#arma3_feedback_tracker
1 messages · Page 1 of 1 (latest)
Normal. More like intentional
=
for what?
In Eden, it supposed to show the default state of vehicles. In Simple Object state, it supposed to show the “idle state” of the live vehicle, not default
hmm, i see
For instance, the suspension of a wheeled vehicle, is pretty high at its default state (in Eden), which is not a normal/natural way to show an alive vehicle that DOES exist there
This is why it sinks it when you make it into a Simple Object
@solid marten Was this comment https://feedback.bistudio.com/T166399#2342954 meant to be for this ticket? https://feedback.bistudio.com/T163762
yes, thank you
Is there a known case where the source (second parameter) of the Hit EH occasionally returns bogus values?
possibly related to vehicle explosion damage.
what do you mean by bogus?
A player who wasn't related to the damage.
So we've discovered that that aircraft passive radar (RWR, anti-radiation sensor) was detecting radars at 2x the distance at which the radar could see the aircraft, instead of the twice the radar's actual range.
i.e, if radar's range = 8000m , and the radarTargetSize of the aircraft is 0.1, then the RWR sees the radar at 2 x 8000 x 0.1 = 1600m instead of 2 x 8000 = 16000m as the documentation suggests.
This leads to problematic situations where PassiveRadars with identical configs behave very differently depending on which aircraft is being used. It also makes passive radar/ rwr useless in some cases and less effective in all cases.
Ticket: https://feedback.bistudio.com/T166838
like the driver of the vehicle?
what 4th param returns?
4th param is either not an object or not a player. Not sure which because this code wasn't set up to check for it. 2nd parameter in the known case was standing about 30m away, although IIRC we had another case where the player was a few kilometers away.
too vague, cant operate on this, present a solid evidence then we can talk 👍
Yeah, was checking whether it's worth trying to replicate, because I suspect it's gonna be a pain.
Worst I found so far is that vehicle explosion damage has the source as the effective commander of the vehicle, which is odd but doesn't explain the issue.
Having this exact issue https://feedback.bistudio.com/T156660
I suspect it happens on official servers, it might only happen if the server has been up for a while or if there's higher than normal load on the server
I've had tons of crashes from this
1 of 5
if your client crashes, please attach the crash reports on the ticket
@ Dedmen Managed to reproduce it, I've attached the crash report as well as steps to reproduce.
https://feedback.bistudio.com/T156660
hey all, how do I submit a bug? Noticed that the stealth balaclava does not work as expected
It stops rendering after certain distance
looks like someone already reported this: https://feedback.bistudio.com/T142683
addAction condition's _this parameter acts weirdly (alternates between _this and selected subordinate)
https://feedback.bistudio.com/T166881
bug or feature? seems (way) more bug to me 🤔
hmm. Could do with the millisecond timings in the log there, to see whether it's triggering in different frames or not.
u break it u buy it
oh no
btw automatic grouping doesn't work in devBranch profiling Eden (tried disable-save-enable-save to no avail) is it known?
So with https://feedback.bistudio.com/T166881, does that mean that every addAction is checked against every local unit every frame? Or is it strictly units in the same squad as the player?
Also what on earth is this for? :P
Oh, selection only. I guess that's fine.
anyone else getting error when using this syntax of this command
private _info = _group getEventHandlerInfo ["EnemyDetected", 1]; // group EH
you are not running dev?
It's not released yet
just the "group" syntax?
oh ok, thx
new dev branch should be this week tho
oh btw, workflow-wise can I fix A3 functions? 😄
WDYM?
as in can I commit SQF? I guess Joris would have to confirm & validate all this, but if I can bring some help on that
Not now, commits are blocked during release phase
Technically you can. But needs approval before commit
yes I meant more in general
> 2.10 + approval then I can I presume
Whatever happened to the community contribution idea or whatever it was called?
still in the process
So is it actually gonna happen at some point?
----> @jaunty narwhal
are new changes still going into 2.10?
re SQF function library refactor .. could work on the submissions now so they are ready when the contribution thing launches
we don't want to refactor, only fix bugs
@solid marten how is that "fileExists" thing going? 😄
Definitely nothing major
@untold sky with the functions library there are lots of optimisations that can be done ... i gather its mainly about fixes instead of optimisations?
No data will be done until 2.10 release is done and safe (not broken)
If it hasn't already happened it will take a bit longer
I could
what Leo meant earlier was "community-provided functions library" which is a different thing altogether
vanilla library fixes might be suggested as well through this medium, but nothing is said/planned yet
(well, we BI could obv)
would have universal benefit to all players
we would just have to agree, no Credits, just "community contribution"
hence why it's on the table
in discussion as well
Eh?
added ages ago?
oh, it didnt work tho, at least the initial implementation ... since "fileExists" always returns true for initplayerserver.sqf (there are those default .sqf files in config)
still got this on server join
21:17:02 User Quiksilver (76561198000000754) tried to remoteExec a disabled function```
optimizations are usually also ok.
