks now. I do not know how to get this to complete.....

fierce vessel
regal nimbus
#

What does it say is updating on the steam download screen?

fierce vessel
#

Will move this issue to there

desert trench
#

is it to be considered a bug or intended behavior that triggerDynamicSimulation needs to be applied also on vehicles even if the crew/effectiveCommander has triggerDynamicSimulation active?

desert trench
#

like this reads directly from the p3d, does it not?

#

\A3\Vegetation_F_Enoch\Bush\b_sambucusNigra_2s.p3d

#

says eliteness

#

ingame

[cursorObject,namedProperties cursorObject]
[1657430: b_sambucusnigra_2s.p3d,[]]

sinful kettle
#

I reported it long ago but it was ignored

untold sky
#

Seems I just scrolled past when you last posted that ticket

outer frigate
#

Is there any point making a feature request for being able to alter how quickly a server syncs time or is that too hardcoded/too much effort

gray wharf
#

hmm, a server setting for syncWeatherAndDayTimeAndStuff = 0; // 0 = disabled, -1 = automatic, any other value = sync delay in seconds meowheart

outer frigate
#

would be very nice 🙂

uncut briar
#

Would make more sense as mission setting not server imo

outer frigate
#

agreed on that front if its not too difficult to do that way around

gray wharf
paper lichen
gray wharf
#

done, thx

desert trench
livid sleet
#

I assume inputController was designed to only work when an xbox gamepad is enabled in scheme mode and not as another joystick input as inputController -1 always returns 0 until you enable it in scheme mode. I'd imagine the best course of action is bite the bullet and use the new modded keybindings via configs?

solid marten
solid marten
desert trench
desert trench
#

can someone confirm the design logic to be correct before i make a FT ticket - a mod defined CfgRemoteExec functions/commands should still be active when a mission itself defines its own CfgRemoteExec? or does the mission definition of CfgRemoteExec make the mod definitions disabled?

sinful kettle
#

Mission takes precedence

desert trench
#

the wording is not clear

#

precedence can mean override or replacement

#

replacement doesnt make sense IMHO

#

like if ACE/CBA whitelists functions, and you would whitelist in a mission too, i can see no reason why you should have to define ACE/CBA again within the mission

#

if thats the current setup, its flawed/buggy

regal nimbus
#

From https://community.bistudio.com/wiki/stop:

Stopped unit will not be able to move, fire, or change its orientation to follow a watched object.
This won't actually stop a unit firing (including reloading magazines) once they've already engaged a target. Not sure if bug.

gray wharf
regal nimbus
#

yeah in A3 the target has to move out of a pretty wide arc.

gray wharf
#

fixed nonetheless 🙂 this description kinda always bothered me, but it never "clicked" that I could edit the phrasing 😄

unborn acorn
wary mason
#

On the feedback tracker site, do I use create new bug report for a request?

outer frigate
#

[feature request]

wary mason
#

So just put that in the title of the ticket?

outer frigate
#

thats what everyone else does

wary mason
#

okay cool

daring wagon
unborn acorn
# daring wagon What do you mean?

for now we have a CheckboxState to unblock attributes in category (eg. Manual Control of rain in weather settings)
But I want to able to block the other attributes when check box was true, reversed veraion of CheckboxState
"CheckboxStateReversed"

#

Example
There are 2 attributes in my object, the 1st is manually sets the number of rockets in pod, the second run script for a random rockets number by Checkbox
Need to lock the first attribute when checkbox running the script is true

outer frigate
#

Just... reverse the functionality of the box itself?

unborn acorn
outer frigate
#

Do Not Randomise Rocket Number:
[this is a check box that unlocks the one below]
Number of Rockets:
[ this is an edit box ]

unborn acorn
#

Yes, but I wouldn't want the checkbox to be true by default, somehow not intuitive I guess

#

We already have the checkboxReversed so why not checkboxStateReversed?

outer frigate
#

you have text and a tooltip right next to it, I don't see how that's an issue imo
i could be wrong and it could take 30 seconds to implement in which case sure, but its really a nonissue imo

unborn acorn
#

add "do not..." to some vanilla attributes, I think it will be so confusing if there was no checkbox reversed

uncut briar
#

Dedmen, as you're doing new commands, what do you think about adding a nulary getPlayerUID which would return Steam ID. I had some use cases where I needed it in main menu/mp lobby for some extension stuff and was quite dissapointed that I could not get it without player object.

