#arma3_feedback_tracker
1 messages Β· Page 39 of 1
lol, really? Because of the nature of that command or...?
uhm great
my bad then.
never noticed the return value nothing on something I was expecting an output π
alright sorry and thanks.
Well, in doubt always check the doc carefully!
sure, thanks.
@gray wharf spam? https://feedback.bistudio.com/T160421#2228507
definitely
@restive bone gone, thx
Thanks 
@gray wharf 2 last comments are for you )
https://feedback.bistudio.com/T63914
"yaaayβ¦"
cleaned up!
Good day. Would love to see EH in addition to 'currentVisionMode' command
https://community.bistudio.com/wiki/currentVisionMode
Something like
player addEventHandler ["VisionModeChanged", {
params ["_mode"];
}];
any chance to get cfgSFX clean-ed up? see issue in the attached file
btw is it a bug that AI still generate a path despite disableAI "PATH"?
it's a waste of performance
probably
made a ticket about it: https://feedback.bistudio.com/T160446
btw is there a ticket for this? or more importantly, is there anyone interested in fixing it?
I think the AI uses the muzzle end for aiming, and when the target "goes past it" they keep aiming back and forth (becomes negative-> tries to turn; becomes positive again -> recenters)
@untold sky Can you please silence the regexFind error m elements provided, n expected when the index is out of bounds?
example:
"abc" regexFind ["d", 3];
```-> `Error 3 elements provided, 3 expected`
and return an empty array silently? no string command throws an error like that so it's confusing.
Perhaps there are also other instances where Ai units are generating Paths when they just shouldn't. For example while they are in the INCAPACITATED state etc... Hope this gets fixes as the performance plumbers when there are some Ai units around and specially when firefights start....
also I had another request
currently when you use capture groups, only the ones that were found are shown in the result, and others are omitted. So it's hard to tell which capture group something belongs to without doing a regexMatch again, which could probably be slower
Example:
"abc" regexFind ["(d)|(b)|(c)", 0];
// sample return
[[["b", 1],["b", 1]],...]
if it could return [] for other capture groups it would be better
Wiki says
has to be in range from 0 to count haystack (haystack length)
too late for 2.06
it's not released yet 
So what
so still not too late
If I tell you its too late then its too late
Not too late for 2.08
Why broken? its exactly as the wiki says
no string command throws an error like that so it's confusing.
string commands never throw out of bound errors
select, insert
even the alt find syntax
_string find [_bla, _index];
Inconsistent yes, not broken.
I'll fix it for 2.08
Aaand done.
Next weeks (if we do one) dev branch
well how about fixing the crash with this find syntax?
is it also too late?
what crash
Maybe
looks like only with -1e9 and smaller
all negatives are broken
Okey fixed
I'm not sure if that return non-matching groups would cause backwards compat issues. But can just add another flag parameter to the command to optionally enable that
yeah that's better
ticket for that please, won't do that soon
just FYI, there are QA processes, it's not "Ctrl+S then push to prod" π
Aren't QA processes passΓ©, in the times of Early Access games? π
a change like that doesn't need QA. It's safe
Β―_(γ)_/Β―
I'll tell them you gave your green light π
https://feedback.bistudio.com/T125415 just saw this, please when you close a duplicate, use the "Edit related tasks" -> "Close as Duplicate" feature, so that it links to the actual ticket
oh, a new option
will do, thanks
I have a potential fix for the "USB Headset sounds crashing game" crash, but I don't have testers with USB headsets that can reproduce the crash π
Anyone know anyone? I already asked in #general_chat_arma a couple days ago but got no reply
Never had it crash for me =/
What is roughly required for the crash?
Have a USB Headset, and your game often crashes when sounds play, especially often when flying around with helicopters
Oh, without any interaction? Then I can 't help as everything was always working fine for me
I have no repro besides "My game crashes when I fly helicopters" "Do you use a USB Headset?" "Yes" "Try removing it" "Oh yeah that fixed it, thanks"
It may have crashed some times but only when I was (un)plugging them, but even that was long ago
And I'm flying helicopters, for the record
i do have a g533 which is a bluetooth headset connected via usb but never had any crash
Would it also work for BT headset? I have such issue with BT headset when using in... headset mode. They are working fine when I'm using those just as headphones - both in wired & BT mode
If you get audio related crashes only with a specific headset then yes
KK do you still need more info for the createDialog ticket or did you just forget to change the task status?
Fix is WIP so I guess thats sufficient info
Waiting for Dedmen to look at it
Would it be possible to get the commands sliderSetStep and sliderStep? See https://community.bistudio.com/wiki/CT_XSLIDER#sliderStep
no
unless this is not what you mean
Well, sliderSetSpeed was always accepting 3rd optional param step which was 0 by default
the getter didnt return it but thats easily amendable
Gonna add step to the getter in a minute
Alright, that does what the sliderStep attribute does in config. Very nice.
rev. 148151
btw any chance of the missing space between the text and the first 'missing thing' being fixed at some point π ?
#arma3_troubleshooting message
OCD?
@restive bone have you got test mission where this message shows?
Dont have one at hand, but it shows up whenever you try to load a mission without all required mods being loaded, so I can make one real quick for you if you want
Yes please just need to test
rev 148152
ah arma needs to update so I cant make a fresh one, I think this should work though if you try starting it via the scenarios menu https://cdn.discordapp.com/attachments/843245525662564405/883705419183321088/kp_liberation.staszow.pbo
Nah. cannot get that message, Get a bunch of different messages though
Can we have some βrichβ sound commands? Like possible to edit curves, pitches, speeds... Also a command to edit/apply what CfgSoundEffects to some sounds, or entire game sounds
I love that 99% of them don't make any sense either! π€£
boom headshot
thanks lou
Created a ticket for a bug I found while creating a UI: https://feedback.bistudio.com/T160612. ctrlShow somehow unfocuses the currently focused control.
I also created a repro mission and attached it
Is it possible to have the Alt key in Eden work the same as in Spectator camera?
Shift -> 2x speed
Alt -> 4x speed
Shift+Alt -> 8x speed
hmm. looks like Shift in Eden is 3x speed?
multiplication of camera systems was apparently a thing in A3 yes π¬
Please not like in the Spectator. It's super annoying there π
navigation in eden is too slow. I always have to use the map to go to somewhere "far" 
yes but you only have two speed modes. one fast and one medium. there's nothing in between
More than fast enough
And it slows down the closer you get to the terrain
What I wonder, does Arma 3 supports a 3D Space mouse ? π
That would be sweet for working on details
alrighty ladies and gents i do be havin some issues that maybe one of you cuties can help solve
Simply put, game crashes shortly after joining my community's server
Comes up as a standard Status Access Violation (0xC0000005) with no logs reporting what happened anywhere to be found
Happened on 2 different servers, so I know it isnt a server side thing
The servers are modded, however, they use completely seperate modlists and I have the same issue on both of them
I've Tried:
- Verifying Integrity of game files
-Reinstalling BattleEye
-Clean Arma Reinstall as well as reinstallations of ALL the mods
-Updating Graphics Drivers
-closing all unnecessary programs other than Arma
Literally ANY input on this would be appreciated as this has me unable to attend events on the server and I am sadge
Have you tried other steps mentioned here? https://community.bistudio.com/wiki/Arma_3:_Unusual_process_exit#0xC0000005_-_ACCESS_VIOLATION
Workthought: Is this feasible nowadays to torture the Engine so turn a GUI display into a texture command/settings, like r2t?
Also, is using a fullscreen camera with r2t without buggy behavior really "will not happen, is engine limitation"?
r2t just depends on the texture
if you can put a procedural texture on a GUI element (afaik you can) then you can put r2t onto a GUI element
if you make that element fullscreen, you have fullscreen r2t
I mean no, the opposite - turn a display into a texture like... whatever I do in a display will become a procedural texture so I can display an actual some info on my laptop object. The answer is a no, I suppose, though. One from my wishlist
no
Good thanks, now I can forget my stupid idea
gone
Since we have drawEllipse/triangle/polygon for UI controls, would it be possible to get those for 3D scenes as well?