But touching a API thats used by thousands is always dangerous, even removing a local variable could break some scripts
agreed ... altho my stuff has been broken many times without complaint 😄
latest gamebreaking one was removing the "get in" action for simulation disabled vehicles ... I used to use that action event to enable simulation ... result was lots of screaming that no one could access vehicles, and explaing to server admins how to fix
it was a good change tho so we just roll with it
_vehicle addeventhandler [
'getin',
{
(_this # 0) enablesimulationglobal true;
(_this # 0) removeeventhandler [_thisevent,_thiseventhandler];
}
];```
cannot see any follow up here: https://forums.bohemia.net/forums/topic/143930-general-discussion-dev-branch/?do=findComment&comment=3458823 I havent looked at how command works and why it finds nonexistent file
Its a amaing beret even with the wrong RVMAT 😂
i think its those default sqf files in the "scripts" folder
it finds "functions\scripts\initplayerserver.sqf" (functions_f.pbo)
there were some tweaks after complains have been made afair
yea all good no worries. you guys are doing great work, i would have moved on awhile ago if not for the continued development
There have been long time of feedback from the time we dealt with those issues. Why all these surprise revelations as 2.10 is about to go live?
they are not big issues i guess ... call it laziness on my part
It is just me or some Arma 3 devs are defecting to DayZ? Hint: The dev in question is not "alive" -wink wink-
to a limited extent they do as they're told
in fact bohemians are kind of like arma AI ... give them a waypoint and it depends on many factors when/if they get there
Great analogy! I guess BI gives them lots of "Get In" orders, because in game when I order my Ai to get inside a vehicle they literally circle around the whole map before even coming close to the vehicle.
We are not tied to just working on A3, if the expertise of a dev is needed on another project they will help out there when they have the time to do so.
Yeah but I bought all of Arma 3's DLCs and cDLCs just so that dedmen works exclusively on Arma 3 and nothing else for the rest of his life until he is too old to type on the keyboard with his hands!
Just kidding! Is pretty cool that you guys are so supportive with each other even between different projects! See, BI does some things right, allowing you guys to choose the place of your pain and suffering on your own!
A friendly reminder thou: Both putin and Xi Jinping consider defection as punishable by death! So don't do that after the new world order is established 😅
Actually... I didn't buy Iron Curtain... and maybe that is why dedmen left 😭
The next time dedmen comes here begging for feedback tracker requests or bug reports I won't give him the pleasure of getting those from me! I hope more of my fellow arma patriots follow me in this protest against defecting, betraying zombie loving devs...
I don't think me helping out for a day or two every couple months counts as defecting.. and honestly your comments feel like an insult, especially your comment on DayZ discord.
Blaming and semi-threatening me for doing my job, you really got nothing better to do instead?
But sure you do keep doing you, I won't miss you, bye and please stay away.
Unable to disable damage to trees 😦
https://feedback.bistudio.com/T166941
.rar? are the one customer that bought WinRAR?
hmm fixed =\
how?
cursorObject allowDamage false;
isDamageAllowed cursorObject; // returns true
exactly
so… not fixed?
I meant 'fixed' .rar to .zip

ah 🤣 I use 7zip so it wasn't a problem anyway
but I can confirm, trees seem unable to be set invincible even through scripting 🙂
The whole thing was just a joke 
Sorry if that wasn't obvious enough... I had been "funny" that whole day on several channels just to leave traces of the fun nature of those posts
I even said that it was cool that you are able to provide support everywhere you are needed
You took me seriously for some reason when most of my posts are semi memes by now...
Might as well add this to biki. allowdamage is only supported for entities and their derivatives. entities are derivatives of objects, so objects are not supported. simpleobjects are derivatives of entities so are supported but have damage handling disabled by design
won't fix
ok
Just for info: I wrote a config for vegetation objects and disabled their damage through the "Enable Damage" attribute, and its works, they are invincible
yeah if it is entity it will work
entities are derivatives of objects
what are objects?
terrain objs? cameras? projectiles?
Object : Entity : SimpleObject, got it! 👍
I mean what objects in the game are class Objects?
these?
spam comment: https://feedback.bistudio.com/T166790#2345544
https://feedback.bistudio.com/T164570
Are we able to get this one bumped? Its still an issue effecting my mod that I made, appreciate everyone is busy but feels like a small issue (hopefully)
Should be able to workaround by forcing setUnitRank in onPlayerRespawn?
Spam comment: https://feedback.bistudio.com/T166882#2346412
suggestion, since pushBack has its pushBackUnique - what about append → appendUnique? 🙃
we have insert now
oh, true! I decide to ignore every new thing coming post 2018 😄
thanks, please dismiss
Added a link 😉
btw is it a bug that when you look close at muzzle flashes the main thread goes crazy and the FPS drops dramatically?
something has to die
either you looking down the barrel, or your FPS 😄
joke aside, maybe a lighting thing?
to repro, run this:
createSimpleObject ["a3\armor_f_beta\apc_wheeled_01\apc_wheeled_01_cannon_f.p3d", player modelToWorldWorld [0,3,5]]
then go to spectator and go all the way close to the muzzle flash (the muzzle flash animation is constantly running for that object)
oh wait I had multiple objects spawned into each other 
👀👀👀 😄
that is the case for many vehicles
for instance the AAA static ship gun
I think also static MG turrets
I know. That was for quick repro.