Worth making a ticket?

untold sky
#

I was making new commands 😄

#

I could also like.. getPlayerUID objNull returns local?
Probably not good idea

uncut briar
#

I think separate would be clearer and better.

uncut briar
solid marten
#

but player UID is steam ID, no?

gray wharf
#

yeah
the only issue I see is if the player has not unit, e.g getPlayerUID player == getPlayerUID objNull
otherwise…?

solid marten
#

well then what dedmen suggested, since player is objnull and it will be seamless

gray wharf
#

if someone messes up and targets a null object, they will have their own uid 😄 muhahaha

solid marten
#

a hacker killing himself, yeah whynot

gray wharf
#

a player-server keeping a ref to someone that got disconnected and deleted 😬

uncut briar
#

yep, separate command makes more sense, or arg of different type like "" but that's messy.

uncut briar
#

Eg. I had this idea where when you join our server it auto assigns you to the slot you've enlisted yourself in our discord bot... No can do without steam id in the lobby screen.

#

I guess I could add steamworks to arma extension and get it from here but that... messy.

gray wharf
#

nular getLocalPlayerUID/getClientPlayerUID? (based on getPlayerUID/clientOwner)

uncut briar
#

imo nulars without get prefix are more fitting for sqf.

gray wharf
#

pllayerUiD of course

uncut briar
#

UiD pepe_screaming

gray wharf
west onyx
#

(assignedItems player) apply {toLowerANSI _x}

#

imagine if toLower took array as arg

#

toLowerANSI (assignedItems player)

#

converting all string/text in the array to lower/upper/etc

# 
_array = toLowerANSI ["tHiS iS a",123,objNull," gOOD IDEa",4];    
// ["this is a",123,objnull," good idea",4]
gray wharf
#

apply is nice in the meantime?

west onyx
#

it is nice indeed

limpid rune
#

Is the documentation on usePip actually correct? It seems bugged to me as the results described on the wiki do not seem in line with the result ingame.

For reference:

Float Gunner head movement will rotate the picture-in-picture (PIP) camera instead of the gunner's head within the vehicle. Used for example in the Blackfish gunner turrets and the Hunter RCWS versions. Since 1.73.141940 it is possible to use usePiP=2, which is going to make gunner following main turret movement - used i.e. for mortar gunner.

In game:
usePip = 0
The gunners head is 'slaved' to the turret when in the 'internal' non-optic camera. Their look speed and limits are restricted by the turrets movement, enabling free look with alt etc gives you the 'normal' internal camera restriction/speed.

usePip = 1/2
These two seem identical, it locks the gunners view forward except when freelook is enabled via the ALT key.

#

Like to me per the documentation when in 'internal' view the turret shouldn't move if usePip = 0, but currently it does. the only way to disable it is to use gunnerFireAlsoInInternalCamera = 0;

#

Which also disables the ability to fire any weapon the turret has from that view - which isn't ideal for say, smoke launchers etc

sinful kettle
#

does it make sense to have a command for setting and getting local variables via strings?
e.g. setVariable ["#hiddenVar", 1] and getVariable ["#hiddenVar", 1]? (unary version of setVar and getVar)
the point is to have local variables not normally visible to (and thus modifiable by) callbacks...

uncut briar
#

you can prevent that with private <ARR>

# 
[_args, _fnc_callback] call {
  private ["_var1", "_var2"];
  _this#0 call _this#1;
};
uncut briar
sinful kettle
#

well that doesn't work for ref vars because if they're seen they can be modified

uncut briar
#

not sure if I follow, if you private them they can't be seen.