Spam comment
gone
seems like these commands are broken:
magazinesDetailBackpack
magazinesDetailUniform
magazinesDetailVest
they all return the same thing as magazinesDetail 
repro code:
_magsAll = magazinesDetail player;
_magsAll isEqualTo (magazinesDetailVest player arrayIntersect magazinesDetailUniform player arrayIntersect magazinesDetailBackpack player)
return true
Those commands all return the same thing
E.g. you don't even have a backpack and you still get all mag details when you use magazinesDetailBackpack
In other words, all those commands seem to be copy pastes of magazinesDetail
Not sure if it's new or it's always been like this
But reproducible in 2.04 up to 2.07
they were debug commands hence weird output format but left by the programmer because they did provide valuable info at the time not available with other commands
i will have a loook
Yeah I'm not sure how they're useful exactly but someone said in #arma3_scripting that they needed the mag id that these commands return
Yeah I use the mag id from it too, honestly would prefer a command that did that in a cleaner way though as having to do string manipulation is annoying
What are you using the mag id for?
when I need to load a specific magazine using https://community.bistudio.com/wiki/Arma_3:_Actions#LoadMagazine
use it in the reload system iron front has for bolt actions etc
speaking of actions, is there a good reason why some of them have some (seemingly) useless and weird arguments?
like in the above example:
unit action ["loadMagazine", target, magazineCreator, number1, number2, weaponName, muzzleName]
magazineCreator: Object - unknown. Can be objNull.
Many actions were not built to be accessible via script
they are internally used things that just so happen to also be accessible via script because they are actions
I can add cr/id to magazinesAmmoFull if this helps
Can we get a uniqueUnitItems equivalent for containers/vehicles?
that'd be awesome yeah
https://feedback.bistudio.com/T62624
Can we get something like this ?
Force gamma setting or a filter like DayZ SA does it ?
so you can just log out change gamma and log back in?
Not sure how you would do it engine side. But something like putting
ForceClientGamma = 1;
Into description.ext or initplayerlocal that will get loaded on mission init.
So restarting and changing the value would have no effect. π€·ββοΈ
Or just two commands, one to set gamma via command and one to block changing gamma via command.
So every mission maker can include it how he wants
not much effort yeah, what usecase?
What about uniqueItemsArray command. Would be more versatile and right now I think we have to use arrayIntersect for that right?
Better solution for this: https://github.com/CBATeam/CBA_A3/pull/1497. Currently need to check all xCargo _container to determine if a container is empty. load doesn't cover cargo with mass = 0. Also would make checking items for saving in persistence etc more convenient.
can just do keys (uniqueUnitItems _unit) for all classnames currently
rev. 148196
now it does
How? _containerA has items X and Y with masses 15 and 0, load _containerA returns 15. _containerB only has item Y, load _containerB returns 0, no?
once you log out the mission or server params have no effect on you. adding server fixed falue for gamma is not an option because all setups are different
If they logout i don't really care what they do. Many players want that on our and other servers. Since not everyone is abusing gamma at night.
What iβm trying to say if they log out and change setting and log back with adjusted setting it is not going to help if you prevent them from further editing gamma in game as they can log out and adjust it as many times a they want until satisfied. the solution to force fixed value on everyone is not a solution because all setups are different and the same value may have different effects on different players. Do you know of a clean solution that cannot be circumvented?
all items have masses, if your item doesnβt have one it is misconfigured. loadAbs will return sum of masses reliably
Quite a few items in ACE have 0 mass, intentionally
If you make a FT ticket for the command I can do it for 2.08, but keep in mind 2.08 is quite a long way away
It would be easy to just make a alt syntax for uniqueUnitItems, but I guess people will want it to be named uniqueContainerItems :/
can always rename it, not too late
Yes too late, it goes out for 2.06 which is already closed, and mods already prepared code to use it
mods can Ctrl+H
We can just say "well a vehicle is also a 'unit'" π
a unit is a vehicle 
is it not possible to see error messages of code in eachFrame EH?
I think it is, nvm
Why that windhose is not returning items
the solution to force fixed value on everyone is not a solution because all setups are different and the same value may have different effects on different players.
I think it is not game engine concern. This is about mission maker vs players.
getters (and setters) for gamma(all video options that affect view. Shadow, brightness, posteffects etc.) will allow mission/addon makers to make kick or force settings scripts. For example if you have 3 hour mission that starts at sunset, you can store gamma settings per player and check if someone crank it up during game or on reconnect.
If your will or your hardware do not allow to play this mission with defined rules, you just skip this mission. And blame mission maker.
Currently it is quite insulting to play night missions when "cyber athletes" instantly turn off shadows and crank up gamma while you trying to play with atmosphere. You try to use flares, torches, NVGs but opponents don't care.
yeah setters that change those video options for a specific mission (but don't change the profile options) can get the job done imo
A simple "force gamma to value while playing this mission" would be rather simple to implement.
But is it worth the effort, I haven't abused gamma in A3 for atleast 5 years π
I know people who do, regularly
Hasn't DayZ implemented this screen grain thing that was supposed to make the game unplayable if you bump up your gamma more than you're supposed to? (while it's not noticeable if you keep the gamma at reasonable values)
dont know how well that works but that sounds like a promising idea as it stops people from just cranking the gamma outside of the game like in their video cards control panel or monitor
I think it's a very neat feature for pvp especially.
Also for night/horror missions.
It supposedly looked like this in 2018 (can't confirm since I don't play DayZ)
https://imgur.com/a/TupH7ir
There are some genuine accessibility concerns with gamma settings, not to mention differences in hardware and viewing environment. A hard force to specific value, while simple, wouldn't be very accommodating. It might be better to allow a range to be specified, with the final setting mapped within that range proportionally to the original settings. So someone who really needs higher gamma wouldn't be [as] completely screwed if the missionmaker has a very bright screen and sets the limit low.
Probably they will do it with monitor options, nvidia panel, sweatfx etc.
I'd argue that even limiting the slider would substantially decrease their number anyway, as just sliding a slider is way simpler than using an external program.
(yes, I've dealt with end-users and they're not that sharp as you'd expect)
Adding the grain, instead of locking gamma could help in this case, I presume.
So someone who really needs higher gamma wouldn't be [as] completely screwed
But do they really need to crank the gamma from 1.0 all the way to 2.3?!? π±
Check how bright 2.3 is. It burns my eyes out
Probably not, but if they need say 1.1 or 1.2, but the missionmaker's screen is calibrated weird and they hard lock it at 0.8...
Then he will get feedback on that issue
I guess its fine if we just provide a getter for all video options (like we already have for audio, can use hashmap for this π) and then you can implement grainy effect by yourself
Why hashmap? 
You can just return an array...
Cannot create object with type[AIGroup], param[group], NMT code[107]:
Cannot create object with type[AISubgroup], param[subgroup], NMT code[111]:
Client: Object (type Type_112) not found.
Client: Object (type Type_108) not found.Cannot create object with type[AIUnit], param[unit], NMT code[116]:
Client: Object (type Type_117) not found.
Client: Object (type Type_118) not found.
any info available what these are about?
(client rpt from briefing state)
Dedmen#7: Group without leader, GE CU Unit
is this some internal debug/logging?
cant find "Dedmen" in A3 configs, nor ours or in the mission (identity definition)
Then you always need to look up at which index the setting that you want is, and if we add new settings (like we just did for PIP) then you have PIP Quality at index 2 and PIP Resolution at index 32
settings select 32
Instead of just
"I want PIP Quality!"
settings get "PipQuality"
More readable, no magic numbers, simpler to extend
yes, telling dedmen that the testfix is working
or rather which of the testfixes are working
ok. an unit in being in null group is fine?
no its not
Wait
no
Units cannot be without groups
creating a unit in grpNull, automatically creates a new group and puts the unit there
ok. let me check with isNull
wat? check what?
wat
Thats unrelated to that log message you posted
and player units aren't createUnit'ed (usually the engine spawns them)
i didnt imply its related. just having this issue, and noticed these two unusual rpt messages - thats all
createUnit is done by CP
// --- spawn a copy of the playable group to calculate insertion positions (uneven terrain etc.)
