#arma3_feedback_tracker

1 messages · Page 2 of 1

real bronze
#

you can just look at the commandmap? i had no issues with this, grammar being
expr: primaryexpr (binop expr)
where primaryexpr is constants, arrays, and unary/nular expr, you check if a unary cmd exists, if it does, lookahead primaryexpr and parse as unary if it exists, else try nular, else parse as ident

sinful kettle
#

well like I showed it's ambiguous. you can show the same thing for nular/binary combo

real bronze
#

a unary variant can exist

sinful kettle
#

you check if a unary cmd exists,
first you should check nular

#

unary and binary don't matter

real bronze
#

either way you need to look at unary map

sinful kettle
#

if not nular, it must have a right arg

#

if binary it must have a left arg

sinful kettle
#

what if it's not? blobdoggoshruggoogly

real bronze
#

written correctly?

#

an error will occur somewhere if it's written incorrectly

#

that totally depends how you implement it

#

i can't see how this is ambiguous, sqf top level expr is literally primaryexpr ((binop expr) | ';') if you follow this rule, it can never be ambiguous assuming you're looking at command map everytime, op op op op [] op op
parse expr
parse primary > unary op, nular op
binop or semicolon > binary op
parse expr
parse primary > unary op, []
binop or semicolon > binary op
parse primary > nular op
simplified in a way as it doesn't do anything with precedence, for binary it's more work yes, you need to lookahead, but unary variant is totally viable

actually, in the example it only works because there is a terminal in there, so yeah binary actually is not possible (didn't think through), but unary is no issue. technically even if it had the same binary name, parser shouldn't be guessing and just parse following normal precedence rules, nular>unary>binary, saying that "assuming code is valid" is not really a thing, that's a developer's error then, , i.e., op op op op op [], and you intend this to be unary unary unary binary unary then you forgot parentheses, it's literally same as +++++[]

alpine tulip
#

Can we have some love for EpeContact EH? Currently it is almost impossible to use them as a proper PhysX event getter

#

Like add returns of vector of force, force position, etc

solid marten
#

Make a ticket so I won’t forget to look at it if there are more params available to return

alpine tulip
#

Rog

unborn acorn
#

Can be added support for underbarrel smoke grenades? like
_projectile addEventHandler ["Deflected", {};
for now I can hit EH only Explode/Deleted for 40mm smokes

limpid rune
#

it's cuz its shotSmokeX which is physX its the same with shotGrenade - neither support it

solid marten
limpid rune
sinful kettle
gray wharf
#

oh, you can ask

solid marten
#

but each intersection returns the object as well can you not compare objects

sinful kettle
#

an object can have multiple convex parts tho meowsweats

solid marten
#

so it was not meant to be used as object drill, just to find surfaces, does it do it well?

sinful kettle
#

Sorry that I wanted to do something a little more advanced than others blobdoggoshruggoogly

#

And no, it doesn't do it well, as I've said before

#

Here

sinful kettle
vocal plover
#

Port old Argo assets into Arma 3

alpine tulip
#

“No promise” - Dwarden

vocal plover
#

some1 already asked? mb

alpine tulip
#

This is actually an FAQ

vocal plover
#

where can I read the faq

alpine tulip
#

I don't mean it is listed somewhere. It literally is Frequently Asked/Answered Question

vocal plover
#

oh lmao tyty

gray wharf
devout wave
untold sky
#

intersting

gray wharf
#

I'll keep it - just edited the URL out 😄

#

aaah no, you hid it already 😄 welp my parallel edit reverted it to public

sinful kettle
fresh hawk
#

https://community.bistudio.com/wiki/actionKeysEx
On the wiki page it says isDoubleTap: Boolean - is the action key a double-tap. Only possible when there is just the Main Key, as key combinations cannot have a double-tap flag
In game I can set and use keybinds such as U + 2xO, which (to me) have a main combination of doubletapping O and U as a combo key in this example.
The command however doesn't return the double tap flag for the example above (and similar ones).
I'm not sure if I have understood the wiki page correctly or if it's a bug?

actionKeysEx
solid marten
#

@untold sky ^^^

untold sky
#

@ NeilZar on monday :harold:

west onyx
#

there appears to be an alternate syntax to "nearSupplies"

#

do we know if it works?

alpine tulip
#

At least [0,0,0] nearSupplies ["",0] doesn't throw an error

#

Didn't throw an error, but I couldn't make it happened neither

solid marten
opaque jungle
#

you must be joking dude the update ruined was was very realistic thermals

solid marten
gray wharf
gray wharf
gray wharf
sinful kettle
slim hemlock
#

Does anyone here use Nitrous Network to host a server?

gray wharf
tribal rivet
#

Not sure if I should report that or #wontfixDontEvenBother, but engine question:

Is it normal that str(9227465) yields "9.22747e+006" already?
I understand that the engine is storing all numbers as floats, so you'll eventually get a scientific notation, but in this case the converted number is "only" 10 millions and the scientific notation gives you a rounding error here, already (9227470 instead of 9227465).

Not sure what exactly I would like to see here. Maybe use the regular notation for all floats that are ints? Maybe (if it's parametrizable) try to convert to scientific notation "later" at the cost of longer strings? (but more precise conversions)

gray wharf
#

I believe the notation is not precise either
try toFixed

tribal rivet
#

for all floats that are ints?
After all, you can't get more than 340282346638528859811704183484516925440 so it's not that problematic to use the regular notation (not even sure how it's called)

regal nimbus
#

I guess ideally it'd only switch to scientific notation when the exponent changes?

tribal rivet
regal nimbus
#

Although the alternative point is that the scientific notation is losing precision due to being one digit short.

gray wharf
#

see the alternative syntax of toFixed

#

it sets the general setting, not per-value

tribal rivet
#

So it's not like the value in memory will already have that error, it's literally just the formatting function

gray wharf
#

use toFixed to prevent any input issue anyway 😛

tribal rivet
#

Sadly 😦

#

But good find, about that comment there

gray wharf
#

I agree it's a pain, but it's also a good practice
imagine we suddenly decide that the significant figures is 3 😄 boom, broken

#

better secure the inputs, especially for extensions

tribal rivet
#

Okay, so in that case, was there a specific reason for "6" to be selected? Besides the fact that "we need to pick a value"? 🙂

Could that value be increased? As I said, losing presision at 10mil is a bit low

solid marten
#

there is tofixed global option added with extensions in mind

gray wharf
#

Could that value be increased?
yes, with toFixed alt syntax

solid marten
gray wharf
# 
isNil {
  toFixed 50;
  "Extension" callExtension ["func", [_myNumber]];
  toFixed -1;
};
tribal rivet
solid marten
#

it cannot be increased due to backward compatibility, there is a ticket somewhere when it has been done the reverted

gray wharf
gray wharf
#

you don't even have to str _myNumber yourself!

tribal rivet
#

I missed the "alternative" word in the sentence and went googling the toFixed part 🙂

gray wharf
tribal rivet
#

Okay, thank you, both of you. I'll see if I can come up with something that works 🙃

#

Interesting to know that changing that number caused issues, though 🤔

gray wharf
#

it always comes to bite you when you expect it the least 😄

tribal rivet
#

There is no way to make it conserve ints as ints, is there?
Because the moment I start using toFixed with any value (other than 0, obviously), it starts adding the fractional part, even for integers. Other than that, it toFixed would be the ideal solution for me

gray wharf
#

ints are not a thing in RV

tribal rivet
#

I know, but str(123) gives you "123" whereas toFixed 2;str(123) gives you "123.00" (the fractional part is always forced).
So that means that RV can omit the .00 by itself, if not needed

#

toFixedMax would be ideal, here 🙂

gray wharf
#

[_bigNumber] call BIS_fnc_numberText splitString " " joinString "" 😆

solid marten
tribal rivet
# solid marten try tofixed 0

Yes, but this then takes away the possibility to pass in floats.
I'm letting people create their own python functions with their own sets of arguments, and then call them from SQF. So let's say that someone wants to do an AAR and log the large ID of a player (or whatever else that is large) and a player location.
They will then do

["logger.log_position", ["Stack", 123456789, 30.12, 64.33]] call py3_fnc_callExtension;

In this case, I can either use toFixed 2 to force all the numbers to have two fractional digits (making a float from the ID) or I can use toFixed 0 to keep the ID correct, but then the floats will be truncated to ints.

Note that if I pass

["logger.log_position", ["Stack", 1234, 30.12, 64.33]] call py3_fnc_callExtension;

then everything will serialize fine. The int will still be an int and the floats will serialize as floats

#

But I'm guessing there is no simple solution to this, based on this discussion 😦

gray wharf
#

yep
floats are floats, ints are ints
make your own function if you will ("if last 3 chars are ".00" then substring(0, -3)")

tribal rivet
#

I was happy to be able to let the users create whatever function they wanted and pass in whatever arguments they wanted, because "it just worked" so I wouldn't want to tell them "but you can't pass in arrays of arrays of arrays anymore, because I have to go over all the arguments and convert them correctly now" 😦

gray wharf
#

you never had ints in the first place, make the extension deal with that?

tribal rivet
#

if last 3 chars are ".00" then substring(0, -3)
Heh, this is actually something that could pan out, because I could do it in C++, when converting individual values to PyFloats, so that would be quite efficient

gray wharf
#

yeah, that's the extension's responsibility I would say
RV can only provide floats, eventually displayed as ints
now it is up to you to secure your output in code, and input in extension

tribal rivet
#

Yup, I just checked and the default behavior of str(12.00) is "12" so I guess I'm gonna go with that.
Thanks! 👍

#

I mean, it's more important for me to let people pass in big integers as integers because I have a gut feeling that more python functions will break when you'll feed them floats, when they expect ints than the reverse (while keeping a reasonable float precision, at the same time)

gray wharf
#

it is YOU who deals with your input in the extension
if a wrong input breaks the function it deserves to be broken

# 
void MuhFunction(string argument)
{
  int expectedValue = int.Parse(argument); // throws an exception and doesn't go any further if you send e.g 'potato'
  StoreInDb(expectedValue);
}
tribal rivet
#

It's the end-user (modder) who controls both SQF and the python code called by my extension.
It is reasonable for them to expect that if they pass in an int from their code on one end, they get the int in python.
Even if technically, there are no ints in RV. That's because as I said, str(<int>) -> "<int>" in most of the cases. If it would always yield "<int>.0", then I'd agree that it would be unreasonable 🙂

#

I totally get your point, though.
As previously said, I was trying my best to make the conversion between SQF and Python as transparent as possible, hence the report

#

Turns out it's simply not possible

gray wharf
#

I think I understood - not certain but yeah, due to float, no "int" really possible

#

it is just a representation

regal nimbus
#

wellll... it's a perfect representation up to 2^24 :P

gray wharf
#

has the canAddItemToBackpack bug mentioned by AgentRev in the notes been reported/noted/fixed anywhere?