#

If you pass them as an argument it does not matter if it's via normal variable or some getVar

sinful kettle
#

oh right I missed the fact that _this is now something else 😅

uncut briar
#

IMO what I provided is enough. Yes you add some overhead due to call and selects and private

#

but so do you by getting a variable via command which needs an array that has to be constructed.

#

the only thing that's bit annoying is having to type out all the vars you need to "secure" as private [] args.
(privateAll 🙏)

sinful kettle
uncut briar
#

VBS sqf docs are not public anymore pepe_screaming

sinful kettle
#

that's what I was thinking about when I said that

uncut briar
#

wanted to link it.

#

What everything else.

sinful kettle
#

other vars

uncut briar
#

Not following again sweating

#

you need to private all you want to be not accessible.

sinful kettle
#

oh right meowsweats

gray wharf
desert trench
#

has the fix inside (plus demo video). should be an easy one and quite gameplay relevant

sinful kettle
#

instead of delete the actual intended speed should be put there imo

gray wharf
real bronze
#

unless you have a namespace per scope for them, but don't see how that's useful 😄

sinful kettle
unborn acorn
solid marten
#

I'm still looking into if it is possible to add scripted object modifier

alpine tulip
#

For instance?

solid marten
livid sleet
#

It's fiiiiine

solid marten
gray wharf
#

oh, you ❤️

cyan basin
#

what priority are launcher feature requests given?

devout wave
#

I believe priority for all feature requests depends on the balance of how useful it is with how easy it is. I'm not sure there's any specific priority for launcher features on the basis of being launcher features.

cyan basin
#

i definitively should post to ft over chat right?

devout wave
#

Yes

gray wharf
#

ye
but asking here to know if it is a worthy/valid request is good too

cyan basin
#

i prefer chat

#

can i ask about the expectations i should expect with specific requests too?

#

i posted a bit on feature requests and a possible bug over at #arma3_launcher

cyan basin
#

uhh, the 3rd party logins do not work, known issue?

#

facebook doesnt because app is dead

#

google plus is dead not sure about api but its failing

#

Unhandled Exception ("Exception") Expected to retrieve an "account" email from Google Plus API call to identify account, but failed.

solid marten
#

logins to where? Arma3?

cyan basin
solid marten
#

here you go

image0.jpg
cyan basin
#

im talking about clicking those buttons

image.png
#

when i attempt to login/register or link an account

cyan basin
# solid marten here you go

feel free to remove them because the version of phabricator that ft is running is incompatible with google, google plus does no longer exist and the facebook app integration has been deleted

solid marten
#

It says on the screenshot I posted that google doesn’t work, did you even look at it?

cyan basin
#

i tried linking an existing account to google and i have the same exception

solid marten
#

I can’t help you, it is what it is.

#

I don’t have admin rights to back end

cyan basin
#

can you disable the providers? i see the close/remove option on your earlier screenshot

solid marten
#

no

cyan basin
#

it should be quick to ask an admin to remove those providers unless there's an actual use for those (during office hours)

solid marten
#

I have option to delete users too doesn’t mean it lets me do it. Anyway, there is no point to talk about it, I can’t help you make an account if i wanted to

cyan basin
#

i already made an account via the register button

solid marten
#

Congratulations

cyan basin
#

simply bringing the issue for attention of the right personnel

#

thank you for your time /gen

west onyx
desert trench
#

@cyan basin check with Dwarden to see if the info can be forwarded to the admin(s)

#

lowpass
anywhere to learn more about this?

west onyx
#

@solid marten can "allowService" be extended to also work for Man rearm?