_slots = playableSlotsNumber BIS_CP_playerSide;
BIS_copyGrp = createGroup civilian;
for [{_i = 1}, {_i <= _slots}, {_i = _i + 1}] do {
_newUnit = BIS_copyGrp createUnit ["B_Soldier_F", [-1000,-1000,0], [], 0, "FORM"];
_newUnit stop true;
_newUnit allowDamage false;
};```
// --- register blacklisted azimuths
if (isServer) then {
BIS_CP_dummyGrps = [];
["Final locations list:"] call BIS_fnc_CPLog;
{
_dummy = (createGroup CIVILIAN) createUnit ["Logic", _x select 0, [], 0, "CAN_COLLIDE"];
_azimuthBlacklistModulesArr = _dummy nearObjects ["ModuleCombatPatrol_LocationAzimuthBlacklist_F", 1000];
if (count _azimuthBlacklistModulesArr > 0) then {
_module = _azimuthBlacklistModulesArr select 0;
_blacklistArr = call compile (_module getVariable "BIS_CP_param_locationAzimuthBlacklist");
if (typeName _blacklistArr == typeName []) then {
_dummy setVariable ["BIS_azimuthBlacklistArr", _blacklistArr];
};
};
BIS_CP_dummyGrps pushBack group _dummy;
[" %1 at %2", _x select 1, _x select 0] call BIS_fnc_CPLog;
} forEach BIS_CP_locationArrFinal;
};```
_tempGroup = createGroup [BIS_CP_enemySide,true];
_newUnit = _tempGroup createUnit [selectRandom BIS_CP_enemyTroops, _buildingPos, [], 0, "NONE"];
[_newUnit] joinSilent _newGrp;```
would be likely one of these
I guess it can happen when you hit max group limit?
everone owns the turret? 
The vehicle is Kajman, the local player is sitting in the pilot seat
Or, this shouldn't be executed on -1?
server command
Yes I'm server
hosted?
make a ticket
Roger
@untold sky ,
I mean - if u need to get multiple values from hash map - must to do something like:
private _var1 = hashmap getOrDefault ["key1",value1];
private _var2 = hashmap getOrDefault ["key2",value2];
I just remember how works set command for hash maps, or params. Would be nice to have something like this:
hashmap getOrDefault [["_variable1","_key1", default Value1],["_variable2","_key2", defaultValue2]];
Thank you!
I guess hashmap set could also support multiples the same way
nooooo 
keys can be arrays
It works for set :)
And I thought of it like this
(hashmap getOrDefault [["_key1", default Value1],["_key2", defaultValue2]]) params ["_variable1", "_variable2"];
No it can't?
Yeah no it doesn't...
hashmap getOrDefault [["_key1", defaultValue1], ["_key2", defaultValue2]]; is ambiguous.
It could mean
hashmap getOrDefault [key, defaultValue]
or it could mean
hashmap getOrDefault [[key, defaultValue], [key, defaultValue]]
so if I were to do that, it would need a new command name, and its not worth that
i used that in my scripts π€¦ββοΈ
no idea how it works without errors
now tested - is works just like you said - just creating new key in hashmap
thanks for saying!
If you want to put multiple keys in the hashmap at once, use insert instead of set
Is it possible to add stream image from Steam compositions to the editor?
To be like editorPreview? something like that:: https://ibb.co/Jt4wmb1
setUnitPos export to SQF issue
https://feedback.bistudio.com/T160837
looks like a "typo" and easy fix
leader loses his place
https://feedback.bistudio.com/T160834
this is quite annoying - right now you have to reorder the whole group manually AFAIK
I added the relevant entries to the ticket. It's just an incorrect array that's causing that.
ty
Yes, But I don't want to, not worth the effort
can you throw errors so that you get the stack trace logged?
or command to print the stack trace to .rpt
[BUG] One-man-tanks (hasDriver=-1) not working in multiplayer
https://feedback.bistudio.com/T153212
@solid marten finally came around to provide the repro. would be great to see this fixed finally π
@desert trench the steam tickets you linked suggest that if you are a host you can drive and shoot which contradicts with your repros
@solid marten On a DS nothing works as far as i can tell. With hosted session its a bit weird. From editor launched it works for me, for a regular hosted session it doesnt.
Overall it seems to be about "locality" / effectiveCommander state. Seems its not synced properly with "hasDriver"
thx that works π
can you spawn the tank on dedicated on your client and see if it works?
isnt that different from assert because it returns variable values?
@solid marten yep works (used (typeOf cursorObject) createVehicle (position player))
noticed cursorTarget returns <null-object> - only cursorObject works. is that normal for empty vehicles?
cursorTarget is older and works on unit's knowledge (press "T" to reveal the vehicle, you will see)
hm it should be, try waiting a bit?
the reason cursorObject was introduced
ok. can i change the locality of an existing (empty) vehicle?
setOwner doesnt seem to work
maybe the fix will not be difficult in this case
it should work π
(but the command is server-side only)
stop breaking things!
oh the line number means scope change in diag_stacktrace. thought it meant variable assignment
same system in the background.
One dumps to RPT, then other returns as an array that you can dump to RPT yourself
each entry is one scope level
kill that spam pls https://feedback.bistudio.com/T159917
done π thx
https://feedback.bistudio.com/T160719
spam comment
thanks, nuked
Would an event handler to go along with cameraView command be too much work?
Stealth spam comment https://feedback.bistudio.com/T159344#2236344
gone, thanks
both gone, thx
https://feedback.bistudio.com/T122960 the last two comments are just useless/spam
also should it be closed? it's a crash from 4 years ago
done, thanks*
fixed a long time ago, please close
https://feedback.bistudio.com/T157560
done, thanks
Can't get headless clients to stay connected to their slots on Tanoa and Livonia, 2.04 prof and stable branches, even with blank mission file except for said slots and a player slot. Mod maps as well as Altis, Stratis and Malden work fine. Haven't tested Creator DLC maps. Ticket necessary?
EDIT to clarify: isPlayer on the HC entities after mission init returns false and allPlayers doesn't include HC entities, alive on HC entities returns true and owner returns 2 (executed on server) or 0 (executed on client)
is it possible to fix linespacing attribute for CT_EDIT and CT_STRUCTURED_TEXT controls? it does nothing 
or is it too much work now?
just these 2 maps?
did it work before 2.04?
Don't have other DLCs to test, but any modded map works fine.
AFAIR, yes.
so let me get it straight, HC disconnects on livonia and tanoa but doesnβt disconnect on modded maps? what about altis and stratis?
Altis and Stratis (and Malden) work fine. HC doesn't disconnect from its entity.
Additional info: bug only happens on Steam accounts without licenses for Apex/Contact.
I thought HCs run without steam
at least on Linux they do
They don't actually need a steam connection π€
I really don't know what to tell you other than I can reproduce it 100% of the time.
Account with the DLCs works with no issues, account without them doesn't.
I don't like guessing, since I have no idea what the source looks like, but the issue is probably in the code that kicks clients out of a map when they don't have the required licenses.
yeah I know the code but it should specifically ignore HCs
I'm happy to do more testing if you can point me to something else, /shrug
Β―_(γ)_/Β―
Also this would completely kick the player off of the server.
If I understood right your HC's just leave their slot?
Yes, as if they aborted mission and went to lobby.
yeah that can't be this system then
Why would it only happen on DLC maps, then? I was also able to reproduce it on GM and PF's maps, same setup mission wise.
I'd ask: does it happen on vanilla? (sorry if it has already been answered)
Nope.
I mean, vanilla + DLC + CDLC, all official content, but without community mod
When this first happened, thought it was a script or mod issue. Went full vanilla, blank fresh mission file, just player slot and HC. Made mission attached in ticket. Didn't work, switched branches, same deal.
Switched maps to Altis, worked.
Tried Livonia, didn't work.
Restarted Steam and my box just to be safe because I've had issues with HCs getting kicked because of a Steam Auth error before, didn't work
Switched to my personal account, worked.
Switched back, didn't work.
I can still reproduce it as of now, so going to buy Apex on my server's account and see if that stops it.
now that's some dedicated testing!
something something conspiracies, something something BI bad
that's how we get people to buy DLCs, the extra mile π
So, would dedmen explode if I said that buying Apex made it work on Tanoa?
Don't distract me I'm playing with lasers
Now I exploded
I've had issues with HCs getting kicked because of a Steam Auth error before
That also shouldn't happen, HC's should ignore steam auth
Just a thought: What about alt syntax for getText/Array ... with default value? So if property does not exist it will instead use the default value provided.
Use case:
private _soundName = getText (_config >> "name");
if (_soundName == "") then {_soundName = _class};
Perhaps there is already a better way of doing that. Let me know if that's the case
Opening attributes on vehicle with storage takes long (especially with a lot of mods)
I wonder if it is possible to make the storage loading after clicking on the "Equipment Storage" attribute?
to set var name or type code in init, it was not required to wait for initialization of the equipment of this vehicle
@unborn acorn The Eden Editor attribute?
yep
It should only be slow the first time you open it
In my mod I did it so that it would preload the attribute when the editor is started.