#

no FT results apparently

sinful kettle
#

also you need many more iterations:

onEachFrame {
  for "_i" from 0 to 3000 do {
    drawIcon3D ["", [1,1,1,1], ASLToAGL eyePos player, 1, 1, 0, str _i]
  }
}
#

with one it's gonna take too long to trigger the freeze

desert trench
#

Trying to call RemoteExec(Call) with 0 targets for func 'bis_fnc_advhint'. Maximum is 1000
A null object passed as a target to RemoteExec(Call) 'bis_fnc_objectvar'

#

would it be possible to get call stack or more context for these?

#

they get spammed a fair bit, but hard to pin point

#

aka worth to a make a FT ticket?

#

Tried to create group while group limit is reached on side Game Logic

#

another case

solid marten
solid marten
gray wharf
#

one BILLION
plus one

solid marten
#

somehow I think even that wont be enough

gray wharf
#

dang

solid marten
#
  • remoteExec everything
sleek scaffold
solid marten
#

but it is 288 per side, so you can create 288 game logics

#

not enough?

#

you can have unlimited number (sorta) of logics in 1 group if you need more

sleek scaffold
solid marten
#

you can have agents instead of units that dont need group, but if you want units then again, 288 units per side? group them

sleek scaffold
#

Would be neat if we are able to add Agents from the EDEN editor perhaps as a checkbox on units... "isAgent" checkbox

gray wharf
#

3000+

solid marten
#

Can pc handle that many units?

gray wharf
sleek scaffold
# gray wharf

I guess server FPS is the true limitation there...

gray wharf
#

iiiiii would suppose so 😂

solid marten
gray wharf
#

yeah but how many 4 real

sleek scaffold
sleek scaffold
gray wharf
# 
private _unit = player;
while { not isNull _unit } do
{
_unit = group player createUnit [typeOf player, [0,0,0], [], 0, "NONE"];
};
count units player; // 10001
``` Cc @solid marten 😃
solid marten
#

before it crashed?

gray wharf
#

it froze quite some times 😆

unknown.png
#

on Stratis 😄

unknown.png
solid marten
#

I wont even attempt it on my 8 core PC

gray wharf
#

given not all are used, safe way

sleek scaffold
#

So the 3.000+ is actually 10.001?

gray wharf
#

not sure about the +1, as it may be a weird null added
I wrote 10k in the wiki

sleek scaffold
#

Can you add another 10.000 to another group?

gray wharf
sleek scaffold
#

For the crash...

gray wharf
#

nah 😛 crash your own PC!

sleek scaffold
#

I do wander which quantum computer was used to define that 10.000 limit... I should be perhaps 1.000

gray wharf
#

I rather believe some "quick table corner calculation" was made and that was it

#

so 288 × 10000 = 2 880 000 units per side, 5 sides…
OK, who is up to try and run 14.4 M units on his PC? 😄

rose cloak
#

and report how many days it took to calculate one frame?

sleek scaffold
sleek scaffold
vocal abyss
gray wharf
#

I didn't even think of that, thanks 🍻

# 
21:02:22 No alive in 10000 ms, exceeded timeout of 10000 ms
21:02:33 No alive in 21062 ms, exceeded timeout of 10000 ms
21:02:44 No alive in 32031 ms, exceeded timeout of 10000 ms
21:02:54 No alive in 42046 ms, exceeded timeout of 10000 ms
21:03:04 No alive in 52046 ms, exceeded timeout of 10000 ms
21:03:14 No alive in 62046 ms, exceeded timeout of 10000 ms
21:03:24 No alive in 72046 ms, exceeded timeout of 10000 ms
21:03:34 No alive in 82046 ms, exceeded timeout of 10000 ms
21:03:44 No alive in 92046 ms, exceeded timeout of 10000 ms
21:04:56 No alive in 10000 ms, exceeded timeout of 10000 ms
note: Minidump has been generated into the file C:\stuff\Arma3_x64_2022-09-05_20-50-28FROZEN22.mdmp
21:05:08 No alive in 21375 ms, exceeded timeout of 10000 ms
21:05:18 No alive in 32000 ms, exceeded timeout of 10000 ms
21:05:28 No alive in 42000 ms, exceeded timeout of 10000 ms
21:05:38 No alive in 52000 ms, exceeded timeout of 10000 ms
```:D
#

sooo…
I tried on Arma 2

unknown.png
#

let's try with logics

vocal abyss
#

Get bigger RAM? 😛

sleek scaffold
gray wharf
# 
0 spawn {

_unit = player;
_i = 0;
while { not isNull _unit || _i < 50000 } do
{
  _unit = group player createUnit ["Logic", [0,0,0], [], 0, "NONE"];
  _i = _i + 1;
};
hint str count units player;

};
gray wharf
#

so… I locked my main computer for now, but Arma 2 is going over 11k "units" (logics)

#

15k*

#

18k… can someone read the code please 🤣

rose cloak
#

19k, it holds me hostage, send help

gray wharf
#

basically

#

I'll try with ArmA after that (if I still have a PC after that)

tribal rivet
#

👀
😢

2022-09-05_22h00_18.png
gray wharf
#

25k 🤣

#

unbeknownst to me the limitation has been entirely lifted most likely 🤣🤣🤣

gray wharf
#

I'll ask another Arma 2 patch just for that

tribal rivet
#

I expected nothing less from you!

gray wharf
#

OK, 26481 units in Arma 2, and one get added every 30s or so 😂
I may stop here

tried to take a screenshot but the PC is unresponsive most of the time 🤣

#

got it!1!

unknown.png
rose cloak
#

just make a photo with your phone. "Don't you guys have phones?!" (c)

tribal rivet
#

Yeah! Take a Screen Shot 😉

gray wharf
#

now, ArmA

tribal rivet
#

It's spelled Arm... wait think_turtle

gray wharf
#

ArmA freezes, IDK why.

0 = 0 spawn {

_unit = player;
_i = 0;
while { not isNull _unit || _i < 50000 } do
{
  _unit = group player createUnit ["SoldierWB", [0,0,0], [], 0, "NONE"];
  _i = _i + 1;
};
hint str count units player;

};
```spawned, even with a sleep in it, doesn't matter
a single createUnit is fine, a `while` freezes the game, even with `_i < 50`
#

800+ units, so no 144 limitation for ArmA

#

2k+, etc

#

yep 5000+, that should do

rose cloak
#

inb4 game actually freezes on radio chatter and not on AI

gray wharf
#

I almost crashed A2 at first, I did a hint str units player instead of a COUNT units player 😄

#

pressed ESC at the right time

gray wharf
gray wharf
#

aaand 12 in OFP. done!

gray wharf
sinful kettle
#

I mean draw3d commands

#

They should not register any draws when the game renderer is not active

solid marten
#

there is dedicated event handler for it

#

there is ZERO reason using oneachframe vs Draw3D. You point shotgun at your foot, pull the trigger and then start running crying to ban shotguns

gray wharf
#

though I hate to say that, but I would tend to agree with Leo here 😛
why can one stack draw events? it's like playSound when sound is disabled, the game should just ignore it

#

of course, if the game is paused we might want to still draw
but unfocused should just drop the drawing

fresh hawk
desert trench
#

they use logic as dummy objects

#

ie if you have enough locations defined in a terrain Combat Patrol or Warlords, plus BI function library, it can create beyond the limit

#

also CAS support module does it

#

post about the bad coding design in the FL a while a go

#

it should cache logics for use, instead of creating a new each time

#

so my question was

would it be possible to get call stack or more context for these?
they get spammed a fair bit, but hard to pin point

desert trench
cold meteor
#

Is it me or there are no sandstorm/snow storm particles effects in Arma???
What a shame.

alpine tulip
#

Then make one?

west onyx
#

just want to climb ontop of the requests and say Thank you to the devs for the attention and changes already implemented

gray wharf
#

Denied, ticket closed

gray wharf
cold meteor
#

Great! Thank you!

west onyx
#

github down for anyone else?

uncut briar
#

IIRC there was a snow model in A3

gray wharf
#

(had this at one point)

unknown.png
gray wharf
#

it still feels a bit weird to me that the rocks are so warm during Winter night but hey, at least it's not all black anymore ^^

unknown.png
solid marten
#

reduce contrast

#

with the new script command

alpine tulip
#

But reducing the contrast doesn't make it (relatively) darker?

solid marten
#

contrast - light is lighter dark is darker

untold sky
untold sky
untold sky
gray wharf
#

the rocks should keep daytime temperature for longer than the ground, then be cool as well and take time to reheat
IDK if it counts as nitpicking or real temperature difference between rocks & surroundings

untold sky
#

rocks use the same cooling/heating up stuff as units and vehicles and houses.
Sadly that stuff is generally pretty crappy, and its also bugged, before noon and after noon happens some weird wrong switchover.
You can best see that on terrain.
People all like "terrain has all same color"
Well, 50% correct
Its sad

unknown.png unknown.png
gray wharf
#

sniff
but thanks for the explanation 🙂

untold sky
#

Also if you wonder why fog is bugged.
Fog works on rgb, thermal is grayscale.
Because reasons, the thermal rendering uses RGB rendering, and for precision reasons splits up the 0-1 thermal range, into RGB values each value doing 0-0.333
Fog is applied to these split up rgb values, instead of to the actual thermal temperature which makes everything all kinds of yuck
And because fog also works with fog color, and everything is built to apply colored fog, and thermals have no color....

Getting these changed would be the next step in thermal overhaul, it was too much work and too big of a change to do it with the first iteration

rare zephyr
# 

private _a = getTerrainHeight [3202.5,11921.25]; 
setTerrainHeight [[[3202.5,11921.25 + 3.75, 28]]]; 
private _b = getTerrainHeight [3202.5,11921.25]; 
_a - _b;

if you run this on tanoa it will return 0.00120163 instead of 0.
i assume that that's a bug with the getTerrainHeight command, since i did not modify the terrain point i'm looking at, just the neighboring one.

#

it happens on other coordinates and maps as well, this is just the first one i found

untold sky
#

0.00120163
ehh, negligible?

rare zephyr
#

not if i try to compare positions using isEqualTo

untold sky
#

Don't do that then

#

floating point numbers anyway

#

The problem is, setTerrainHeight works on grid points, getTerrainHeight does not.
It converts to grid, and back to normal coordinates, and then gets terrain height of these normal coordinates. With slight precision errors it might not hit the point exactly and do stuff like that
What you have there is 1.2 millimeters

rare zephyr
#

i see, in that case i'll try to work around it

untold sky
#

Just looking through my old thermals stuff, this image is the old thermals before the changes, but one had its histogram adjusted by hand.
Stones and trees were always glowing :harold:

screenshot-12.png screenshot-11.png
daring wagon
solid marten
#

it already triggers event

daring wagon
#

Oh, I thought

forceEH (Optional): Boolean - will force lbSelChanged EH to fire
would prevent it from firing the EH if set to false.

#

Nevermind then.

regal nimbus
#

What's the change there? In previous versions it always fired the EH, correct?

#

So the default to forceEH is true?

solid marten
#

no default is false

#

added to biki

regal nimbus
#

Is that a change then? Because I'm pretty sure I hung Arma with that a couple of weeks ago :P

#

might have been the checkboxes though.

#

on that note, cbSetChecked could do with a similar EH parameter.

solid marten
#

Does it not trigger CheckedChanged EH when you execute the command?

regal nimbus
#

It does. I'd rather it didn't, so you can distinguish between a user action and a cbSetChecked.

west onyx
#

would be great to have a way to detect when a vehicle is being repaired/rearmed/refueled from vanilla mechanics (like coming close to a Bobcat)

solid marten
#

hence the param

#

because cannot just hard fix it after so long

solid marten
west onyx
hallow sun
#

So i made a "strip corpse" function that puts all player corpse gear into a groundholder to avoid dupes and other bugs related to dead player gear (https://www.youtube.com/watch?v=TnAUPw96Ors).

  • Player loots uniform/vest/backpack from a player corpse and it's still on the corpse.

It uses this below to clear the corpse (Executed on the server side).

_targetCorpse setUnitLoadout (configFile >> "EmptyLoadout");```
So sometimes this doesn't work and the corpse still has all gear.
Also tried this below, which didn't work either on the same corpse.
```sqf
removeUniform _targetCorpse;