Anyway it was my own mistake
I just noticed that diag_scope and diag_stackTrace only work within a file
can they be extended to work across multiple files?
I was trying to make an in-game debugger. but it can't be done properly without those commands
ah wait isNil resets the scope to 0 and clears the stack trace... 
it's already a command 
it's used to check if the result of the code is nil, and to run code unscheduled
it has nothing to do with "fast"
maybe I mix it with something else then but isn't unscheduled faster though?
it finishes faster but it's not faster (at least not enough to notice)
so faster in that sense 😄
I was just thinking why people use it to run in unscheduled
It is used to define thread-safe atomic operations (just like the synchronized keyword in Java), not to improve execution speed.
oh didn't know that
maybe a new command for that would be in order and documentation
makes the code more readable
"nearestSimpleObjects" wen ???
save us from (allSimpleObjects []) select {((_x distance2D _position) <= _radius)}
do inAreaArray
hmm?
allSimpleObjects [] inAreaArray [_position, _radius, _radius, 0, false]
think is faster?
ill test it out
depending on how many objects you have this could end up being 10x faster

nice
0.396983 ms
Cycles:
2519/10000
Code:
(allSimpleObjects []) select {((_x distance2D player) <= 500)};```
0.0967 ms
Cycles:
10000/10000
Code:
(allSimpleObjects []) inAreaArray [position player, 500, 500, 0, false];```
400 simple objects in VR
it will be even faster if you don't use position 
i think it requires position
I think in this case since you have so many objects most of the performance is being spent on collecting those objects
try
diag_codePerformance [{_this inAreaArray [getPosWorld player, 500, 500, 0, false]}, allSimpleObjects [], 10000];
vs
diag_codePerformance [{_this select {((_x distance2D player) <= 500)}}, allSimpleObjects [], 10000];
for 100 objects the first one is ~13x faster for me
nice
we tend to have several hundred simple objects on terrain at any one time, but good to know its much much faster 😄
going through seeing where i can use inAreaArray now. been using for ages but not for simple object detection
i also am doing stuff like this
if ((_allPlayers findIf {((_x distance2D _spawnPos) < 300)}) isEqualTo -1) then {
not ideal 😄
spawn or spam? 😁
Are “this part of this model is glitched” “this part of this model can't be retextured” or such graphical minor issues subject to be reported as of yet?
they have been in the past and it is possible to report them, just "quite unlikely" to be fixed as it would be Mr Reyhard to be eventually working on it
if it's a smol thing, the chances increase
sad nois
ye, data is quite a very low chance as of now (unless it's some 3D config like "hey this surface is actually texturable" or whatever 3D artists say, I am not one :p)
commands, functions, bugs, crashes, these are "code only" and way easier to bring 🙂
well it doesn't check after every expression if the 3ms timer ran out
And also running only scripts a bunch, instead of a bit of script and a bunch of other stuff and a bit of script again, is nice for the branch predictor
I know, but shouldn't be noticeable, because sqf is much slower than branch predictor
ok if you say so 
idk you tell me 
when you check the execution time for 1 (which is the fastest thing to run in SQF) you get 0.0003ms = 300ns
is the branch predictor slower than that?
[]spawn {
_start = diag_tickTime;
for "_i" from 1 to 20000 do {1;};
systemChat str ((diag_tickTime - _start) toFixed 10);
}
0.003 seconds
unscheduled 0.002 seconds
diag_tickTime is sadly limited to millisecond precision, so you can't really measure it properly.
can't use diag_codePerformance as its unscheduled
Also running a script with one instruction, won't train the branch predictor for running scripts
But it saves a the timer check. Its faster.
No one small line of script is probably not measureably faster, but if you run your whole mod in unscheduled, it adds up
_start = diag_tickTime;
```_`private`_`_start` 😝
is this what makes having many spawn "threads" active the same time, or rather the constant swapping between each?
one spawned script can run at most 3ms per frame (longer in loading screen) after that its stopped
is it a bug that diag_stackTrace doesn't show the for variable?
i.e:
for "_i" from 0 to 0 do {
systemChat str diag_stackTrace
}
_i is not listed
however, _x and _forEachIndex are listed in forEach 
A divine signal that forEach is superior.
have you tried from 0 to 1?
yes
welp
divine victory indeed
0 to 0 is valid anyway
oh yes, just wondered if it was initialised (well, it should but who knows the shenanigans behind it)
It'll happily list local/private vars that you set to _i.
My first attempt at checking this caused a hang on hashmap circular referencing :P
one spawned script can run at most 3ms per frame (longer in loading screen) after that its stopped
well yes, but i was asking something different 🙂
people on the discord here (and elsewhere) say you shouldnt have too many spawn uses at the same time
my question was why that is specifically (context switching or your said check for 3ms limit after each expression - or sth else altogether)
because the scheduler only processes 3 ms of scripts every frame
now if you have 100 spawns, and each one takes more than 3 ms to finish, and you run the game at 60 FPS, that means each script is processed every 16ms * 100 = 1.6 seconds
so a script that would take 7 ms to process, now takes 3.2 seconds
so it got 3200/7 -> more than 400x slower
If you have a lot of scripts that are eating unlimited CPU time then you're doing it wrong anyway, but merging them into one spawn is in practice a good way to prevent worst-cases wrecking everything.