What if... we can improve a3\3den\UI\Attributes\AmmoBox.sqf?
Not really sure we can.
I've already looked into it
There are so many exceptions and things one needs to work around. Would be nice if it was simply looping through cfgWeapons and adding them to a hashmap π
Neither, I don't have an idea what it does π
Here is the important part of the script
But yeah this seems complex enough to make a nice freeze
I've just noticed. Sound and Voice are exactly the same. I guess that's not intended? Worth a ticket?
Also, previewing the voice thingy does nothing because the code aims for cfgVoice instead of cfgSounds
Isn't a 20 yo thing (aka intended)?
Idk
sounds worked in OFP, so no π
removed, thx
could it be made possible to allow different color text in the server parameters?
"different color text"β¦ where?
server parameters screen
oh like to have a red title for your server you mean?
most likely not
I mean allow different colors for the parameters
you could highlight important parameters that way
do you mean mission parameters then?
yes
ok, because to me this is server parameter ^^
this
yep
given your screen, I would give a higher priority to categorisation rather than colour
but I already have categories =/
that are made of empty settings right?
settings that do nothing
yep, so I'd rather have an official way to have that π
oh right
if your setting is so important, either set it at the top or start it with [IMPORTANT]
colour would only bring rainbow situations IMO
and bold
and italic
and font size Γ 2
lol
honestly, you can make a ticket but I hope it's not something that will happen π
lol, great... π
Here's the ticket for that, which you all have been waiting for: https://feedback.bistudio.com/T160964 π
runs away
I noticed you can improperly use macros with class variable and not get error msg
#define MECH_PRICE(a) (1000 + 1 * a)
class Test
{
value = MECH_PRICE;
};
No argument passed to the macro but no error message
if you do something similar in sqf then you get the error message though
While testing a linux port of my extension, on WSL2, I noticed that if I did
./arma3server (...) "-servermod=@Pythia" and @Pythia was a symlink to /mnt/e/somethingsomething/@Pythia, then the 32bit server hang right after SteamAPI initialization failed. Steam features won't be accessible!
Replacing the symlink with a copy of the real directory makes it work. Symlinks to WSL directories work correctly as well.
Oddly enough, ./arma3server_x64 with the same parameters boots correctly and works fine so it's just a case of the 32bit binary and a symlink to the windows drive π€
Is this ticket-worthy or too niche?
Last interesting strace messages are:
openat(AT_FDCWD, "/proc/self/statm", O_RDONLY) = 14
fstat64(14, {st_mode=S_IFREG|0444, st_size=0, ...}) = 0
read(14, "18816 9443 5546 10022 0 7283 0\n", 1024) = 31
close(14) = 0
access("@Pythia", R_OK|W_OK) = 0
fstat64(14, {st_mode=S_IFREG|0777, st_size=440, ...}) = 0
and a lot of futex(0xf70d8400, FUTEX_WAKE_PRIVATE, 1) = 0
btw using the /mnt/drive path to compile the project is the default behavior with Visual Studio when selecting the WSL cmake setting
I'd say thats pretty too nieche
is it possible to have a bool parameter for hashmap set to set the data only if it doesn't exist?
right now I do:
_hm merge [createHashmapFromArray [[_key, _value]], false]
```
#define SET_UNIQUE(x,y) merge [createHashmapFromArray [[x, y]], false]
_hm SET_UNIQUE(_key, _value);

just check hasKey first?
in general yes thats possible and pretty easy but I won't do that this week
yeah but I'm lazy π
so is it ok to make a ticket?
yu
Using loadAbs on a container that has not yet been opened in the mission returns 0. Ticket needed? 2.07 Dev Branch.
Here's a small Eden Editor bug https://feedback.bistudio.com/T161001 (Cannot reset search result of the locations list)
Is it known that you cant use the scrollbar in Eden triggers if you have too many items in cfgSounds/cfgSFX?
Spam: https://feedback.bistudio.com/T160983
and this one: https://feedback.bistudio.com/T160980
both gone, thx
@daring wagon general issue for combo boxes or "just" as part of Eden?
For me I only had issue in Eden Editor. For example selecting the scroll bar sometimes makes them collapse again, or the list get drawn behind other controls.
Any comments on this? or is it somehow wanted behaviour?
what does it return in the end config?
with getNumber I think it was zero
try getString (or open config browser), it might return "MECH_PRICE"
you mean gettext, it doesn't return anything..
well, check the config browser (and yeah getText)
config browser? I forgot to say this is in mission description.ext
config browser has description.ext config too
Its intentional.
Unknown blabblygoop that doesn't make any sense -> string entry
ok thx I will try to find the config entry..
hmm cant find "Test" in config viewer
ah found it
the value is just empty string
so, yep, all good
but no error =/
Are you sure you get error for invalid macros in SQF?
Because.. I remember ACE having dozens of invalid macros and never noticing until I compiled with SQFVM which correctly detects and reports them
yeah I get error in init.sqf
sqf```
#define TESTM(a) (1000 + 1 * a)
sleep 0.1;
systemchat format["TESTM: (%1)",TESTM ];
error:
systemchat format["TESTM: (%1)", ];>
13:13:20 Error position: <["TESTM: (%1)", ];>
13:13:20 Error Generic error in expression
that's SQF error, not compiling error
it ends up with this code```sqf
systemchat format["TESTM: (%1)", ];
yes makes sense
I try sometimes
SQFVM = https://github.com/SQFvm/runtime ?
yes
cool π
I think the release build was a bit outdated and had errors
you'll have to get the latest build (either compile yourself, or find one in repository's Actions)
ok, can it read configs like description.ext?
no I don't think so... 
ok
nope, you don't
they just fail silently (replaced with empty string afaik)
just wanted to get errors when trying to use those macros without arguments in description.ext so i don't have to go through all the configs looking for errors π
yes
nice π
dealt with, thanks
on it
done, thx for the report
Is there any hope for the introduction of an autowalk (essentially toggle W) keybind at this point?
are you that lazy?! π€£
My wrist hurts after holding W for the better part of the past two hours 
(following a day of more gaming)
Just get a flashlight or some small glass bottle of magic health oil that you can just stand on W key
or just use AutoHotKey?
I'd never heard of that one π
AutoHotKey is basically what AutoIt v2 was over 10 years ago
17 years ago π
thats the "programming language" I started with some 10-12 years ago
10 years ago
yeah probably. the reason I learned about AutoHotKey was because I was trying to play Project Reality but it had no toggle crouch. So I googled how to enable toggle crouch for it and I found autothotkey! π
given how old PR is I think I shouldn't be surprised π€£
I'd prefer a built-in solution over installing third-party software, but I don't know how much trouble introducing that would be.
Won't do
dunno about AutoIt but AutoHotKey is pretty lightweight 
like only a few MBs
I got a scripted version but I think it isnβt local to the client, so youβd have to be fine with the network traffic of the animations
I can just loop the walking animation 
player playMove "AmovPercMrunSlowWrflDf";
player addEventHandler ["AnimDone", { player playMove "AmovPercMrunSlowWrflDf"; }];
Is that the same you are doing?
Yep, and account for if you can sprint, the terrain slope, swimming, damage etc
Oh, fancy! Do you have a link?
Would have to make one, on a private repo, itβs a 168 line function for every type of walk/run/swim, you could customise it for your mission
https://forums.bohemia.net/forums/topic/236063-release-auto-move-function/
https://feedback.bistudio.com/T160940 Some folks seem to only have a limited set of languages available? Anyone knows the cause of this?
They say reinstalling and/or verifying the game does not help.
Does Arma have a Steam language selector thingy?
Maybe they selected some language other than English there 
Don't think that makes a difference
I got all languages no matter if I set it to German or English in Steam
I just added some intel in FT comments about this
so when do we get the ++ operator? π
We won't. This has been requested before.
"I always thought it was because I bought game outside of Steam store. Thought I had a game copy only for my region or smth" how elese can you buy Arma3 if not on Steam?
ok thx π
same goes for += operator?
There was a time when one could buy arma 3 as physical disk, wasn't there?
dunno, I think it was always steam, maybe you are talking about Arma 2?
Is Arma 3 out in stores in Canada? I thought it came out in stores and steam on Sept 12? Well, where is it? I see it on Steam but nowhere in stores. Am I missing out on something? Will it come late or what? If someone could clarify this issue Id be very happy. Thanks!
Seems like there was, at least in europe
2 ΠΌΠ°Ρ. 2014 Arma 3 Retail - Russian for me said Steam
So perhaps that's the connection
but what about the other guy? 