So i tried to execute it again and again on the serverside and it refuses to work.
I thought it might be due to locality because the player didn't re-spawn yet.
So i tested it a few times on players that didn't re-spawn yet and it worked without a problem.

So what could be the issue here ? Maybe the same reason that causes the problems with gear (uniform/vest/backpack) from dead players in the first place ? 🤔
This bug is super annoying for years now notlikemeowcry

solid marten
#

@hallow sun how long the corpse has been dead

hallow sun
solid marten
#

yeah i can believe it, in about a minute the corpse leaves the group and then later it becomes just a model

#

who knows what stops working

hallow sun
solid marten
#

maybe person leaves server?

hallow sun
#

doesn't matter either ^^

solid marten
#

or disconnects but it doesnt register straight away

hallow sun
#

just tested on a player that died 20 mins ago when i wasn't even on the server when he died.
he was also disconnected already and it worked (he disconnected two minutes after he died).

sleek scaffold
devout wave
devout wave
#

Only now realising my proposal example has an unary command with the argument on the left. i am very smrt

gray wharf
#

I'd add an is for is***Disabled getter

desert trench
#

it seems the action menu width spacing doesnt take into account the hold action key press action naming

#

worth to report on the FT?

west onyx
#

that looks like some formatting + align right issue

#

imo not a big deal

desert trench
west onyx
#

maybe experiment with whatever formatting hold action uses

desert trench
#

it looks like its ignoring special characters to compute the width

west onyx
#

can create an alias kju_fnc_holdactionadd?

desert trench
#

//add the action
private _actionID = _target addAction [_title, _codeInit, _arguments, _priority, _showWindow, ACTION_HIDE_ON_USE, ACTION_SHORTCUT, _condShow, ACTION_DISTANCE, _showUnconscious, ""];

#

p:\a3\functions_f\holdactions\fn_holdactionadd.sqf

# 
private _keyNameRaw = actionKeysNames ["Action",1,"Keyboard"];
private _keyName = _keyNameRaw select [1,count _keyNameRaw - 2];
...
_title = format["<t color='#FFFFFF' align='left'>%1</t>        <t color='#83ffffff' align='right'>%2     </t>",_title,_keyName];```
west onyx
#

what if you remove the “call” from the action text, still over length?

desert trench
#

so it seems BI tried to work around it adding those space characters as part of the string

#

its certainly not about the overall width of the action menu

desert trench
# desert trench

like these use no extra magic and just the plain string as shown without extra whitespaces

dreamy bane
#

i find it bit tricky to work with extensions when you run both server & client on same PC because one of them gets to read/use the DLL but the other cannot because the DLL becomes locked , or something.. and you get the CallExtension 'extNameHere' could not be found ... message

#

I solved this by creating copy of the DLL , one for the client and one for the server. but of course this is not ideal 😄

solid marten
#

why don you set server in own folder?

dreamy bane
solid marten
#

yeah I guess ctrl+c ctrl+v is overkill

dreamy bane
#

maybe I just write batch script to keep the DLL in both folders up-to-date 🙂

dreamy bane
#

wouldn't it be good idea to improve those CallExtension "could not be found" messages for locked files, it would help people to solve the loading problems

solid marten
#

what does it say now?

dreamy bane
solid marten
#

how do you repro it from 2 different exes?

dreamy bane
#

run server, call extension from there. run client , try to do same thing and you get that message

untold sky
#

That error message is correct though, could not be found means could not be found. If it is found but fails to load, you get could not be loaded: reason

dreamy bane
untold sky
#

Ah the "file exists" check, is not a "file exists" check, its actually a "can I read this file" check

#

But that shouldn't fail even if someone else has the dll loaded

sinful kettle
#

.NET is weird. you never know 😅

#

like that 2 DLL issue meowsweats

dreamy bane
#

had this problem just yesterday, client & server needs their own DLLs

sinful kettle
#

can you check with a C++ extension?

dreamy bane
#

umm maybe bit busy atm

untold sky
#

ah, the "file exists" check is actually a "can I read this file" AND a "can I ImageLoad this file" check.
https://docs.microsoft.com/en-us/windows/win32/api/imagehlp/nf-imagehlp-imageload
This might fail if someone else has the DLL loaded.. maybe this needs write access for some reason.
This is done to read whether the dll is x64 or not.... That can be done with simple read access too instead of using this thing

And this failing is considered "dll does not exist in folder", which.. is nonsense because you just did a "can I open this file" check just before this, so you know it exists already...

solid marten
sinful kettle
#

@solid marten btw I noticed that when you delete a unit in 3den their "holders" are not deleted.
could it slow down the editor when you place and remove a lot of units?

dreamy bane
#

will get back to this topic a bit later...

solid marten
sinful kettle
#

I think they're the containers on uniform, backpack, etc.

untold sky
solid marten
#

those are out vehicles, do they get deleted if you create unit delete unit with script in preview?

sinful kettle
#

yeah in preview it's fine

solid marten
#

3den is not simulated in the same way so i guess they are marked for deletion but need a simulation to deal with it, either way unless it causes problem i’m not touching it

sinful kettle
#

yeah I figured that could be related to some of the container bugs that people report meowsweats

#

but if it's not the same then nvm

solid marten
#

you can start preview and see if they are still present in the mission, if they are then it is a problem

sinful kettle
#

in SP it was fine. let me test MP

#

MP is fine too...

#

btw is there a way to simulate high ping on LAN?

untold sky
#

there are tools for it yes, commonly used by cheaters

sinful kettle
#

I found a way to break the game 😅
server:

unknown.png
#

client:

unknown.png
#

is that useful?! meowsweats

solid marten
#

only if you have repro

sinful kettle
#

I do

solid marten
#

then it is useful

sinful kettle
untold sky
#

yes

#

First thought that may be related to the thing where units run around without clothes, but.. doesn't seem to be as you are just switching?

sinful kettle
#

switching?

#

after doing some thing the server's player appears naked to the client, and vice versa meowsweats

untold sky
#

I have hopes its the bug that sometimes causes players to appear naked to others

sinful kettle
#

what I did was pretty hacky tho meowsweats

#

but I hope it's at least related

fresh hawk
sinful kettle
#

@neon swan 👆

dreamy bane
#

so, sounds like ded got the message thing figured out?

untold sky
#

no, still doesn't make sense that it cannot open it, it should be able to even if another process has it loaded

dreamy bane
#

hmmm well i have no ideas but Ive learned arma is quite picky on the file access to other processes in general....