I'm somewhat curious about the scheduler properties though. Like if you have a few hundred spawns on (very) long sleeps, does that have a significant cost?
well this is all clear and obvious, but again not the question raised 🙂
again various people stated having fewer spawns (doing the same code) is more efficient - as Dedmen brought up the 3ms runover check done after each expression, i've asked if thats the reason, context switching between the different scripts/"spawns" or something else
if the question is
why you shouldnt have too many spawn uses at the same time
to avoid clogging the scheduler, take the least print in it?
let me an example
[] spawn {
{_x setVariable ["XXX",time];} forEach allUnits;
};```
vs
```sqf
{
[_x] spawn {(_this#0) setVariable ["XXX",time];};
} forEach allUnits;
[obviously this doesnt make sense - yet in generalized form - process allUnits "at once" or give each its own spawn "threat"]
ie if you have some delayed check you can spawn+sleep, or have a loop with a timestamp on the unit to determine the execution
well, it creates/compiles 50 spawns / scopes + registers them into the scheduler then run
SPG-9 gives 2d sound in 2500m radius
https://feedback.bistudio.com/T167073
quite similar to the problem "2d sounds of car brakes"
request for limiting the rotation/elevation of turrets
https://feedback.bistudio.com/T164568
again not what the question is about 🙂
I am afraid I don't get it then 😄
I still don't get it?
you ask "why not use many spawns", I answer "because scheduler" "that's not the question"
gniii 😄 I am missing something perhaps
the 3ms thing I believe checks after every statement (;-separated things)
i am asking a detail question. not a general rule of thumb assessment
if the 3ms check is done after every statement, it shouldnt matter how you structure your code for scheduled
obviously there can be easily a higher initial load by using multiple spawn, but if its just an one time thing, it wouldnt matter either
also unscheduled doesnt have to do the 3ms check, and the run code can be optimized by "the engine" better/more easily than scheduled
but the question was about scheduled vs scheduled - one spawn vs multiple doing the same code/execution result
hence my suggestion its likely the computation to switch from one "script" to the next
ie the engine likely stores each script on a stack
well yeah scheduled (one big spawn) vs scheduled (multiple small ones): you split the same code in multiple "threads" for the scheduler to check instead of just one, that's about the overhead there is
executes them one by one, when not finished within 3ms, will be put back on the stack (but on the bottom)
so if you use multiple spawn you cause more switching from one to the next
but all this is basic guessing
and the engine may well have some optimization mechanisms
Might be possible to set up test cases with the profiler.
The scheduling algorithm of the RV engine is described on https://community.bistudio.com/wiki/Scheduler.
Like Lou said, both scheduling and context switching take time. I played around with the following code (in an Editor preview on an empty VR map) to see when performance starts to degrade:
private _target = 10000;
Counter = 0;
_target spawn {
private _startTime = time;
for "_i" from 0 to _this do {
[] spawn {
sleep 5;
Counter = Counter + 1;
};
};
waitUntil {Counter > _this};
systemChat format ["%1 s", time - _startTime];
};
```I was going to report results here, but they are far too inconsistent (e.g. most runs with `_target = 9000` finished in around 5.05 seconds with minor to no impact on FPS, but I've also had one run with `_target = 9000` that went down from ~165 to ~25 FPS and took 40 seconds), so I suggest you try it yourself if you're interested.
Around `_target = 10000` seems to be an interesting region. At some inconsistent point (e.g. `_target = 11000`), the FPS will drop to single digits and stay there for a long time while the code is being processed.
So it seems the RV engine can handle a couple thousand active script handles - but keep in mind: Every script handle running the code above does nothing but immediately telling the scheduler "no thanks, I'm sleeping, give the remainder of my execution time slice to the next one in line" most of the times it receives an execution time slice. During normal gameplay (*especially multiplayer*), FPS are also much lower to begin with, so if you had several thousand active script handles during normal gameplay, I suspect that there could be several seconds between two execution time slices for the same script handle even when scheduler / context switch overhead (or whatever else is causing the massive performance decrease produced by the code above) don't impact performance yet.
Hey I just thought of something:
Is it possible to have a new command @ for hashmap get that is the same as # for arrays?
like, "the same but with higher priority"?
(in that case I would overload #)
what I would like to see is an exit comeback - that would have the exact same behaviour as if (true) exitWith {}; 🙂
yes
hashmap # hashmapKey
why though, are there really cases where get needs more prio?
it's shorter 😛
bad Leo, bad! 💦🐱
well, my request is equivalent to breakOut "" as well
btw the reason I asked for this was for chaining # and @:
_hashmap@_key#0@...