Going to ask him.
the BIS store used to sell the game
yes, but it sold Steam keys
Hey devs, you see how when a player tries to walk on a fallen wall or a destroyed building they would sometimes walk backwards even if you are trying to go forward... it is super annoying so why not make it that if you press W you just go forward!?

I'm sure thats something the devs never thought of
Dedmen and the other devs are propably going Woah why didnt we just implement it like that?? right now
You mean like how the Ai units were checking constantly if their flashlights we ON even during daytime and it was fixed now after like 10 years?
Miracles do happen you know
I have a feeling there's a difference between something like that and 'just making the player go forward if you press w'
Prhaps tt has something to do with the GEOM lod of course... perhaps the player looses its ROADWAY lod when over those fallen/destroyed objects?
Then a solution could be that if the player is on a GEOM lod, it just considers the ROADWAY lod under it when calculating its movement
if it is what I think it is, "moving forward is moving backward" is to make people "exit" object geometries they might have entered - which is counterable by⦠moving backward to move forward.
You can buy keys online and activate them in steam
Please keep your crappy trolling out of here...
I know what the problem is
and I have an idea for that too
If a player enters a object geometry that it should not enter, the game should just make it no longer be able to move forward, and the player himself will press S to get out of it... right now it feels like a bug the way it is implemented
Changing two lines of code is not a miracle
"just" sure.. no
Finding the issue after 10 years when several other "coder eyes" could not find it looks more like a miracle π
Well it didn't break anything
so it was not something anyone was looking for
And you don't find such issues by looking at code
Thank God someone who tries to find work arounds rather than just "it is an old game let it be" mentality (Pi)
it's not just that (although less sarcasm from Piβ¦ π)
it's also that issues must be prioritised - we can't throw devs at every bugs we know/discover π
and the cloning protocol has failed to this day
Spend potentially days trying to fix "when you walk on a fallen wall or destroyed building (which rarely happens) you would SOMETIMES walk backwards for a bit"
Or use that time to work on things that actually matter
Completely agree! Just the most glaring and annoying ones first... like the issue I pointed is very prevalent and happens on so many missions
"it is an old game let it be" mentality (Pi)
Ah yes thats exactly what I said
like the issue I pointed is very prevalent and happens on so many mission
I don't remember ever experiencing it. Atleast I never saw it as a problem, nor have I seen any tickets.
Your message is the first time I hear about it. Doesn't smell like a very important problem..
to you ^.^ this is what I mean. but hopefully after doing a FPS *= 3; operation Ded' can check other stuff :3
After more fps I'll work on lag spikes and features that make modders happy
What I was trying to point out with my comments was that, at least to me, you were sounding like an asshole with the way you wrote your message that thought that nobody that has ever worked on Arma 3 had enough brain cells to think that maybe "if you press W you just go forward!?"
Not on things that a few players rarely experience and that mildly annoys them
and -easy- with the used wording/lexicon still π please
defending is a thing, attacking another - we're all here for Arma luv, thanks
meh
re-read in an hour or so, you'll notice it could be "what is the reasoning behind that?" rather than "why is it so stupid?" π
luv, I say β€οΈ
what? was my question somehow not ok? sorry if it was asked before
countless times
btw if you wanna know, the problem is that when the game detects that the player has moved into an object, it applies a velocity opposite to the "would be" velocity
while it is not a bad idea for stuff like falling, etc., it causes problem when the unit is moving willingly (although letting the unit move in whatever direction he wants is not the solution either)
or should I say it "moves" the unit in the opposite direction of the velocity
because it doesn't change the velocity
only reverses direction of motion
so you want a ticket?
Completely agree, and also I think that fixing the reverse velocity thing will make the composition makers quite happy too π (Are composition makers considered modders too?)
Won't fix. Simple.
You are right... I did perhaps sounded like a douchebag, but believe me that is only because English is not my main language and so the way I translate it in my mind and in my language doesn't sound bad...
My point being that I am not as bad as that comment by me could made me look like...
Even so, perhaps a ticket to preserve @sinful kettle's findings would be good, perhaps other dev would fix it one day π
Is it worth to torture the engine nowadays to have an expanded ability for ctrlSetAngle to do it with more control types? Or maybe just for controls groups? I would like to see a angled texts... for some reason
Can anyone confirm that the Eden Editor attribute "EnableDebugConsole" actually does nothing? I set it to "Available for everyone" but it's not availabe in SP and MP hosted (DS not tested)
have you tried putting GetNumber(missionconfigfile >> "enableDebugConsole") in the console watch field?
Returns always 0 (Only available in Editor)
doesnt work for me either
Why would that return anything when you didn't add it to description.ext? 
I thought editor setting would overwrite it
but I don't know where those editor settings are stored
It returns Eden Editor setting if no missionConfig setting is found
Ah wait. Sorry. I was talking about getMissionConfigValue
That will return the correct value set in Eden Editor. Still no debug console in SP or MP
Vanilla debug console right?
yes
maybe only missionconfigfile is read for dedi clients
I am talking sp
oh sorry
I used dropdown menu yesterday to set avail to all and it worked. you will need a ticket and a repro
https://feedback.bistudio.com/T161136 Here you go!
It has been like this for me since ages.
worked for me on friday 
Should weaponState syntax 3 muzzle return weapon's current muzzle, or first?
Syntax 1 returns same as currentMuzzle, syntax 3 returns weapon's first muzzle.
23:00:25 Error: Mismatched Begin/EndContexts
Any hint on what that could mean?
dunno video doesnβt play and there is no more context
Why does the video not play?
Can you check if debug console is available in the mission I added?
Programming error related to scripts, you don't happen to have repro steps for that?
I think it started when I used endl as pattern in regexMatch
I can recheck later if needed
It should have to be something bigger than script commands, one level higher.
A script command itself shouldn't be able to do that 
Then idk. I did tons of UI stuff yesterday.
Video works for me and I'm not even logged in.
well it doesnβt work for me
try on a PC perhaps? I had some buffering hiccups on Firefox/PC but it went well after that
why not use youtube?
I was on desktop Firefox.
I just dragged it into the input field. Never had any issues with that.
I've just noticed. Every time I place an "empty" object (prop) in Eden Editor it writes "<TIME> Trying to create an invalid center EMPTY" into the rpt."
It also happens if I create a bunch of props with create3DENEntity. Any chance to get rid of that spam?
it writes 17:01:05 every time? yeah that's weird
fixed it for you
muh trolling!
Huh.. I'd assume someone would've noticed that if that has always been the case 
Hello Arma commuinity,
10 months ago i created this ticket on the feedback tracker
https://feedback.bistudio.com/T155018
The FV-720 Mora behaves weird and immersion breaking on destruction.
That bug basically renders the vehicle useless for use in any arma mission.
Did this ticket maybe get overlooked?
No don't create duplicate tickets 
I always skip tickets like that, because I know I cannot do anything about it.
And if noone who knows that stuff looks at them, they get ignored/forgotten
It's probably a bug in the model
so I guess reyhard is the one that might be able to fix it
https://feedback.bistudio.com/T85239 A very old post... It is about a Revive Eventhandler. Is it possible that you take a look at this? π
Revive functionality is scripted, it canβt be engine event handler only scripted
https://feedback.bistudio.com/T82480 who would something like this fall under?
It's honestly the most annoying part of making attachments to duplicate a model rather than just reskin it. Or would this be considered out of scope for A3 in end of life?
there is some weird race condition(?) going on with textures in 2.06, see #arma3_scripting message π€ i can post the driver code for generating that array in pastebin(2.06) if needed, i downgraded to 2.04 and it was empty
you should use BIS_fnc_initVehicle to customise textures. Vehicles that have random textures applied by the game will apply them in post init so yes you will have race condition
ok yeah that seems to be the case π€ but why wasn't it like that in 2.04? π€
it was 
iirc someone asked me a similar question about texturing. I wrote a script for them but they said it didn't work, but it worked when they used debug console. so I just told them to add sleep and it worked 
private _veh = createVehicle ["C_Heli_Light_01_civil_F", getPosATL player, [], 0, "NONE"];
_veh setObjectTexture [0, "a3\air_f_heli\heli_transport_04\data\heli_transport_04_base_01_black_co.paa"];
[{ systemChat str getObjectTextures _this; }, _veh] call CBA_fnc_execNextFrame;
this code produces different output on 2.04 vs 2.06 on the same exact mission
this is what im wondering
Me, should be relatively easy but I wonder if its really worth the effort?