#

sometimes have to alt-tab back to arma to "release" file

untold sky
#

yes for pbo's and config files, but this is not using the Arma filesystem

dreamy bane
#

ok

tribal rivet
#

Wild guess but make sure your AV isn't interfering.
I spent over 4 hours a week ago debugging my extension that caused interpreted (!!!) python files to literally silently disappear from the filesystem (but they couldn't be replaced with a new file because they magically were present... somehow). Schroedinger files, so to speak.
In the end, after 4h, it turned out that... I really need to get rid of Avast 🙄

sinful kettle
gray wharf
solid marten
#

@gaunt depot Are you on profiling? I'd like you test deleteVehicleCew

gaunt depot
#

@solid marten
Works on profiling build ["Arma 3","Arma3",210,149939,"Stable",true,"Windows","x64"]

#

Alternative syntax doesn't work at all anywhere, only primary one

#

Ok, doesn't always work

#

Seems to work fine for dead units on turrets, no matter where you call deleteVehicleCrew from, body always deletes instantly everywhere

#

When unit is on passenger seat, it breaks again

#

When you do the delete from body owner's client, it instantly deletes it locally, then after few seconds it deletes on another client but never deletes on server

#

When you delete from other client or server (where body is remote), it deletes it instantly properly

#

Deleting dead body from driver works fine from all sides too

#

In total, what i tested and results with Listen server and 2 clients:

  • Deleting driver seems to be working regardless of locality or where command is called from, I'm not sure I tested all combinations, but most I tried worked
  • Deleting turret produces same picture, works fine in most locality combinations
  • Deleting cargo is broken when vehicle is remote if you do it from where dead body is local, if you call it from where body is remote (both server and another client), it works fine
#

With so many combinations this might need some kind of scripted test to check all possible combinations

#

Either way, alternative syntax never works in any of these

#

Ok, from more testing the only non-working case is when client (not server) deletes local body on a passenger seat of a remote vehicle. Remote vehicle might be server's or another client's.

#

Alternative syntax doesn't work at all

solid marten
#

Ok, I kinda hoped to avoid modifying move out message but I guess there is no choice if you want to do it properly

#

alt syntax should work it just does it for each crew what command does for given crew

gaunt depot
#

In all my tests it did nothing

#

Oh wait

#

deleteVehicleCrew vehicle

solid marten
#

yes?

gaunt depot
#

vehicle, not unit

solid marten
#

lol

gaunt depot
#

So nevermind my comments about alt syntax then

solid marten
#

i guess it will fail for cargo

#

the problem with cargo you have to move out where vehicle is local but for the rest where unit is local

#

i’ll make it proper

gaunt depot
#

So much less RPT spam once I stopped using deleteVehicleCrew, I guess it was the main reason for it

# 
2022/09/13,  7:34:12 Client: Object 532:461 (type Type_462) not found.

Still plenty of this stuff though, wish message names weren't secret to try to guess what's breaking.

solid marten
#

My keyboard is busted, T is not working properly, impossible to write code and stay calm, so will redo it when new KB arrives

tribal rivet
#

If you were a youtuber, you could record a "No-T-letter programming challenge!" and earn an extra buck with it, instead 🙃

gray wharf
#

a youuber you mean

#

the day KK became a chauffeur

solid marten
west onyx
#

ive been having some issues with keys too

#

reboot seems to fix, but very odd. quite an expensive keyboard too

hallow sun
#

Not a big deal, but when changing PIP quality settings ppEffect effects stops working for PIP till you restart the game (modded, not tested in vanilla).

gaunt depot
solid marten
#

they are membranes, anyway got exactly the same kb second hand this morning

#

seems working ok

solid marten
gaunt depot
#

Tested it on a turret with isPersonTurret = 2, but not turned out (so static animation, not using FFV)

#

It deleted fine regardless of locality combinations

#

Tested many combinations but somewhat randomly, maybe not each possible one

solid marten
#

so on FFV turret it works but not in nonffv when body is local?

gaunt depot
#

Are FFV turrets that much different from normal turrets?

#

It worked fine on turrets (ffv in my test) and driver

#

But failed on cargo seat

west onyx
#

requesting "taskDestinationVisual" pls

#

or at least when task is attached to an object, ability to show task icon in render scope

#

also thank you for pos1 getDirVisual pos2

neon swan
# fresh hawk Any update on this?

Fixed in rev. 149942, which should be in next week's Dev Branch. I incorrectly made the assumption, when creating the command, that a key combination could not contain a double tap; it can but only for the main key. I have updated the biki page as well to reflect it.

solid marten
gaunt depot
#

I meant engine-wise, they're still turrets

#

So my test with FFV turret shouldn't be much different from normal turret

solid marten
#

turret is locally simulated seat in vehicle. normal cargo seat is proxy so it is local to vehicle but ffv is local to unit in it otherwise it won’t respond to controls fast and smooth. not sure what you mean by engine wise

gaunt depot
#

I meant that normal turret (tank turret) isn't that much different from FFV turret (Offroad back seats)

solid marten
#

my expectation ffv should work just as normal turret for removing and deleting unit

solid marten
#

@gaunt depot Ok it works fine on what I tested so going to ask @untold sky to add it to the next prof to try again

gaunt depot
#

👌

gaunt depot
#

But in general there is no reason not to update it to use object's render scope instead of simulation

#

Nobody will ever want their task to jump around

west onyx
#

i guess most tasks are stationary

hallow sun
#

So after doing some more testing with a strip player corpse script and revive script, it looks for me that most of the issues with the gear from corpses have to do with locality.
Sometimes the game doesn't seem to change locality to the server for dead players and commands like setunitloadout (to set a empty loadout) setPosATL or even deleteVehicle stop working (serverside) and you have to remoteexec them even tho the corpse shouldn't be local to the client anymore.
maybe i am totally wrong but that's my observation 🤔

regal nimbus
#

Err, there is in general no autoswitch back to the server, other than after clients disconnect.

#

so "shouldn't be local to the client anymore" seems to be a false assumption.

#

All the commands you mention are supposed to be global argument though, so if they don't work correctly in some cases when the target isn't local then there's still a problem.

hallow sun
#

Happened even on my local test server, so it can't be a issue with the connection.

regal nimbus
#

Is there a good reason that setMass has very slow network sync?

gaunt depot
rose cloak
#

"Local Argument", though

gaunt depot
#

will probably work on remote entity

solid marten
limpid rune
#

I'm noticing a weird issue in 2.11 - but i'm not sure if it happens in previous versions. But setDamage seems to cause 'force' on physx objects

#

It's really noticeable on tanks as it causes their suspension to compress

#

yeah happens in 2.10 too

solid marten
#

2.08?