and the reason I didn't want an overload for # was to distinguish between array and hashmap easily (same way we have get, while it easily could've been an overload for select)
btw KK did you add the multi column tooltip support for listNBoxes?
it doesn't work properly:
https://feedback.bistudio.com/T167088
it only supports 2 columns
Wasnt me, maybe @untold sky
A QoL idea: adds the icon in ArsenalEden's Assets Browser like this. This feature is from one of my Mods, but it does heavily scripted solution to avoid getting removed when I search the browser. The thing, if it is implemented into Engine, is pretty
Arsenal or Eden? 
there's an opfor FIA? 
Also urmmmm... can we have a news about so-called Community Involvement?

👉👈
I thought FIA was indfor and blufor 
Yeah FIA has all BLUFOR/OPFOR/Independent equivalent since Bootcamp update IIRC
Spam comment: https://feedback.bistudio.com/T167136#2348647
https://youtu.be/IXDaHrGk2ak?t=96 any idea on why this happens (keep an eye out on the fps hint)? when you teleport AI around and change their behavior to either CARELESS or SAFE (others are fine) there are decent fps drops, i'm guessing there's some caching or something going on? (it becomes worse and worse with the teleporting, timestamp at last 20 seconds as that's where it gets really bad)
only CBA loaded, tried without it, same thing
Depends where you teleport them.
Afaik AI in careless and safe always try to use the road
Now if you teleport them the planning of the path has to be redone to find a road
So I guess that explains the FPS drop
Tho path finding is async 
and I can't imagine that drop lasting more than a couple seconds if it was doing planning
^
Actually iirc a path keeps retrying for up to 13s
So is this a persistent drop or only for a short period after execution?
persistent
tried for a minute straight
at the end switched back to "AWARE", and it was normal again
well, that's horrifying :P
If they're not teleported then there's no fps difference between aware and safe?
Or what if teleported a very short distance?
see the beginning, no difference
haven't tried
I wonder if it resets if you recreate the group.
We can't afford to recreate units because our loadout code is expensive :P
it should yeah, didn't have time, have some other stuff to do, but that was my next thing on the list
i was creating the AI at [0,0,0] and setting pos manually (then realised that my fps was shit for some reason), as it's alot faster, unless you use the can_collide
We have a lot of code that creates AI on a point and then places them afterwards, so potentially similar issue.
if you do set behaviours to either of those, then yeah
is it possible to get bombs to trigger the incoming missile event
I don't think so.
would it be a good idea if remoteExec would report error if the function it tries to call doesn't exist? because now it just silently fails
like client would report error if the function it's supposed to have doesn't exist
report error, no
log on client, I would say "why not" but I am not a network dev (and again, could be abused by an ebol scripter)
what's ebol?
a funny writing of "evil"
hmm they're already shotmissile... and even dumb ones AI treat like a guided missile
it seems the engine queues sounds played with playSound if you are tabbed out, and plays them all at once when you tab back in
worth putting on FT?
(needs -noPause obviously - havent tested with MP yet)
yes, probably maybe. but its probably for reasons I don't want to fix
Why too many?
before every frame the scripts are sorted by who didn't run for the longest time, gets to run first.
if there are looots, then that can be slow
Yes but I hate it 😠
it would be nice for chaining arrays and hashmaps tho 😛
and easily telling which's which
I don't remember that, is that new?
yes:
Since Arma 3 v2.01.147037.
I like it, but not sure with mods some will have icons and some don't?
mh can't remember
ok. any ideas for a workaround? tried to enclose the playSound with https://community.bistudio.com/wiki/isGameFocused but doesnt seem to help from my testing
and isGamePaused
same with hint sound, if eachframe it then alt tab it all plays at once
you have to give an example when it accumulates on tab out
ok thanks will add isGamePaused too
we use playSound for conversations, so you can hear them best
these play either during the briefing phase, or mid mission
another case is for ambient sounds that should not be directional / 3d
Some actually don't. But majority of Mods do have configFile >> "CfgFactionClasses" >> faction >> "icon"
PS. What the icon supposed to do? Is it used anywhere in the game?
Spam: https://feedback.bistudio.com/T167194
+2 spam comments: https://feedback.bistudio.com/T166738#2344851
Spam comment: https://feedback.bistudio.com/T166800#2349627
Made ticket about freeExtension not working with C# DLLs https://feedback.bistudio.com/T167202
Hello can someone please update the BTR-K to BMP-K , BTR implies its on wheels and not on a tracks
It does not
also, no BTR-T in Arma 3 vanilla
K?
well, this request?
the channel is about #arma3_feedback_tracker , no?
You probably misread it Lou 👀
It is probably still night time on Lou’s planet
we have btr-k in A3 ?
kamysh
Yes it is BTR-K
Not T, you know, the difference between T and K... Slightly different you know? Even K has more line
you’ve been ARMARED!
mom come pick me up, people are mean to me here
is this just about the naming, or tracks not moving in MP since some network optimization with animations?
@frigid quail?
https://en.m.wikipedia.org/wiki/BTR-T not all of them
ah that explains the name then
but anyway - i assumed that was what the complaint was
same, but awaiting confirmation ^^
Yes that’s the complaint now first question does the kamyesh is IFV or more of a APC ?