It would make my day if you could, would mean making attachment variants much simpler.
I put it in my notes but not sure if its that likely to happen
ah, the joys of feedback tracker.
Okay how can I use a scripted Eventhandler for this? The workaround in the comments did not work for me π
Things I just wonder:
- Can we have a feed in Launcher that shows changelogs of recent Mod updates?
- Can we have a proper gun fire smoke/light effects when the unit is attached to something?
because in 2.04 it was even worse, the textures were transmitted before vehicle was ready to receive them on remote machines so you end up with different textures or no textures
I will need to find where in scripts to add it, so for this I will need a ticket i can test, so ideally a light mission with exact steps and expected outcome
not the same, that code linked used next frame execution
PostInit is one frame after init right?
Might be that cba execNextFrame fires before we fire the object postinit EH, so it might be off by like a quarter of a frame
no, execnextframe doesn't do anything here, its just for illustration, if you call setObjectTexture in the same frame as createVehicle, the texture will be changed back the next frame(the init vehicle stuff im guessing, what KK said), disabling the randomisation fixes it and it works as expected i.e.
disableRandomization[] =
{
"AllVehicles"
};
Ah nvm
Didn't look and thought the code was this
private _veh = createVehicle ["C_Heli_Light_01_civil_F", getPosATL player, [], 0, "NONE"];
[{ _this setObjectTexture [0, "a3\air_f_heli\heli_transport_04\data\heli_transport_04_base_01_black_co.paa"];; }, _veh] call CBA_fnc_execNextFrame;
yea
though there are still inconsistencies, or i guess the randomisation is not implicitly enabled on all vehicles? for example, spawning T-100 has no effect
private _veh = createVehicle ["O_MBT_02_cannon_F", getPosATL player, [], 0, "NONE"];
_veh setObjectTexture [0, "a3\air_f_heli\heli_transport_04\data\heli_transport_04_base_01_black_co.paa"];
meanwhile on T-140K it works ok, the texture is set
private _veh = createVehicle ["O_MBT_04_command_F", getPosATL player, [], 0, "NONE"];
_veh setObjectTexture [0, "a3\air_f_heli\heli_transport_04\data\heli_transport_04_base_01_black_co.paa"];
yeah, if a vehicle doesn't have textureList[] in config defined, then it doesn't retexture it, that explains it π
So ded, are you interested in fixing the "halo jumping" bug?
When your feet stop touching the ground and your altitude is high (I guess it uses AGL?) you go into halo jumping anim
make it use getPos select 2 for altitude (AGLS) and the bug will be fixed
simple 
or if not you maybe KK is interested? π
π
can you describe the bug in detail plz?
if you place yourself on top of a structure that is too high from the ground (e.g. in this video AGL height was ~175m), the unit goes into halo jump animation as soon as he stops touching the ground
but if you check the AGLS height instead, it won't happen anymore
you can get the AGLS height using getPos _unit select 2
here's the more annoying version
when you reach the top of a staircase your feet stop touching the ground for a split second (due to the sudden change in slope). that's more than enough for the bug to trigger
here are 3 more bugs that @sleek scaffold sent me
Enhanced Movement mod fixes all three of these simply by spawning a helper roadway object under the player's feet and moving it every frame (but it doesn't fix the halo one)
(btw you guys were right; that's why I was seeing the setpos 0,0,0 error in diag exe
)
here's the test mission that @sleek scaffold sent me where you can try the above bugs yourselves:
https://drive.google.com/file/d/1FnCr69Wii-Q0eEbo-caUWYLX9-pHU654/view?usp=sharing
with enhanced movement π
is there a ticket?
dunno. I'll check and if there isn't I'll make one
ok ping me
File description.ext, line 97: Config: ';' used as a separator in the array
File description.ext, line 98: Config: ';' used as a separator in the array
File description.ext, line 99: Config: ';' used as a separator in the array
possible to get mission context for these please (on a DS/when going hosted - so when it parses all the missions desc.ext)
@solid marten
This is the halo jump bug (already existed, I just added two videos + possible solution):
https://feedback.bistudio.com/T75547
there were some tickets referencing the other bugs, but they were too specific (e.g. falling thru the floor of carrier, etc.), so I made a new one
https://feedback.bistudio.com/T161233
Impressive ticket, I haven't seen such detail on one in a long time, well done π
GONE
I assigned the second ticket to myself without reading, so now I read it and unassigned it, reason, it is way too complicated, way over my head, dont want to give you false hopes
okay 
At least we have Bad Benson's EM mod who was able to fix the issue years ago (http://makearmanotwar.com/entry/ypXAHjfPun#.YV8q3tqZNhE)
Fix or workaround?
workaround in a way that wouldn't be acceptable for us
Yeah well, from his position any other thing than a workaround wouldn't be possible...
Having the players stuck on animations, not being able to move and having to kill themselves could be considered even less acceptable and immersion breaking to be honest
Have you guys watched the Splendid Squad mission? Right at the start one of the players had to kill himself because he was stuck on the halo animation: https://youtu.be/Zoi671enFwE?t=11
KK has claimed that one. he was talking about the movement bugs
what Dedmen means is that it's too lazy, and not the optimal way to do it
he's right. it should be fixed properly
but I guess it won't happen 
Yeah but modders don't have access to the engine code, so of course it was not going to be optimal... And yes it could've been even moded better probably, and there has been lots of optimisations done and new scripting commands that were not available to Bad Benson back then...
But as you pointed, too bad it won't be fixed via an engine solution. TBH by now most of us the players have already learned to "laugh" at these bugs... how else would be get Arma'ed!? π It is just bad publicity when potential buyers see a yt video promoting the game like the Splendid Squad and sees those unresolved bugs. Someone should've told Liru the Lance Corporal (the mission maker) to just require Enhanced Movement on all the players
Related to the HALO bug, ~7 years ago I made a feature request (https://feedback.bistudio.com/T79436) to have a command that disables this behavior per unit on demand, since I assumed it was a desirable feature.
This annoyed me while I was dealing with a jumping out of a C130 in MP. The model had a micro roadway gaps where the ramp and the cargo hold connected, which caused the switch to freefall anim.
Also happened when they got pos'd into the cargo hold, sometimes the positioning wasn't precise enough in MP as the C130 was "micro-stuttering" over the network on the client, which resulted in everyone in the hold balancing on their bellies. π
My thinking at the time was that this might be fixable if this behavior had a grace period for triggering, and reverting from it when coming into contact with another surface/roadway.
My game is freezing when I execute:
ropeCreate [cursorObject, [0, 0, 0], cursorObject, [1, 1, 1]]
This is breaking H&M 
Can you send a freeze repo.... oh crap these are only fixed in profiling v2 
Can you send a freeze repo....
?
Arma should generate freeze reports, that tell me where it froze
but that been broken for years
and the fix for it isn't release yet 
Isn't the file I added in the ArmaReport_log zip file ?
Are you going to fix when you have the time or should I create a hotfix for the H&M?
I'll look at it on monday
π€
I got it
you want to me to add it to the ticket?
or DM you?
ticket
Ah eh fix at most on profiling branch, currently there is no update hotfix planned
:/
But I am able to use ropes, and I know others that are able to use ropes. so, its something specific to you
it only happens when you attach a rope from an object to itself
apparently
the freeze dump I sent was from attaching an mrap to itself
also interestingly, I think it froze after I unpaused the game
why would you attach a rope to yourself :u
Β―_(γ)_/Β―
Because you can't attach to a dead vehicle
wdym?
So to fake a towing, you attach the dead vehicle to the alive vehicle
And create a rope attach from the alive vehicle to the same vehicle but to a position where it looks like you attach the rope to the dead vehicle
This picture show an alive vehicle but it also works when the vehicle is dead
really? I thought you spawn a dummy object, attach it to the towed vehicle, and attach the rope tothat
Not necessary before 2.06...
is there any reason that setVariable's public flag doesn't support OBJECTs?
like remoteExec does
if not I guess I'll make a feature request...
plz do, I remember making the mistake this year π
I made it anyway! π
https://feedback.bistudio.com/T161258
dunno why I ask?! π€£
14:45:34 Error: Object(2 : 3700) not found
14:44:42 Creation of object C Alpha 4-1 failed, state LOGGED IN
14:44:42 Creation of object C Alpha 4-1:<Empty main subgroup> failed, state LOGGED IN
14:44:42 Client: Nonnetwork object a2c24e0.