limpid rune
#

i'd have to switch branches

#

this is just running

[this] spawn {
_obj = _this # 0;
waituntil {_dmg = damage _obj; _obj setdamage (_dmg + 0.01); sleep 1; !(alive _obj)};
};
``` from the tanks init line
#

yeah happening in 2.08 too

#

probably been around for awhile just not super noticeable as you would rarely run setDamage in a loop like that

#

It's significantly worse on smaller objects (like statics) causes them to bounce around a bit, but I only tested that with modded ones. with tanks it seems to be a consistent downward 'push'

#

setHitpointDamage etc don't do it

gaunt depot
#

Probably because of damage going to tracks game wakes object up to simulate deflated tyres after damage was done?

#

Try setHitPointDamage on track specifically?

limpid rune
#

It does it to statics aswell with no tracks

#

but ill try it

gaunt depot
#

Either way tank rollers jumping like this is a visual bug

limpid rune
#

Yeah nothing with setHitpointDamage was running

[this] spawn {
_obj = _this # 0;
waituntil {_dmg = _obj getHitpointdamage "HitLTrack"; _obj setHitPointDamage ["HitLTrack",(_dmg + 0.01)]; sleep 1; !(alive _obj)};
};

and didnt move (and the hitpoint was taking damage)

gaunt depot
#

Probably physx scene isn't present for a frame so they fall through it or something

limpid rune
# gaunt depot Either way tank rollers jumping like this is a visual bug

Well its more of an issue with statics if anything as it causes them to jump and bounce, this is particularly noticeable on big statics like modded AT guns/Artillery

To me it looks like something is pushing downwards on it, and for tanks this just causes compression of the suspension which is... alright but weird, but for statics which don't have wheels nor suspension it causes them to bounce as they have nowhere to go

#

yeah testing with a vanilla 50 cal it bounces and clips through the ground partially

gaunt depot
untold sky
#

Helo, I need some Intel GPU people for help. 🙏
We have a new fix for the flickering issue, and our QA confirmed it. But before we release it in the next hotfix I would like to test some more (as the last flicker bug also didn't appear everywhere) to confirm that its actually a real fix.

For that I need some people who were affected by flickering in the original 2.10 update.
I'll send custom exe and a mod to be loaded, it'll be ready in about half an hour

sinful kettle
untold sky
#

nu unless I make a second notlikemeow

sinful kettle
#

what happens if I edit the exe and replace it myself?! does it work?!

untold sky
#

ehhhhhh dunno

sinful kettle
#

ah wait it was win 10 fucking with me 😠

#

I did replace nvOptimusEnablement and change it to 0 but it didn't work meowsweats

#

yeah I have the flicker problem too meowsweats

sinful kettle
untold sky
#

:yay:
Sadly delayed a bit longer than 30 minutes because steam is in a bad mood

#

ah no nvm apparently it works anyway

solid marten
#

i had some recent auto updates to win 10 including some major ones

lean mica
regal nimbus
#

I wondered why ACE was doing a setMass remoteExec 0 in its crate-carrying code, and it turns out that this is the only solution to killing people with the things. The global sync is so slow that you can easily pick up a box and kill someone with it before it catches up.

devout wave
gray wharf
#

done, thanks

livid sleet
#

We know we can hide objects on the server/HC and AI will basically phase through them and path like they don't exist but it'd be neat if we had the ability to apply that behavior to an object without visually hiding it allowing local MP/editor to use those kind of fixes without negatively effecting the host (or dev testing) but I'm not sure if that would be worth attempting.

livid sleet
#

I'm not sure if anyone has looked into this at all (last update was from 2017) but I've got a workaround I mentioned in the ticket for when this issue occurs and it'd be neat to see it fixed in the engine at some point
https://feedback.bistudio.com/T123262

sinful kettle
#

btw KK, would you please add AIPathOffset to getRoadInfo's returned array?
I think you missed that one when you added the command: https://feedback.bistudio.com/T147630
it's an important property for knowing which side of the road the AI must drive on.

hallow sun
livid sleet
#

I'm not sure about how it behaves in MP but by staying away from any decay value non-zero and below 1e-4 has fixed it in SP and presumably when in MP since it won't ever be on those values with the watchdog.
I'm talking about this particular issue which is 100% reproducible with the values 0 setFog [0.422057,2.46173e-036,0]; https://i.imgur.com/evmGlIn.jpeg

Imgur
devout wave
outer frigate
#

Using setHeadgear on a player with any vision mode other than normal always sets the vision mode back to normal even if the headgear has integrated vision modes (pilot helmet with nv for example) -- is this intentional and is there any reception to changing it so its persistent?

solid marten
west onyx
#

any ideas where this error comes from? its been coming up for years

#

near mission start

untold sky
#

its a config

#

of some vehicle

hallow sun
#

RHS mods i think
We have the same error

west onyx
#

john jordan just posted in scripting re a “findNearest” command, imo would be a good addition for mission makers

#

<array(pos/obj)> findNearest <reference(pos/obj)>

#

that is a common code block for sqf mission scripters

livid sleet
# west onyx

LT_01_scout_base_F -> MFD -> MFD_Gunner_Ready_To_Fire -> Draw -> condition

#

no idea what its supposed to be for such use-case

west onyx
#

trigger when AI changes attacked targets

#

There are lots of things I want AI to do based on what target it is attacking

#

for instance, tank switching shell type from armor piercing to HE when firing at infantry or a building

#

or AI unit having increased accuracy when new target is further away

#

would be huge for AI gameplay

# 
    'AttackTargetChanged',
    {
        params ['_unit','_target'];
        _unit setSkill [
            'aimingAccuracy',
            [0.13,0.4] select ((_unit distance _target) > 500)
        ];
    }
];```
solid marten
outer frigate
solid marten
outer frigate
#

Essentially:

  • use headgear with integrated NV
  • change vision mode to NV/TI/whatever
  • addheadgear with a helmet with integrated NV
  • vision returns to normal
solid marten
#

it is intentional

outer frigate
#

pain

solid marten
#

new headgear = reset. pain? you have no idea how much pain if you don’t reset

outer frigate
#

fair its just a pain in my case when im using a script which changes the players headgear if they change vision mode to have a glowing visor version and therefore doesnt work with integrated nv/ti helmets because it instantly switches back

solid marten
#

store it and then set vision mode if supported on the new item

outer frigate
#

would require me to rewrite most of the script as im using visionmodechanged eh which fires when thats reset

gray wharf
snow cairn
gray wharf
#

also, well, to be fair you use setPosATL without the moveOut :p

snow cairn
gray wharf
#

weeeeee, thermals and fog 😛

unknown.png
#

and roads hide well 😦

unknown.png
sinful kettle
gray wharf
#

set as "Not Fixable" 😄

#

(I hesitated with "Not a bug")

solid marten
#

looks like p*sstake

desert trench
gray wharf
hallow sun
#

Can we have that bug fixed when a client logs into the server and the vehicle nearby just explodes from simulation kicking in ?
Even when there is no object close to it and it's parked on flat ground/surface.

gray wharf
#

no.
I like that one, it stays.

regal nimbus
#

Did you want it to explode earlier? :P

livid sleet
#

Have you tried touching it

#

just give it a little tap

west onyx
solid marten
#

@gaunt depot deletevehiclecrew didn’t make it into prof, next dev then

untold sky
#

but as far as I can see, it should work same as setPos

gray wharf
untold sky
#

does setPos handle it correctly, or is it same broke

gray wharf
#

@snow cairn?

solid marten
#

it is move out that is local, I would imagine it does set pos but because unit is still considered inside the vehicle it goes back

gray wharf
#

if it cannot be fixed in anyway and it stays like this, I will keep the Effect Global but add a note that setPos* commands should ideally not be used to move units out of a vehicle, and if so remoteExec it

snow cairn
#

Um let me try setPos today

snow cairn
#

@gray wharf setPos is even worse haha. It moves the remote player out, then moves him back to the vehicle, but on top of the vehicle and just glitches out. stops after a couple of seconds, but the remote player is still in the vehicle

gray wharf
#

sooo @untold sky yeah, not good 😄

solid marten
#

setposX commands can be tweaked to do move out if used in units inside vehicles

gray wharf
#

even for remote units?

solid marten
#

oui

gray wharf
#

that would be wunderbar then

solid marten
#

jawoll

dreamy bane
#

would command like: object setPosX [pos, doEject]; make sense?

gray wharf
#

I wouldn't say so as the engine can detect whether or not the unit is in a vehicle, why?

real bronze
#

it should just always eject, no point in giving a choice, don't see where it could be used (imo)

dreamy bane
#

yea just a thought

#

maybe error check when ejecting would be more useful, if u don't want crew to get accidentally ejected

hushed birch
#

im using BIS_fnc_holdActionAdd and i set a timer of 20 seconds to the action, it works all fine, but after 12 o 14 seconds the icon in the middle of screen fades out, its this fixable?

devout wave
#

not at present, no

#

someone was asking about it not too long ago, so there's some possibility it might get looked at in the future, but right now you just have to live with it

desert trench
real bronze
#

the action menu does fade out, but not the progress icon (as far as i know) 🤔

devout wave
#

The progress icon does fade out

real bronze
#

not for me 😄

#

it only goes away if i stop looking at the object

#

just tried it

snow cairn
#

just look at it like how you look at your crush. that should make it disappear real fast.

gray wharf
#

as a male, how to make friends? just tell a girl you love her

hushed birch
oak mauve
#

I recently discovered and reported a bug with broken pathfinding for AI groups placed in the editor - https://feedback.bistudio.com/T167439
I wanted to ask if the developers are aware of this bug because there are no replies to my post.

regal nimbus
#

huh, that is kinda fun actually

#

It looks like the editor platoon is trying to stay in wedge formation while the zeus-created platoon switches to column, but the settings are identical as far as I can tell.

oak mauve
#

This issue breaks absolutely all missions that spawn more than one vehicle in AI groups - Warlords, ALIVE, every CTI mission, squads spawned by the DMP mod or NR6 HAL mod spawner.

sinful kettle
spare vine
#

solution: set the feedback tracker to noindex, same as all the links. This will keep those bots away (at least most of them)

sinful kettle
#

btw is it possible to return all parameters of a command? (I mean array ones too)

#

e.g. does the game store the parameters of the right hand array of addAction somewhere?

#

if so a command that returns them all would be really useful. that way even if wiki is not updated we at least know what the command needs

#

and also I can write a better SQF linter 😅

spare vine
#

I can't think of any language which supports that 🤔
Only method I know is reading out the inline documentation or reading the source code of each function/method.

sinful kettle
# spare vine I can't think of any language which supports that 🤔 Only method I know is read...

well I expect that when you register an SQF command you register the parameters too
to automate the type checking and not have to write it in the command function itself
e.g. something like functionList.Invoke("command", leftArg, rightArg) and then in Invoke method: if (!params.left.isValid(leftArg) || !params.right.isValid(rightArg)) return ...; commandFnc(leftArg, rightArg)