It's kind of debatable, and if we were starting from scratch I might advocate for it to be called a BMP, but ultimately it doesn't really matter and there's 9 years of existing content that refers to it as the BTR-K. It's kind of late in the game to give a damn about it.
and that ^
True
It’s better to think it’s referring to the BTR-T as nex gen
That makes more sense thanks
BTR-50 was also tracked. BTR does not mean "it has wheels" xD
BMP basically means the same as IFV
BTR means "light weapon carrier"...
Kamysh is more the second than first IMO. But that's debatable.
Or am I mistaking it with the AA one 
no idea, and anyway not renaming so we can close the case 😄
BTR literally means armoured transporter, no reference to wheels or tracks, while BMP means IFV, again no reference to the wheels or tracks
the same way Peugeot is known to make cars, but they also make excellent pepper grinders 😛
I mean, Kamysh is probably more an IFV given the Marid is the APC. It's heavily armed and armoured too. Feels like a mix of a BTR-T and BMP-T.
But yeah like Nikko said i think we've all made our peace with vanilla naming schemes 😄
(e.g. the Marid meanwhile gets a MSE designation... What does that mean??)
(Most Squishy Enclosure?)
From https://dev.arma3.com/post/spotrep-00105:
Tweaked: IncomingMissile Event Handler now provides the actual missile as argument
Thus, https://feedback.bistudio.com/T157204 can be closed as resolved.
done, thanks! 👍
It looks like new thermals broke nightvision on some maps ( including livonia) for winter months at darkest times
the thermal changes didn't touch nightvision tho 🤔
My bad thermals
But is it accurate that thermal can't distinguish thermal difference? What i would expect here is that bot would be bright as sun on somewhat visible terrain features
I assume there is no difference :/
I need to check tomorrow. The terrain min/max temperatures were adjusted
24 november 2035 midnight on tanoa
it hust looks like camera will not go below some threshold because even if will remove that bot environment would still be pitch black no matter how i look at it
I suspect the temperature is negative
Please run ambientTemperature in debug console and send me the output
[-4.33386,-4.33386,-4.33386] 24 november 2035 at midnight Livonia
yeah thats it. @sterile nova terrain temperatures cannot be negative, thermals only work on 0-60°C range.
At negative temperatures, everything gets clamped to 0, so everything is same black temperature and you see nothing

needs a hotfix
The fact that you haven't doesn't mean that nobody did.

Who would have thought so, right?
oh no my role as a cas gunner has become more than just mindless point and shoot oh nooooooooo
me
read please
does my split personality count as somebody else
look, it's not a vote to whoever asked or whatnot.
even if one single only lonely person asked, and BI deemed the request as valid, it's valid, period
the TI could use some improvements, they were provided
there is a drawback that is not out of reach of a hotfix, hold on for a while
ok whose fault is that? bi or the mod makers for not making sure their mods work in rc branch?
ok bro
its clear nobodys going to have a coherent conversation with you isnt it
how?
where are your tickets?
your feedback tickets for these fucked up gamemodes and mods
speaking of thermals... seems there's a banding issue over the ocean
🅱️eport 🅱️he 🅱️ug
how is this breaking game modes (and mods) 😂
I haven't heard anything about that?
oh nooooo a tiny team couldnt test every single aspect of your update 🥺
(stop bitching about shit if youre going to take no steps to help make it not shit)
The red X was a incorrect display, it didn't break anything
hey, language and no flaming, though
let's keep it civil
what is flaming
There was the busted 32 bit bug. Trouble is that everyone who could actually test the dev branch was running 64-bit :P
but the paid servers often ran 32-bit to save a bit of RAM.
look up flamebait - in short,
flamebait is posting something that is intended to elicit strong responses
lamebait 
i thought you werent part of any team or server
it just seems like youre trying to search for problems where theres very few (that youre not actually mentioning ngl)
we get that
oh, game update, and a feature is broken, fine.
it doesn't mean that the update is shit, the game is entirely broken, etc etc
you report an issue, it can be fixed. think a bit, is BI willing to bother their players or to improve their experience?
so yeah, it's a pain, but don't take your frustration out on the first person that walks by
the flamebait warning above also applies to you.
(and use the legacy build in the meantime if its that much of an issue for you 😎)
Cooperate than hate
...?
Can you define your issues then?
the issue is known, it is a temperature range to fix
hes not actually said thats the issue
true
I mean its better to cooperate w the devs by reporting concerns and issues than hating the update
I mean I have pictures of what thermals looked like before 2.10 and uhhh.... I wouldn't exactly say they were even working
hes just come in here and gone i hate everything about <random feature> without describing what is wrong with it
the known issue is that at certain dates it's -5°C, but the thermals only range 0..60°C
hence something full black that shouldn't be
i think blood showing up as warm in thermal is pretty fuckin neat too 😎
I got that 😉 thanks for the message btw
yep
top of the tower 😄
I see one immediately yes
pahahahaha
if it's taken during the day the tower being so bright (hot) is normal
Thermals are no longer unrealistically totally OP.