#

the rearm action

#

box allowService 0

#

unless they are totally different systems

daring wagon
#

Just empty the inventory

#

Even if you disable the rearm action ppl can still grab stuff from it

gray wharf
#

I never like the "rearm" thingy, never figured out what it tried to achieve 😅

azure crane
#

yup, lack of any info what you're missing only made this option worse

#

it depends on what default loadout your unit has no? I noticed that during campaign or in other instances I would pick up chemlights or smoke grenades that I didn't need or couldn't even use

gray wharf
#

I believe it tries to get to your default classname loadout, but that's just a guess

azure crane
#

that's a very useless feature

alpine tulip
#

Yeah, useless because I love to streamsniping

#

😛

gray wharf
#

it's a "quick refill" useful in intense firefights ("click and done"), but tbh given I don't know what it is… I don't use it and it is more of a bother

azure crane
#

I understand its premise but I'd rather have Arma remember what weapons I currently have and adjust the "rearm" detection for that, not for what the original unit has, as of now it does both adding magazines I'm using and clogging rest of my inventory with ugl smokes

#

it's not an issue as long as the base unit is the unarmed one, but there is plenty of missions where base units are any other units with edited loadouts...

gray wharf
#

as I said, I guess
either that or some "2 chemlights + 2 handgrenades + 5 mags for main weapon" thingy

solid marten
west onyx
#

if it does then disregard ... im having trouble keeping up with all the new stuff 😄

gray wharf
#

lockInventory also prevents normal access to the inventory I believe

azure crane
west onyx
#

the Rearm action is generally useless in all community scenarios (which use setUnitLoadout instead of config loadout)

daring wagon
#

Is it a bug or a feature that a Guarded by .... trigger can no longer be returned by 8 allObjects 7 nor allMissionObjects "EmptyDetector"

desert trench
#

its a special type of trigger. probably not covered by the "normal" triggers

cyan basin
alpine tulip
# 
[0,1,2] select [0]```This (Syntax 5) does not return an error but just an empty array, which does seem to be unreasonable non-error. I've confused about this since I tried to do just like the string syntax (Syntax 4). Is it going to have a *fix* or nah?
gray wharf
#

I believe it should, as [0] should be a 2-items array

solid marten
#

or it should do what string syntax does, return everything from given index to the end of array

gray wharf
#

that too 🙂

sleek scaffold
#

Hello guys, I know I said I wouldn't talk about this topic anymore after repeating myself over and over again for the past few months and once every other week or so but... I just can't... I can't drop the subject. So here it is again:

#

This 2023 is going to be a great year for A3 and BI!

#

Arma 3 is doing great and the lunar year brought many new players in according to steam charts

#

Month Avg. Players Gain % Gain Peak Players
Last 30 Days 12,191.6 +1,707.7 +16.29% 25,319

#

Anyway, that is all, if anyone tough that I was going to talk about something else was mistaken 😉

#

I hope our community and devs threat these new players well 🙂 It is important to always work on player retention... We all know A3 is not an easy game to get a hold on in the beginning. A3 is more of a love due to endurance and perseverance rather than "at first sight"

solid marten
#

@desert trench ```import CfgRemoteExec as CfgRemoteExecMod;

class CfgRemoteExec : CfgRemoteExecMod
{
class Server : Server
{
class Functions : Functions
{
jip = 0;
class BIS_fnc_effectKilledSecondaries
{
allowedTargets = 0;
jip = 0;
};
};
};
};```

solid marten
gray wharf
solid marten
#

Yeah

gray wharf
#

Cc @alpine tulip ↑

alpine tulip
#

Ye, thanks

solid marten
#

merci

desert trench
uncut briar
solid marten
#

no 2.14 we branched last week

devout wave
#

Did the SPG-9 audio fix make it before the cutoff or was it pushed back because of the second problem that got found?