but that's just speculation

real bronze
#

i just got everything off wiki for mine

#

(it's actually not that bad, there are a few inconsistencies you have to handle due to wiki formatting)

sinful kettle
#

yeah those inconsistencies are exactly my problem 😅
I have decided to add a manual "patch.json" to patch commands that were parsed incorrectly but still a fully automated approach would be nicer

real bronze
desert trench
#

might be fps/scheduler plays into it

untold sky
gray wharf
#

ah wait, just got what you meant

sinful kettle
#

no I meant the types 😅

gray wharf
#

yep… nah, only the method in the engine deals with that - no "deep signature"

solid marten
#

technically this can be done for the newer commands as there is rather handy template to evaluate array params but even then it is a lot of work for very little gain

gray wharf
#

in-game accurate documentation is not exactly "little gain", but I would say it would be coming late to the party.
an internal extract to make sure we didn't miss a syntax, though, I can take it!

the possibility to have a direct wiki link however would be nice 👀 + possibility to set link prefix somewhere in e.g player profile (one can dream)
https://community.bistudio.com/wiki/ + command
https://community.bohemia.net/wiki?title= + command
ftp://myOwnStuff.fr/wiki/ + function

maiden willow
gray wharf
#

@maiden willow…?

maiden willow
#

Or whts this for

gray wharf
uncut briar
#

Is it me or is getAllPylonsInfo returning wrong "Assigned turret" when used on clients of dedicated server.

#

On server it reports correct values while on clients it seems to return default/config values. Not reflecting current state of the vehicle.

fallen elk
regal nimbus
#

@uncut briar You're aware that setPylonLoadout only has local effect, right?

#

When I saw this I initially assumed it was a wiki error, but no.

uncut briar
#

Yes I do. I'm testing on vehicle customized via eden.

#

Rest of loadout has correct values, just the owner turrets are wrong on client.

naive rapids
#

come back with the thermals the way they were, this is horrible, only server for those who want mission experience, in wasteland mode it was horrible on servers that are full thermal.

#

This should have an option for server owners to make changes to, rather than being a permanent fixed option.

stray hill
naive rapids
#

a lot of people complain but no one comes here to express their opinion, I've mentioned it a few times, this should be changed or have options for change.sorry i'm boring.

fair wing
#

It always says this and i cant seem to fix this when i download a new thing from the workshop

#

Please help

#

is this the wrong channel

gray wharf
sinful kettle
#

Try repairing the game 👇

woeful novaBOT
#

Steam file verification process:
Right Click on game -> Local Files -> Verify Integrity of game files
Arma 3
Steam Workshop Mod repair process:
Open Launcher -> Right Click on mod -> Repair
The first process will automatically verify Steam Workshop items as well.

solid marten
uncut briar
# solid marten Can I repro?
  1. Place Blackfoot in editor
  2. Change one side of pylons to the driver
  3. getAllPylonsInfo heli on client and dedicated server
  4. Observe that client sees initial/config vehicle state instead of the real one
#

In my tests it looked like only ownership state is mismatched

#

The loaded magazine, magazine id were correct on both parties.

#

@untold sky when local only objects are created?

We've tried to use Deformer mod in a mission (which AFAIK uses only setTerrainHeight via 3DEN attribute).

Mission objects placed as normal objects are working fine, objects placed as local objects are misaligned.

#

I guess they might be created too early before the terrain changes are synced?

solid marten
uncut briar
#

exactly what is seen in the screenshot

devout wave
#

Whether the weapon is controlled by the gunner or pilot

uncut briar
#

Pylon owner turret is not correct on client

untold sky
#

they are right next to normal objects, and pretty sure after, not before

#

or.. they should be

uncut briar
#

hmm, weird. Will experiment more and try to cook a repro without mods

solid marten
#

Ah wait it is driver, lemmie check the code

atomic walrus
sinful kettle
#

Hi devs,
since I won't be bothering you guys for the next ~1-2 months, I'd like to bring up a few of my tickets that are essential to my AI mod one last time, so if you guys happen to have some free time and looking for some tickets while I'm away, please take my tickets into consideration as well!

  1. getter for bones hierarchy: https://feedback.bistudio.com/T167461
  2. model animations: https://feedback.bistudio.com/T150103
    (if not possible, this will work for me too (forcing object animation): https://feedback.bistudio.com/T165117)
  3. vehicle handling (steering/gas/break): https://feedback.bistudio.com/T150072
  4. more versatile switchMove: https://feedback.bistudio.com/T164731
  5. weapon aim and head orientation setters (only for AI): https://feedback.bistudio.com/T150070

thank you and goodbye for now! salute

sleek scaffold
west onyx
#

can the “sort” function be made to accept an array of positions and sort by nearest/furthest

#

<array of pos> sort [origin,true]; or something

ionic spindle
#

retexture does not work on 4w jeep

alpine tulip
#

*MB 4WD, and only if the texture is "". Vague feedback like yours won't work but try to elaborate as far as possible

ionic spindle
#

the mb 4wd jeep cant be texutred to be invisible

devout wave
devout wave
solid marten
#

add some logging, i.e position of the plane at the moment of the drop and which completion statement of which waypoint is executed, position of the way point, distance between plane and waypoint, then we can investigate

regal nimbus
#

For some reason I didn't figure that ground vehicles will still stop dead for a few seconds after waypoint completion.

atomic walrus
#

(regarding this topic)

regal nimbus
#

If the mission was setting a high completion radius that was previously ignored, that could certainly break it.

atomic walrus
regal nimbus
#

oh, I didn't read the code. Thinking aloud.

#

Surprised it can actually do 5m with an aircraft at all :P

atomic walrus
#

Would it help to increase the completion radius to a higher value? What would be suitable?

regal nimbus
#

I dunno, this one feels like a bug and should be looked into.

#

What sort of aircraft is this typically?

atomic walrus
#

The Vanilla VTOL

regal nimbus
#

The NATO one?

atomic walrus
#

B_T_VTOL_01_vehicle_F

regal nimbus
#

Ok, I set completion radius 5 and it completed at ~1km :P

#

This is probably not working as intended.

#

Default completes at 100m or so.

atomic walrus
#

In the mission it should spawn a box in the players zone but it spawns the box really early and box come down before airplane reach the player zone.

regal nimbus
#

Yes, because it's completing the waypoint at 1km

atomic walrus
#

ah ok

regal nimbus
#

Doesn't seem to matter what you change it to. 5m, 100m and 1km all complete at 1km.

#

I tried wheeled vehicles and it worked fine, although sadly they're still stopping for a few seconds at each completion.

#

aha, if you set 1.5km then it does complete at 1.5km rather than 1km.

#

So it looks like it's capping to the default, except the blackfish has 1km rather than 100m for the cap.

atomic walrus
#

What is the reason to set a cap for that? If someone set a dedicated radius, he would live with the consequences. Why to restrict him?

#

or is the "set special radius" broken at all and overwriten by the "default value"?

regal nimbus
#

Because AI vehicles are incapable of moving accurately, so in many cases they'd never reach a completion radius lower than a type-dependent value.

#

In fact the vanilla wheeled APC can't even reach the default completion radius reliably.

#

But the caps are clearly 10x too high for air vehicles here. Tried with a Blackfoot as well, 50m completion with 0 completion radius, 500m with 5.

#

If you want to fix the mission, just remove the setWaypointCompletionRadius call, but it is bugged.

atomic walrus
#

I would prefer that it would be fixed in the performance build. Till this happen, i can stay on V2.

devout wave
#

The vehicle's default waypoint completion radius is nominally controlled by its precision config value. On main branch, that's 100 for the Blackfoot, 200 for the Blackfish. Someone on profiling should probably check that to confirm whether the problem is that vehicles have newly incorrect precision, or that the precision value is not being interpreted correctly.

regal nimbus
#

Either way it's using 10x the value when you use setWaypointCompletionRadius.

atomic walrus
regal nimbus
#

Editor repro is trivial.

#

Place air vehicle, give it waypoint, set completion radius to something non-zero.

atomic walrus
oak mauve
#

I'm sorry for using the wrong channel, I don't have time to report it properly and I'll be away from my computer for a few weeks.

I found that all tracked vehicles never try to unstuck themselves, they will endlessly ram into obstacles that block their path - trees, poles, sandbags, rocks or the walls of a house
I can't pinpoint the exact location to reproduce this, but it happens on all maps randomly with all tracked vehicles.

#

While the wheeled vehicle will move backward and try to get around the obstacle, the tracked vehicle will do nothing

20220925022912_1.jpg 20220925023038_1.jpg
west onyx
#

thats arma ai driving for you

regal nimbus
#

The wheeled vehicle unstick logic is pretty busted too, but it does try. Sorta.

#

There's one mode where it headbutts the object repeatedly, I think randomly trying different angles, none of which work.

#

And another where it turns extremely slowly for several minutes, despite having plenty of clear space in front of it.

solid marten
solid marten
solid marten
limpid rune
#

I do wonder if it's done for the sake of things like fences/trees so they know a direction to fall down...

devout wave
atomic walrus
solid marten
#

If precision is greater than given radius then the greater one will be used. Not a bug but intended. Are we done now?

devout wave
#

It seems like the precision is being treated as bigger than it actually is, e.g. the Blackfish has a config precision of 200m but is behaving as if it has a precision of 1km

solid marten
#

repro please

devout wave
#

This is distilled from what the others are reporting, I'm not on profiling myself:

  • Place a Blackfish with a waypoint, radius more than 0 but under 1000
  • Place a reference object in the same place as the waypoint
  • Waypoint On Activation: systemChat (this distance ref_obj)
  • Run the mission and see what distance is reported on wp completion, compared to the radius set
  • Repeat with a radius over 1000
  • Result should be that the Blackfish completes any waypoint with radius < 1000 at or around 1000, as if it has precision 1000, with any radius over 1000 respected more accurately
azure crane
#

uhh can we have this box wider or list points shorter so people can select proper task type on first try? There's no hover tooltip so it feels a bit off (I'm mostly concerned about cDLC bullets)

unknown.png
desert trench
#

given the fog banding issue seems to be investigated, is there also a chance to look into the black fog during night time?

livid sleet
#

I'm not aware of that issue, is there an easy repro?

desert trench
#

might be the Contact DLC (Livonia) brighter night config actually

#

like fog color may need adjustments alongside

devout wave
alpine tulip
#

Just a thought: can we slap Dedmen so he'll make a working visible laser pointer like non-IR variant of acc_pointer_IR?

daring wagon
#

Just a thought: can we slap Dedmen [end of quote]
yes

gray wharf
#

WTF
Dedmen only takes tickets and hugs
But don't slap him please
you can eventually whip him

deep plaza