Thats not a bug or issue, thats intended
ok because theres been a change to a feature youre so used to
...because its more realistic
its how it always should've been
Get used to change
arma is a milsim game not a roblox game
if the environment is warm, the person won't stand bright white in it
at night however… it's Christmas
yeah there is, the thermals
but only now
Then you're playing a different Arma than I am.
Which yeah, fair. if you don't play Arma as a realistic game you may not want it to be realistic.. but.. Its still supposed to be a realistic game
and the night vision
omg ive just had a great idea
blinding people with ir lasers
oh sh—
"authentic"
:sadcat:
an authentic military experience as you make the same repeated war crime joke as your vehicle flips because you hit a rock
The new thermals are still shit, but atleast alot better
maybe because you dont sit at a chair and computer in irl warzones
that's still the closest you get on the market, in that combat range
…and that still won't revert the thermals 😄
If your only issue is that its more realistic and less OP now, then sorry, thats not a reason for us to change it back, because thats exactly what we wanted to achieve
yep, that's by design
no, because the difference of the heat of the building and the person isnt as great during daytime when the sun is shining on everything
set your time to night or a different time when the people are in the shadows
thats how thermal should be working and thats how thermal now does work
There are lots of issues with the new thermals, but that one isn't one of them
could you see players through windows in the old thermal? iirc you could?
great you cant do that any more
which is realistic
Thermals in general are going to still be effective in open ground but in a large city it should be fairly difficult to spot infantry
Indoor temp affects person’s temp?
that's called "moving the goalpost" and it's seen as bad etiquette
any other question or are we good?
peoples bodies clipping thru the walls isnt an issue with thermal either
It just will require a change in tactics based on your environment
okido!
yet even if we disagree, thanks for your concern about the game
Q tho
oh no
NO
you mean this?
In the recent update, does the building’s indoor temp affects the body of the person?
not as far as i can see
No.
The main thing that changed was that sun now heats objects based on sun direction. (That was always supposed to be there, but there was a bug in code that broke it since it was implemented 😄 )
but it should affect the temperature of other objects, because of shadow 
meh being able to actually see peoples blood through grass and shit is actually pretty nice so it has the kjw seal of approval 👍
(a very prestigious award)
If I may ask, what is ambientTemperature doing on the backend to calculate those temps? I assume its largely based time of year for the terrain's lat/long or is that something in the terrain config?
terrain config has temperature ranges
yeah. i can post a couple more images too if required.
Ok good so if I wanted I could make a patch mod for the terrains I utilize until an engine/data fix is applied?
What about it?
There is a bit too little variance in the grass, unless you have grass disabled there
you can i already did this
...as well as manage to break the night sky when doing so
yeah
Well I think there might be one near the center, but I'm not sure. Probably wouldn't notice it in actual playing
whats the range on that screenshot
Very realistic
^
Good, exactly as intended then
yep?
I don't think I actually tested ghillies
good, more realistic now
for ex. https://youtu.be/oTex0jB4xNQ
There is information out there about how someone can just wrap a space blanket AKA survival blanket AKA emergency blanket around themselves and they just disappear off thermal. There is another one saying just shoving an E-Blanket under your coat makes you invisible. And another that a Ghillie suit will mask your thermal signature so much you d...
ghillie made with actual ground material performs better too
so, are you going to post a screenshot of every change you notice or accept that there have been changes?
ok
Peace to us all thanks for the update devs
The ghillie image should have more variance on the ground thermals.
But engine limitations mostly
What did I read? 🤔
You are right of course this is why arma footage is repeatedly used by news agencies who can’t tell the difference. Now could you please go away and let people post real issues so they won’t get lost among your whining?
@untold sky once you said you would put on your list to do, adding illuminating under-barrel grenades (illuminating flares) to vanilla, since there are none. there are just those that indicate position that don't illuminate the surroundings, but only themselves
Well I guess this hack doesn't work anymore in 2.10. It was "working" previously and was an attempt to do some potential optimization on server/HC performance
// CAManBase init via CBA Ext EH
if (!hasInterface) then {
(_this select 0) addEventHandler ['FiredMan', {
params ['_unit','','','','','','_projectile'];
if (!local _unit) exitWith {deleteVehicle _projectile};
}];
};
Seems the deleteVehicle call is now getting networked to all clients for the projectile, previously it wasn't and the projectile still behaved correctly.
We need every thing licensed guns and other things in arma 4
I don't think it's necessary to restart that conversation
@vocal plover
you will fire Cold M17 and you will be happy!
doesn’t matter
It was a subtle warning, don't flamebait.
I am not the person adding content/items, I won't be doing that
then maybe @solid marten or @sterile nova can add illuminating flares to vanilla, since you said it wouldn't require much work, besides localisation
It has been on the todo list since quite close to the games release
This is a fine place for feedback, though FT is better as you dont get multiple conversations crossing over.
In general Feedback works best if its constructive.
You tell us an issue, how you got the issue and why its an issue for you.
Bringing up previous updates and issues that were unrelated didnt help.
cheers
For some reason my tanoa at that time is 18.9°C 
December midnight is even 20°C
Or right its southern, so june is winter
nope june is also warm :/ what wrong
So “Press X to Win” in 2.12 pls.
can mods change that?
yes but I have none :sad:
ye but then the other person might ^^
Maybe ace?