solid marten
#

dunno, but another problem is separate problem

devout wave
#

Well I mean it was more like a second layer to the same issue. Fixing the first appeared to expose the second, so I would've thought the first fix wouldn't go live until the second was also fixed, given it apparently makes the SPG-9 inaudible from further than 30m.
Anyway if you don't know then 🤷

west onyx
#

the ability to spawn vehicles empty (no inventory) and also spawn units naked would be handy. i use setunitloadout on units so if i spawn 2000 AI in an 8 hour session, thats 4000 sets of proxies and inventory items. the game really likes to add green chemlights to everything. it would save a lot of unnecessary object creation wear & tear

#

just thinking aloud

solid marten
#

if cancelling load out at creation stage is simple then why not

regal nimbus
#

Yeah I'm not sure how much network load it'd save, but we'd certainly use it on everything.

cyan basin
waxen tundra
cyan basin
#

oh, i see. maybe ask ft admins?

#

it should be possible to disable it without much hassle

waxen tundra
#

i'm ft admin, this needs deeper change

cyan basin
#

ah okay, yeah fixing the google integration requires a phabricator update or maybe it got removed

waxen tundra
#

we have plans to deal with FT but no eta

cyan basin
#

disabling the integrations should be doable through?

waxen tundra
#

phab is obsolete

cyan basin
#

oh really? good to know thanks

#

so its soon to be replaced, got it

#

this issue shouldnt be a worry then

west onyx
# 
clearItemCargoGlobal _v;
clearWeaponCargoGlobal _v;
clearBackpackCargoGlobal _v;
clearMagazineCargoGlobal _v;
#

im not sure if thats the most efficient way that it could be done

devout wave
#

The UCAV Sentinel has apparently always had selections for changing its ID numbers, but they were never connected to hiddenSelections to make them actually usable. 🤔 https://feedback.bistudio.com/T170116

solid marten
west onyx
alpine tulip
solid marten
outer frigate
#

Any possibility of having an addItem syntax that lets you ignore inventory space or is that a deep engine limitation?

untold sky
#

I thought we already had that

#

atleast you can overfill boxes

outer frigate
#

Creates new item and tries to add it into inventory. Inventory must have enough space to accomodate new item or command will fail.The item can also be a weapon or a magazine.
According to wiki -- seems to be correct from my testing
Would also be nice to be able to add weight to the inventory without taking up space however probably too niche to be added

devout wave
solid marten
#

additemcargoglobal will ignore

#

there are many inventory commands so find the one that does what you need @outer frigate

outer frigate
#

(Want to be able to "add" an extra 5kg for example to the player's inventory without it taking up inventory space so they can still add more things on top of that extra 5kg and so on)

solid marten
#

Not sure I understand. additemcargoglobal would add it over the allowed limit

azure crane
#

i think KJW wants to add 5kgs to unit weight without adding any item to its inventory

outer frigate
#

yes that
additemcargoglobal still takes up actual space in the inventory

solid marten
#

Odd request tbh, what is the purpouse?

outer frigate
#

Personally want it for my two primary weapons mod to be able to simulate it being added to the player's weight -- however also would be useful in mods which modify how much weight the player technically has without influencing their inventory and such; plenty of armour plate mods and such which that'd be useful for 🤷

As I said -- niche -- just curious if it's an easy thing to add or not.

uncut briar
#

could manipulate loadCoef, but that's far from being accurate.

outer frigate
#

Yeah, already tried that with an insane headache and no success at the end of it unfortunately

solid marten
west onyx
#

no worries

unborn acorn
solid marten
#

That might be the case, I'm going to scan entire config just to make sure, all INT instances are caught

solid marten
#

@unborn acorn Hopefully 2.12

dreamy bane
#

do the arma functions ever get updated? i mean is it worth posting about bugs in those?

gray wharf
dreamy bane
#

i noticed BIS_fnc_spawnVehicle only supports "airplanex" but not "airplane" simulation

gray wharf
#

so case "airplane"; @ L60 & _sim == "airplane && " L84
@solid marten, are you a lovel— would you still like to help on some functions? 😄