#

does it make sense to create a ticket so that the vanilla debug console is enlarged a bit and gets a history like cba?
basically a similar debug console to cba but without using a mod for it. 🫣

alpine tulip
#

There's already histories that can store up to 9 or 10

devout wave
#

The vanilla debug console has a history, there's just no GUI buttons. Use PageUp and PageDown to navigate it

deep plaza
#

yes, but the window is very small, so would it make sense to create a ticket just for that?

alpine tulip
#

I don't really think so, since if I need to write a long code there, I'll just use external text editor to write and paste

devout wave
#

You can use Leopard20's Advanced Developer Tools which significantly improves the debug console. It is a mod but unlike CBA it won't create a dependency or screw up anything else.

devout wave
deep widget
devout wave
gray wharf
#

not gonna lie, I liked the disclaimer [not spam] :D

vocal thistle
gray wharf
#

I admit it would be nice

vocal thistle
#

It would make our lives sooooo much easier - speaking on behalf of the Apache mod, the Hatchet's H-60 pack and our F-16

#

and surely some of the others too

#

(+higher quality mods)

indigo kiln
#

Yes it would be very very very nice to have, and I think my suggestion is fairly reasonable

#

I'm also in prague next week so just tell me who I need to bribe with booze

gray wharf
#

I accept the booze, but I don't promise any result other than me downing it 😁

vocal thistle
#

Damn, I would've joined in with some extra Polish booze as well if my car hadn't broken down

devout wave
gray wharf
real bronze
#

what is this with this spam stuff meowsweats

azure crane
#

Bots love phabricator and it haven't received any security update for a long time now

gray wharf
#

phabricator itself is abandoned, a free alternative has been building on top of it but we haven't made the transition to it (yet?)

devout wave
alpine tulip
#

Actually, is there any reason why we can't retexture weapon accessories but only magazines? It is really unreasonable I think, is it because an accessory is a proxy of a proxy? (I probably missed someone's explanation, but hey)

alpine tulip
tacit kettle
#

Hi guys i have a question is there any news about AAR for arma 3. It would be cool if we could have ingame or at least on map recording of players and ai. Other wise we need to use Third party applications to achive this.
I searched the Feedback tracker and the last mention was in 2016.Reference: https://feedback.bistudio.com/T66212

devout wave
#

It would be a pretty major undertaking at a time when the A3 dev team is rather light on resources - and those third-party applications already exist. I mean, never say never, but I'd rate that as "unlikely" at best

gray wharf
#

super unlikely confirmed indeed

tacit kettle
gray wharf
devout wave
gray wharf
hallow sun
#

When using the localOnly option in the editor for objects i get a bunch of these in the server rpt on server startup.
Does this have any negative effect ?

4:19:36 Ref to nonnetwork object 27d06d30100# 1239648: netfence_02_m_8m_f.p3d
4:19:36 Ref to nonnetwork object 27d06ce0b80# 1239654: netfence_02_m_8m_f.p3d
4:19:36 Ref to nonnetwork object 27d06c16080# 1239656: netfence_01_m_4m_f.p3d
4:19:36 Ref to nonnetwork object 27d06b18b80# 1239702: billboard_f.p3d
4:19:36 Ref to nonnetwork object 27d06acf580# 1239704: signsad_sponsor_f.p3d
4:19:36 Ref to nonnetwork object 27d06aceb00# 1239705: signsad_sponsor_f.p3d
4:19:36 Ref to nonnetwork object 27d06ace080# 1239706: signsad_sponsor_f.p3d
4:19:36 Ref to nonnetwork object 27d06acd600# 1239707: signsad_sponsor_f.p3d
4:19:36 Ref to nonnetwork object 27d06accb80# 1239708: signsad_sponsor_f.p3d
.....
outer frigate
#

no thats what local only does

#

makes it local to each client

hallow sun
#

When using hold actions with longer duration (+10 seconds) the icon fades out while progressing meowhuh

alpine tulip
#

I actually think if there is a command to force-show or force-hide Action Menu

hallow sun
alpine tulip
#

Why you think it doesn't look like so?

hallow sun
alpine tulip
#

I'm confused, why you put a 🤔 on my idea if kju has the same problem?

hallow sun
#

found his post afterwards 😅

alpine tulip
#

I'm confused still...

desert trench
#

well its needs a fix on the engine side

#

or expose configuration via config or sqf

real bronze
#

are we talking about this icon?

uncut briar
#

attach it to some other object

#

not yourself.

#

In SOG we had this issue with our medical actions so we created our own holdActions which use cutRsc instead of the scroll action icon.

real bronze
#

ah i see the issue, the action can appear normally without you interacting with the action menu (showWindow and high enough priority), then it doesn't fade out

#

but if you start the action via the action menu, then it fades out

lean mica
#

Can someone tell me is it a bug, or I missed something? Why kick does not work in onUserConnected in server cfg?

#

onUserConnected = "ban (_this select 0)"; - works
onUserConnected = "kick (_this select 0)"; - does not

devout wave
devout wave
hallow sun
#

Previous messages about the problem with more details:
#arma3_feedback_tracker message
#arma3_feedback_tracker message

I did some more digging regarding the wonky setUnitLoadout (configFile >> "EmptyLoadout") on a player corpse to avoid some MP problems (duped gear etc).
Tested on my local server and our live server .

Maybe noteworthy: We move dead players to a graveyard group created by the server and re-spawning/revive creates a new player object.
So when i execute this code below (server-side) for my old player object after i died and re-spawned the setUnitLoadout seems to work like intended and cleans the corpse reliably.

_obj = objectFromNetId "2:6061"; 
(group _obj) setGroupOwner 2;

After running this code once, every setUnitLoadout execution to clear corpses seems to work without any problem.

devout wave
azure crane
azure crane
#

oh and I made this: https://feedback.bistudio.com/T168176, I was working on a simple Western Sahara mission (no mods, just the CDLC), and as I wrote there, any dynamically simulated units will freeze as long as player is in a vehicle with AI driver or AI passenger (with no driver), so Dedmen fix pls

azure crane
gray wharf
#

D E N I E D

azure crane
stone tree
stone tree
gray wharf
daring wagon
azure crane
# daring wagon You found the cause ?

not the cause, but part of the issue is that dynamically simulated group has it set as false even when I try to make it true, so I think this is just a symptom

#

loading additional content through startup parameter doesn't do anything, setting dynsim through scripts during scenario also makes dynsim squads break. Only thing I haven't tried is diag exe but I assume it also won't change anything

paper lichen
daring wagon
desert trench
devout wave
dreamy bane
real bronze
#

rpt is good enough imo

#

you wouldn't get much more diag info anyway

#

for stacktraces you can just launch the game with -debug

dreamy bane
#

it would be easier to view the call stacks and variables of each error

west onyx
#

is there a faster way to get result of allmissionobjects 'weaponholder'

#

like 28 allobjects 4 or something

west onyx
#

so far i am getting this

# 
0.137495 ms

Cycles:
7273/10000

Code:
(12 allObjects 0) select {_x isKindOf 'GroundWeaponHolder'};

Result:
0.214087 ms

Cycles:
4671/10000

Code:
allMissionObjects 'WeaponHolder';```
#

so (12 allObjects 0) select {_x isKindOf 'GroundWeaponHolder'}; is significantly faster

#

tested with 100 "groundweaponholder_scripted" and 100 buildings to add noise to the (12 allObjects 0) array

#

if I pump the buildings to 500 to add more noise, it gets predictably slower (0.35)

real bronze
#

and yeah, you sadly have to filter

#

which eats runtime big time

hallow sun
#

Sometimes when leaving a static machine-gun the turret stays local to the client even tho it should not do that (The static gun was created by the server) 🤔
Repro: Setup a static gun on a MP server and check if its local via after getting in and out repeatedly.

(objectFromNetId "2:3321") turretLocal [0];
gray wharf
#

I think that is how it works for vehicles

hallow sun
#

"When gunner leaves turret it is supposed to change locality to the locality of the vehicle."

#

It works most of the time but sometime it fails, even getting in and out again doesn't revert it

gray wharf
#

it is a main turret, so idk

#

and if it is inconsistent, it stinks indeed

limpid rune
#

it got changed so that the vehicle becomes local to the effective commander aswell now

#

like regular vehicles its local to the driver with highest priority but because static guns dont have a driver, the effective commander (gunner in this case) it becomes local to

devout wave
gray wharf
gray wharf
deep plaza
#

Is it a bug that you can't attach a #lightpoint to a bullet (ammo) created with createVehicle in MP ? but somehow attaching the #lightpoint to a throwable ammo works.
The lightpoint is visible with a throwable ammo but not with a bullet ammo

rose cloak
#

i'm not sure bullets are synced over the net at all

deep plaza
#

But they do damage and the impact is visible

rose cloak
#

bullet is created on every PC with the same staring params, so theoretically it should behave identically on every PC.

limpid rune
#

yeah it's that - would be better to just createVehicleLocal the lightpoint and create it locally for each clients specific 'projectile'

deep plaza
limpid rune
#

no as said that's whats already happening

rose cloak
#

run EH on every client so every client creates local lightpoint on its local projectile, (and not only one EH on the source that tries to stream whatever it's doing to others with remoteExec)

devout wave
#

It sounds like they are needing to create artificial bullets and there are no existing ones to use

devout wave
gray wharf
#

niouked

stone tree
devout wave
azure crane
deep widget
desert trench
#

you can quite easily identify spam posts via such new posts query. seems more useful someone with respective permissions to get through that once a day than this individual posting here imho

#

could easily automate the identification with some regex too with high matching certainty

gray wharf
desert trench
alpine tulip
azure crane
last jolt
atomic walrus
west onyx
#

yikes i didnt notice this typo before 😄

#

i guess too late to change

#

CargoPlaftorm_01_green_F

gray wharf
#

unfortunately

daring wagon
#

Fixing typos now will cause more dammage than do good

gray wharf
#

I see what you did here

daring wagon
#

😄

west onyx
#

yep, typo fixing canceled

#

regarding actual bugs, not sure if ever been reported, but with tacticalPing=3 the ping is supposed to appear on the map

#

hint: it doesnt

#

tested in dedi MP

desert trench
#

@hasty acorn ^

hasty acorn
# west onyx tested in dedi MP 
class DifficultyPresets
{
    class CustomDifficulty
    {
        class Options
        {
            // Simulation
            reducedDamage = 0;        // Reduced damage        (0 = disabled, 1 = enabled)

            // Situational awareness
            groupIndicators = 1;    // Group indicators        (0 = never, 1 = limited distance, 2 = always)
            friendlyTags = 1;        // Friendly name tags    (0 = never, 1 = limited distance, 2 = always)
            enemyTags = 0;            // Enemy name tags        (0 = never, 1 = limited distance, 2 = always)
            detectedMines = 1;        // Detected mines        (0 = never, 1 = limited distance, 2 = always)
            commands = 1;            // Commands                (0 = never, 1 = fade out, 2 = always)
            waypoints = 2;            // Waypoints            (0 = never, 1 = fade out, 2 = always)
            tacticalPing = 3;        // Tactical Ping        (0 = disabled, 1 = enabled)


            // Personal awareness
            weaponInfo = 2;            // Weapon info            (0 = never, 1 = fade out, 2 = always)
            stanceIndicator = 2;    // Stance indicator        (0 = never, 1 = fade out, 2 = always)