What could be causing that? Are there any commands that print that or so?
Modded Keybinding - Problem with "MODS" Section
https://feedback.bistudio.com/T161263
Youβve run out of groups?
arma could use higher group limit
what is current limit?
not enough?
talk to @gray wharf pretty sure this is per side
in my CTI mission some people have turned the number of groups spawned so high from the parameters that they reached the limit, have to tell them to lower those params π
it's hard to estimate how many groups the mission needs so the limit could be higher...
or I just tell them to lower those params π
Its not my mission.
Never had that issue myself. Thanks for the infoπ
Looks like there is something wrong with BIS_enableRandomization & disableRandomization[] = {"AllVehicles"}; in the description.ext config.
_vehicle = "I_C_Offroad_02_unarmed_F" createVehicle position player;
_vehicle setVariable ["BIS_enableRandomization", false];
Vanilla (No mods /stable): https://gyazo.com/babe0cafe0bdd1f6002e50c94485527a
Modded (Persistent game-mode / server perf build):
Server Restart-1: https://gyazo.com/107be08aca0f3a53f667c13a04cd6100
Server Restart-2: https://gyazo.com/1d44b663a0d63e7ad0a94cad02e3de82
π€
https://gyazo.com/85e101252a4b40e39567e7ce8b4f8d14 its done two colours for me
yea and the stuff like overhead protection, doors, headrests etc also change
[Arma 3 Launcher] Add link to CDLC compatibility mods to simplify user experience
https://feedback.bistudio.com/T157714
wasnt this already implemented? cant see it in the Launcher
getSensorTargets on an empty vehicle seems to crash the game
Exception code: C0000005 ACCESS_VIOLATION at 0796BCA8
Version 2.06.148221
Fault time: 2021/10/09 19:26:49
Fault address: 0796BCA8 00:0796BCA8 Unknown module
also weird that it reports the player as its own sensorTarget, lists the friendly tank nearby in the player group (expected) but doesnt pick up the other tank not in the same group
player knowsAbout cursorTarget turns 4 too
100% repro
Is there a "performance" reason of why is the "FiredNear" EH limited to 50 meters? Or could it be increased now?
AFAIK BIS_enableRandomization has not been working since the new config based disableRandomization[]
Have you used config option? the script in debug console should do nothing
mission param works but not for animations
I used both to be sure on the server test ^^
It still randomises the vehicles to a certain degree (didn't switch colour)
but not for animations
what does this mean?
748.426: No alive in 2000 ms 748.426: check another threads -- add ProgressRefresh()
oh I think I got it. it happens when the game freezes for over 1 second
and it keeps dumping .mdmp's in my A3 logs 
1.5 GBs of mdmp's... wat?? 
can I just disable this?
stop breaking the game
Oh blind iam, thanks π
Will this ever change again ? Or is there some other way to do it now ?
I wasn't breaking the game. this is what I tried to run, which is a perfectly correct code:
_types = createHashMap;
_objs = nearestTerrainObjects [[worldSize/2, worldSize/2], [], worldSize/sqrt 2, false, true];
{
_m = getModelInfo _x select 1;
_o = _types get _m;
if (isNil "_o") then {
_o = []; _types set [_m, _o];
};
_o pushBack _x;
} forEach _objs;
_delta = {
params ["_x", "_y"];
_a = abs(_x - _y);
_a min (360 - _a);
};
{
_ns = [];
{
_vy = vectorNormalized vectorDir _x;
_vz = vectorNormalized vectorUp _x;
_vx = _vy vectorCrossProduct _vz;
_yy = asin(_vx#2);
_xx = _vy#2 atan2 _vz#2;
if (_ns findIf {
[_x#0, _xx] call _delta < 1 && {
[_x#1, _yy] call _delta < 1
}
} < 0) then {
_ns pushBack [_xx, _yy]
};
} forEach _y;
diag_log [_x, count _y, count _ns];
} forEach _types;
btw the game force closes itself for no reason when I run that (only because it takes too long to run, but I'm ok with that)
I had to run it scheduled, but to prevent it from taking too long I isNiled the code inside forEach:
{
isNil {
_ns = [];
{
_vy = vectorNormalized vectorDir _x;
_vz = vectorNormalized vectorUp _x;
_vx = _vy vectorCrossProduct _vz;
_yy = asin(_vx#2);
_xx = _vy#2 atan2 _vz#2;
if (_ns findIf {
[_x#0, _xx] call _delta < 1 && {
[_x#1, _yy] call _delta < 1
}
} < 0) then {
_ns pushBack [_xx, _yy]
};
} forEach _y;
diag_log [_x, count _y, count _ns];
}
} forEach _types;
```which then caused the game to dump many mdmps for me
(the code inside isNil was taking too long sometimes, that's why it dumped many)
so long story short, I'm not breaking the game. it's gotten broken now. the game shouldn't close itself just because a code takes too long
so I guess this "fix" has broken the game
looks like this only happens in diag exe 
normal exe doesn't force close
@sinful kettle is _y defined in your code?
my code is correct
no need to analyze that
and yes, _y is defined
it's a magic var
ok ic
forEach on hashMaps got _x and _y @dreamy bane
ah right
if I ever get arma crashing its because of infinite loops or recursion
it is a bug, so I guess
setObjectTexture has become "Erratic" since the last update on Main Branch. It will sometimes set the texture and other times it wont. On units the command always works, it is just vehicles that don't work 100% of the time. Don't know if there is a limit, I am loading almost 100MB of vehicle textures from the mission directory
it has to be executed with delay now
Good to know! Thanks!
or use BIS_fnc_initVehicle
@untold sky is it a bug that the game force closes itself on diag exe when it freezes too long (script taking long to finish)
I'm pretty sure this didn't use to happen
is it a bug that group members whose leader is dead ignore domove commands? I can make a repro if needed
yes, in the DLC tab
performance
why are you freezing so often?
That message should only be printed in diag build?
Yes, diag/internal exe force kills itself if its been frozen for over 30 seconds
As far as I can see this fix worked exactly as intended, it fixed the freeze logging.
The normal timeout is 10 seconds..
I didn't expect that normal players game will freeze for 10+ seconds multiple times per game run.. And i still don't
The code has always been there, but the freeze checker was completely broken.
I fixed the freeze checker, so now this code works again
You can completely disable freeze check with -nofreezecheck start parameter
I limited the number of dumps per game run to 4 in total. 2 per distinct freeze
Not shown if you own the CDLCs?
Might be yes
copy
Button would be here
Is this one new?
I was testing the code I posted a few messages above, but since the diag exe was force closing, I ran it scheduled and isNiled parts of the code... 
so the game froze multiple times during the isNils
no, atleast A2
there is also -forcefreezecheck and -hidedebugger. But normal people don't need these
I don't remember them being on the biki. Thanks for calling me "normal" π
At times tickets get custom policy with some prio set even if I dont set it. Ref: https://feedback.bistudio.com/T161306
Any idea whats causing that or how to change it back to normal as author?
You created a private ticket
Also priority on immediate oof :U Don't do that
Oh I guess private bug report automatically set prio immediate too
yes it does
Afaik you cannot fix it yourself, you need a moderator for that
thanks
admin actually. I cannot edit those tickets
I remind you if I don't forget it myself
So we both agree to remember to not forget, good
Sounds good π
the moveTo command wont work if player is dead group leader, should i make a ticket?
moveTo
https://community.bistudio.com/wiki/moveTo
moveTois a low-level command used in FSM called withdoFSMorcommandFSM. SeedoFSMfor more information about this. UsedoMoveeverywhere else - even in FSM executed withexecFSM.
I was first trying to use domove but that didn't work when the group leader dies then I tried moveTo (recommended by Leapard20) but that has similar issue with dead leader
try move
also be sure to not have any mods loaded at the time of testing
ok disabling mods
wiki says move moves the whole group
ok tried without mods but same thing
move does nothing, running it for each individual
tried move with player's group as arg but nothing happens , probably because player is the leader...
that, yeah π
want me to create ticket?
dealt with, thanks π
there was not even a link, it's just a genuine recipe without advertisement/SRO hack! π
oh, I missed that. sneaky sneaky
I believe that's only a ZEN feature
you don't use ZEN then?
do you use any Zeus mods?
no. like I said, afaik this feature doesn't exist in the vanilla game. it only existed in ZEN (and maybe Achilles) and got broken by the update
iiit's not a #arma3_feedback_tracker thing
try #arma3_questions / #arma3_troubleshooting @steep marten
@livid sparrow are you aware of that ^ ?