            staminaBar = 1;            // Stamina bar            (0 = disabled, 1 = enabled)
            weaponCrosshair = 1;    // Weapon crosshair        (0 = disabled, 1 = enabled)
            visionAid = 0;            // Vision aid            (0 = disabled, 1 = enabled)

            // View
            thirdPersonView = 1;    // 3rd person view        (0 = disabled, 1 = enabled, 2 = enabled for vehicles only (Since  Arma 3 v1.99))
            cameraShake = 1;        // Camera shake            (0 = disabled, 1 = enabled)

            // Multiplayer
            scoreTable = 1;            // Score table            (0 = disabled, 1 = enabled)
            deathMessages = 1;        // Killed by            (0 = disabled, 1 = enabled)
            vonID = 1;                // VON ID                (0 = disabled, 1 = enabled)

            // Misc
            // mapContent = 1;        // Extended map content    (0 = disabled, 1 = enabled)    // before Arma 3 v1.68
            mapContentFriendly = 1;    // Map friendlies        (0 = disabled, 1 = enabled) // since  Arma 3 v1.68
            mapContentEnemy = 1;    // Map Enemies            (0 = disabled, 1 = enabled) // since  Arma 3 v1.68
            mapContentMines = 1;    // Map Mines            (0 = disabled, 1 = enabled) // since  Arma 3 v1.68
            mapContentPing = 1;
            autoReport = 1;            // Automatic reporting    (0 = disabled, 1 = enabled)
            multipleSaves = 1;        // Multiple saves        (0 = disabled, 1 = enabled)
        };
        
        // aiLevelPreset defines AI skill level and is counted from 0 and can have following values: 0 (Low), 1 (Normal), 2 (High), 3 (Custom).
        // when 3 (Custom) is chosen, values of skill and precision are taken from the class CustomAILevel.
        aiLevelPreset = 1;
    };
    
    //this isnt used because aiLevelPreset isn't set to 3. 
    //class CustomAILevel
    //{
    //    skillAI = 0.5;
    //    precisionAI = 0.5;
    //};
};

So, the base game "regular" difficulty, where tactical ping does show on map, has tacticalPing set to 3, and does not define mapContentPing
however, the base game also defines the default Custom Difficulty, and in that, it sets mapContentPing = 0;
Above is my custom difficulty config, I have this running on 14 arma servers and they are all actually using the custom difficulty and tactical ping works.
So TL;DR, based on your testing, it would appear you need BOTH mapContentPing set to 1, and tacticalPing set to 3, for it to appear on map

#

this is guesswork though, would love for someone to go look at the source and figure out what these undocumented config options really do

vocal abyss
fresh hawk
rose cloak
#

link points to something that looks like a browser game

trim acorn
#

Its a browser game, likely ad bot

hallow sun
gray wharf
#

but it is noice

regal nimbus
#

What are you doing with it where that's an issue?

hallow sun
gray wharf
#

use Fire EH on others and check distance

hallow sun
#

Would be nice if someone could look into that, it's an old bug and spams the rpt like crazy.

21:37:53 Server: Object info 2:6527 not found.
21:37:53 Can't change owner from 0 to 2
21:37:53 Server: Object info 2:6528 not found.
21:37:53 Can't change owner from 0 to 2
21:38:07 Server: Object info 2:30192 not found.
unknown.png
desert trench
#

Duplicate magazine LIB_45x_XXX_HE detected (id D:DDD) in slots player and close
Duplicate magazine LIB_45x_XXX_HE detected (id D:DDD) in slots player and short
Duplicate magazine LIB_45x_XXX_HE detected (id D:DDD) in slots player and medium
Duplicate magazine LIB_45x_XXX_HE detected (id D:DDD) in slots player and far

#

we are seeing this spammed for both client and server since recently (from tanks and AT guns with split muzzle setup to allow AI to switch between main gun ammo)

#

was there some change in logging for 2.10 to make this appear?

#

or could this come from magazineWell definitions?

limpid rune
#

seems to happen when AI is gunner and player is commander and orders the AI to load a different magazine, strangely though we have AI only muzzles for all ammo types but AP and it only complains about the HE ones

hallow sun
#

800.000 times "Can't change owner from 0 to 2"
70 mb rpt meowsweats

unknown.png
gray wharf
#

you know you don't have to select all the text to then "search in selection" right

desert trench
#

isnt this some faulty script code, or what's the deal here?

hallow sun
#

i will try to narrow it down with the help of #logEntities like i did with the stray Entities bug from CanAdd

#

maybe i can find some more parameters/details blobdoggoshruggoogly

last jolt
deep widget
dreamy bane
#

I think the "LandedTouchDown" may be bit bugged because when landing at carrier it gives the _airportID as number parameter while it should be the carrier object, like "Landing" EH gives as parameter

#

the params passed to LandedTouchDown are: [plane, airfield_index, <NULL-object>]

#

testing in Altis the LandedTouchDown gives 5 as the _airportID but since altis has 6 airfields it's index to one of them and not for carriers

#

I doubt carriers support any kind of index

#

anyway, "Landing" EH does it right

rose cloak
#

LandedTouchDown just seems to get the ID of the nearest airfield as its parameter, judging my me randomly ramming the ground on Altis

dreamy bane
#

but it's not -1 when landing on carrier

rose cloak
#

it's not -1 when ramming ground at any random spot as well

dreamy bane
#

right

#

I first thought LandedTouchDown was for autopilot landing only but it still doesn't work when landing on carrier (the parameter issue)

last jolt
outer frigate
#

its not changing temperature, its the thermal adjusting to the ambient temperature around

last jolt
#

So this is in fact the effect intended to be displayed by the developers?

azure crane
#

yes, it's called aperture

last jolt
#

Okay thanks. I was hoping it was a bug. Is there an option to retain previous thermal system?

outer frigate
#

no

last jolt
#

Nice

#

Anyways thanks for the info. 👍

hallow sun
#

About the rpt spam from

16:59:16 Server: Object info 2:5290 not found.
16:59:16 Can't change owner from 0 to 2
16:59:16 Server: Object info 2:5291 not found.
16:59:16 Can't change owner from 0 to 2
16:59:16 Server: Object info 2:5292 not found.
16:59:16 Can't change owner from 0 to 2

If you count the netID's it appears that it's the same error count for each SupplyXX container from inside the "mother container" -> AeoG_WoodBoxLightBig_Static.
(2:5291 & 2:5292 error both 155 times).

#LogEntities executed after server is done with loading everything (Base objects, containers etc) and is ready to accept players.
https://cdn.discordapp.com/attachments/646531827397296138/1037439073209561109/unknown.png

.Rpt at the end of the mission runtime.
https://cdn.discordapp.com/attachments/646531827397296138/1037439498566516756/unknown.png

The same crate has multiple SupplyXX (backpack/uniform/vest) appearing with Object not found and can't change owner 🤔.
You see the netID's from the #LogEntities log in the first screenshot and the subsequent netID's of the SupplyXX containers erroring in the server .rpt.

Conclusion: Something does setOwner on vests/uniform/backpack inside other containers (in this case base containers).

lean mica
west onyx
#

who do we have to bribe for an "attacktargetchanged" event handler

desert trench
#

this can be closed from what i undestand

#

can be closed

#

wasnt this solved?

desert trench
#

can be closed

real bronze
outer frigate
#

Is there any reason using grenades underwater was removed? Would like to suggest re-implementation as would make diving easier with chemlights & ir grenades etc

west onyx
#

It would be great to be able to put a player into incapacitated state before death, without using a handledamage event handler

#

Right now we cant apply incapacitation mode to all units in a mission, because we'd have to add a handledamage event handler to all those units , which is outside performance constraints

#

so something like a hardcoded way to divert units into incapacitation state instead of death

#

and a "lifestate" mission event handler to manage this

azure crane
#

just found a GAMEBREAKING BUG, male01engfre when in Stealth says "[numbers] team Blue" while units are being assigned to team Green

#

where can I apply for a refund after ~7k hours?

#

hmm, actually all are mismatched, dude calls out Yellow when assigning to Blue mega_thonk

regal nimbus
#

Try the stable build, might be a new bug.

hallow sun
#

When closing client inventory the server sometimes logs this below, i disabled our inventory saving to exclude problems in our scripts (Client & server on latest prof build).
6:29:31 Error: Object(4 : 12) not found
The netID should increase subsequently.

Can someone replicate that with a "vanilla server" ?
Just repeatedly open and close your inventory and this error should appear in the server .rpt.

dreamy bane
#

have you guys thought about new commands for the lazy eval conditions, something like &&! or &&? so you could do if( cond1 &&? cond2 ) then {}; - just a thought

real bronze
dreamy bane
real bronze
dreamy bane
real bronze
#

yes?

#

a && b:
evaluate a
evaluate b
call &&

#

same story is for every other command

dreamy bane
#

ok

#

well everything is possible if wanted but I see this is a bit problematic

untold sky
untold sky
untold sky
untold sky
untold sky
untold sky
untold sky
untold sky
untold sky
untold sky
untold sky
untold sky
untold sky
untold sky
dreamy bane
untold sky
#

Without the {} code block, that would change the basic fundamentals of SQF

dreamy bane
#

alright

alpine tulip
untold sky
#

Should be easy tho

vocal thistle
untold sky
#

It will come to next dev branch update. Please check it out then and see if it works well enough for you

untold sky
outer frigate
untold sky
#

nu

azure crane
#

Dedmen, have you found time to look into this problem of dynamic simulation I brought up some time ago? Not asking for fixing but if its even patchable, it would help me decide whether I should wait with publishing my stuff that's already utilizing dynsim or should I start redoing them

untold sky
#

I found time to look onto the ticket and make a internal ticket for it

#

Probably tomorrow

#

Also I'm confusing your avatar with R3vo

untold sky
untold sky
#

So
group player targets [false, 100, [WEST]]
->
[B Alpha 1-1:4,B Alpha 1-1:3,B Alpha 1-1:2,B Alpha 1-1:1 (dedmen)]
(Only one group on map)

desert trench
#

isnt the own group's members obsolete to list as they will be known all the time - or is there any situation/case where this is not happening?

#

like you have the extra filters from targets command, but filtering your own group out again (engine level via extra true/false parameter, or via sqf is probably less efficient)

#

in short no wouldnt list them

#

is anyone else seeing a practical reason to do list the own group's members?

untold sky
desert trench
#

11:41:57 No alive in 20203 ms, exceeded timeout of 10000 ms

#

Are you still interested in dumps from these?

#

seems to happen from group spawning by combat patrol, but hard to pinpoint exactly

dreamy bane
#

let me know if I forgot something, the debug info is in both systemchat and .rpt file

#

I also noticed that while the plane is on carrier deck it moves a little bit from time to time and causes the LandedTouchDown to trigger again

west onyx
untold sky
#

ugh

west onyx
#

too difficult / time consuming?

untold sky
dreamy bane
untold sky
#

Ah yeah found it.
Someone forgot to do a copy-paste, which is why it finds the closest static airport and ignores all dynamic airports.
The correct position was calculated, but it was forgotten to actually return that...
It thinks the carrier is at 0,0,0

dreamy bane
#

I mean if LandedTouchDown doesnt work for carriers then when landing on carrier the _airportID shouldn be -1

untold sky
#

it is intended to work

#

and it would without that typo