^ moved to: #arma3_questions message
I had a general question: do you guys still fix models too, or is it just game bug fixes due to EOL status?
Models can still be fixed, but depends on which model and how much
like recently reyhard (I think?) fixed up some models with broken geometry
yeah actually this is what I wanted to ask. I noticed that there are several tickets on FT mentioning stutter problems near rock objects. can this be improved if their model complexity is reduced a bit?
or is it more related to unoptimized collision math and stuff?
I don't think changing their geometry to be less complex would improve much.
If you have a repro how to reliably get alot (preferably like.. below 10 fps) of stutter due to rocks, maybe I can find some optimizations there
hmm...ok, I'll try! π
huh, I just tried those missions in the above tickets and I got 20+ FPS drop! π€―
The lower it gets the easier it is to find the bottleneck
well the first one (driving bug) kinda freezes for a bit for me
when I reach near the rocks
but after a bit it stabilizes (due to caching maybe) and my FPS goes down to 40-50 (from stable 60)
the second one again gives me a bit of freeze for like ~0.1-0.2 s when I shoot and the bullet hits the target
I recorded at 60FPS so you can see the freezes better

i think SOG had challenges with fps from big rocks
may want to ask @steady panther about it
Yeah we did and Dedmen has done some optimization for collision checks due to that.
I really need constant low fps if possible.
But maybe I can find the cache and turn it off
I heard with helicopters you get constant low fps? but that might be a seperate issue
well I do get 10 FPS with a heli near those rocks! π
actually no it goes up again after a few seconds to ~46 
too complex collision models at least seemed to be a factor
Yiss that looks pretty bad! (I mean the stutters, the video is great) π
The stutters happen when you hit Ai units rather than just from shooting...
yes that's what I meant... 
Perhaps when a unit is hit the engine does some collision checks that the nearby rocks slows downs or the check just takes too long for some reason
Β―_(γ)_/Β―
but rocks much simpler than vehicles as an example. Why 2 vehicles driving around does not cause such stutters?
Good point... perhaps it is also happening for other objects too not only the rocks but just on a lesser degree?
I believe in that case collision is handled by Nvidia's PhysX, otherwise it's partly handled by the engine 
some of the rocks have hundreds of components
something looking simpler does not necessary mean it actually is simpler
or bug free
So I replaced all the 22 rocks on the repo mission with one of livonia's most complex buildings with interiors (Airfield Control Tower) and tested again... not a single stutter! So it seems these rocks are far more complex and resource intensive than even the heaviest buildings!

that can happen
in essence they are all just "vehicles"
with slightly different simulation types
Well, the fact that it happens doesn't mean that it should happen. Rocks are merely decorative and serve almost no other purpose while other "vehicles" have the right to be more complex and intensive but not the rocks... π
rocks and builidings both share same purpose
they are essentially same
Im just trying to break this illusion of building looks more complex so it has to be
Large size of an object has always been a bane of Arma, hasn't it? π
quite
This was posted on the Necroplague server and it was very funny: "Let's face it, these rocks where probably done by an intern or FNG."
So if I replace those rocks with an even larger object like for example the USS LIBERTY I should get even worst stutters then... right?
Maybe, if it's related to size.
Those giant rocks with VC tunnels from SOG always come to my mind when talking about large objects in Arma π
Hahahaha, so I replaced the rocks with 22 USS LIBERTY destroyers, the most complex object I know in Arma, and had CERO stutters!
So it is not the size (said no girl) π€£
likely some sort of buggy geometry part
Im looking to improve my code that checks to see if a groups knowledge of the players changes. Right now I use code that I repeatedly call using the BIS_fnc_loop function. I was hoping there would be an event handler which would be more efficient but I dont see it. Thanks.
meaningful to make a FT request for that?
Hm since arma already simulates the changing water levels, would it be feasible to create a script command that can change the water level across the map
?
plz no 
use getPos π
would be nice but should asl measure from new water level or old?
Hm, how does it work with the existing tide system?
havenβt seen proper tides
You can already make the water level go very high, but I never checked to see what asl does then
you mean you can mod?
You can edit a maps config to change the max height of the tide, so to say
it breaks many things, let's just forget about it plz
For a video of what I mean: https://www.youtube.com/watch?v=ASkmBry76pI
all good? it breaks the AI cached water levels
(according to Dedmen)
all good as in if you change world asl it never going to change from the moment you load world
script command implies dynamic change
and yes I think it is going to mess up the lot
Also, since the ancient times a lot of arma cool ideas were rejected if they were not going to work with AI as well as human player, this is still the case even today
well you guys also reject ideas simply because you think it can be misused.
like the setAimDirection and setHeadDirection commands I requested which according to you guys it can be used "to cheat" (which btw, scripting = cheating, as you see people cheat every day in public servers)
or the other thing I requested was being able to get all object model surfaces in a certain LOD (only GEOM, VIEW, FIRE, and ROADWAY) instead of having to do millions of lineIntersectsSurfaces
and you guys say it can be used to "rip" models...
first of all, there are better ways to rip models, otherwise how do all these rippers get their models already?
second of all, who would rip a model from those LODs? They are extremely bad looking and are used for very specific purposes 
Dedmen or Killzone Kid would you please look into a few getter and setter related to AUTOCOMBAT please.
For SP play/player under AI command/AI groups the long timeout for groups to return to normal is affecting the experience way too much in negative ways (very slow movement and the combat shout spam even with no more enemy nearby - 5 minutes timeout!).
There is disableAI yes, but removing it altogether is not desired either. Just shorter timeout/flexibility to adjust dynamically would help to improve AI behavior a great deal.
Thank you for the consideration πββοΈ
ref:
https://feedback.bistudio.com/T156436
https://feedback.bistudio.com/T161510
https://feedback.bistudio.com/T155105
Yes but oh please no.
Its already not properly supported concerning AI pathing, people would expect that to work right if it becomes a script command
tru makes command for only himself
it was made with the unofficial 2.00 OFP patch iirc ^^
Might be possible, can't really say without looking into it.
And I won't look into it without a ticket and free time
thing is, should it impact ASL commands
Yeah possible, but do I want to invest the time? eh
As mentioned before, what happens at the moment when the sea level changes a lot?
Any chance you can put it somewhere at the bottom of the ideas list at least
?
Ah ye, would be interesting to test what happens when you mod the sea level 
Making ASL heights change by that might not be feasible.
It depends on how commands are implemented
LOOOOOTS of commands expect that ASL == World and never check actual sea level
updating these is not an option
Yee
If I get around to it later, I'll test what the current behaviour is with strongly changing sea levels
waiting on choppers to be spawned into the water
Its fun when you load into a map in 3den and there's just 2000m high water above the map
Just as fun as my livonia island I imagine
I tried to enable tides on "our" map and it broke lots of things. I think there is a reason why tides & bigger waves are disabled.
(Invisible waves damaging vehicles parked close to the shore / missions spawned under water even tho i used a surfaceIsWater exit condition).
With bigger waves you just pushed through the waves and your boat sank ^^
With bigger waves you just pushed through the waves and your boat sank ^^
Oh I've got a video for that, one sec
This is from back when I broke the tide system a little bit too much https://www.youtube.com/watch?v=idDldrwKpuw
But ye, the tide system falls apart pretty quickly once you start playing around with the values
or well starts causing weird things (like the ones you mentioned), it doesnt fall apart that quickly
I tried with much more reasonable values, still didn't work ^^
class Sea
{
seaTexture = "a3\data_f\seatexture_co.paa";
seaMaterial = "#water";
shoreMaterial = "#shore";
shoreFoamMaterial = "#shorefoam";
shoreWetMaterial = "#shorewet";
WaterMapScale = 20;
WaterGrid = 50;
MaxWave = 0.25;
SeaWaveXScale = "2.0/50";
SeaWaveZScale = "1.0/50";
SeaWaveHScale = 1.25;
SeaWaveXDuration = 5000;
SeaWaveZDuration = 10000;
};
That's the max i found still working for the waves.
seaTexture = "a3\data_f\seatexture_co.paa";
hmm now I kinda want to start playing around with different textures
would be nice to have a command that changes the max scheduled script running time
it can be misused a lot
plus you can already it do it yourself using isNil to execute more commands at once
a lot of things can already be misused π
I use EachFrame for faster execution but would be nice to give more time to scheduled scripts too

rip