@lean mica sorry accidentally deleted your message.
Mods shouldn't need to adjust, we'll make a engine change that should fix it
Something must've gone wrong with steam updating my game, I still have the old temperature configs 
why was i tagged
But is't it somewhat correct considering it's geographic location. Close location to Tanoa
https://www.timeanddate.com/weather/vanuatu/port-vila/climate
Dunno, but whatever issue was there is now fixed
It from the start was not affecting all maps. Mostly those located in north Eurasia like BI Livonia and CUP Chernarus
I could repro on chernarus winter, and fixed it there
OK than
Stratis, 2035-07-06 (default date) 00:00:
road is warm, and heats a bit over time (or the thermals are calibrating)
2035-01-01, 00:00 - horizon banding (and nothing else visible because < 0°C but it's known)
Don't want to be annoying but can i get an answer for this ticket https://feedback.bistudio.com/T163286 Will there be a solution for it or too deep inside engine? And for clarity i provide another name for it - No ammo sensor can lock beyond ObjectViewDistance, no matter what config settings for sensor say.
That's the normal overlay used by all vanilla NVGs except fullscreen ones like the Viper helmet.
since when did that appear on thermals 
shouldve gone to specsavers
twice you say that, on the third occurence I bring you in myself
i am actually the start of the arma 3 campaign
the S in CSAT stands for specsavers
i have code in extension that causes buffer over run in the version function, if you guys are interested?
Why would we be interested that you have a broken extension?
because the fault maybe in how arma handles extension output
no there is none, if you are causing a buffer overrun then its you
well i can show the code if you are interested then you can say what's wrong
my bet is on "no" 😄
he's keeping us exited 🙂
I think there are unlimited number of ways how to crash arma with corrupted dll.
Let me guess, you pass something unintended as plug-in version and engine does not sanitize it.
what I'm doing is something I have done many times before but it doesn't work inside a DLL i guess. though i saw nothing wrong while debugging
not in 2.10
start writing in C++…Did you sort out why you had double dll running?
must be arma loading it twice
You just need to crash extension and Arma crashes, but this is expected
everytime I use callextension the DLL gets loaded twice
but not if you load C++ extension, right?
yeah C++ is fine
Do we have C# specialists here?
I did work with it for 4 years, but I wouldn't call myself a specialist no 😄
I'm not a specialist but I guess it's due to .NET runtime refusing to let it go 
doesn’t explain why it loads twice
Dedmen said this:
#arma3_scripting message
In regard to the new thermals, with the goal of them being more realistic, has the addition of variable brightness options been considered?
yes
CLR will be loading the dll a second time, nothing we can do something about
^
but if you're using C#, you shouldn't need to reload the dll, you can just edit the code when you're debugging (look up edit and continue debugging C#)
you can also use plugins (unload/load assemblies, and have the extension drive them) (MEF like) kinda thing if you want to 😄
ive tried that but it did not work
which version of .NET do you use?
4.7.2
it has the "apply changes" button but the new code isnt read in
i can try it again later but I was already tinkering with it without any luck
sometimes i cant get even the debug break points working, like now... i wonder what part the .pdb file plays in this
or maybe that stuff doesn't work because there are two DLLs loaded
As a temp fix I made a patch mod that updates CAWorld's config to clamp all negative Weather temp values to 0 and that actually applied to most modded terrains I had loaded, only a couple specific terrains had a hardcoded value.
thats fixed in tomorrows profiling
sweet
https://feedback.bistudio.com/T167206
Has anyone else here experienced this?
yes
yeah doesn't seem to work for me either 
just use C++
Angara not commander variant has misaligned zasleh
So the inventory bug was never completely fixed: #videos_arma message
"the" inventory bug lol
Are you editing the RVExtension function, or a different sub function or class
Neither worked for me
In RVExtensionArgs I added another output.Append("_New") but it didn't work
I made another method called by RVExtensionArgs that printed to the console Console.WriteLine("test")
I couldn't change that console message either 
Debug build
RV
this change is not popular
Fixed: Incoming missile warnings were played through the radio instead of the effects channel - FT-T163245
this affects "ear plug" scripts which decrease effects volume, especially in noisy helicopters
so pilot when using ear plugs is not able to hear missile warning
need to script a replacement 😛
cant see any easy engine based solution, do you @west onyx ? (other than the old state which isnt great either)
I'd like to request commands for filtering objects quickly when the filter parameters are constant.
I can only think of class and side filters.
https://feedback.bistudio.com/T167261
Any other ideas?
so pilot when using ear plugs is not able to hear missile warning
makes sense, no
Why do I have a feeling this has the same vibe as thermal overhaul? Turn everything down, and only listen to missile locks, sounds like an exploit
a bunch of videos at the end of that link, what is it
as a RL helicopter pilot you are wearing sound suppressing "ear plugs" too
and im sure military aircraft have important audio cues delivered into the headset
i think most arma pilots would be using reduced Effects volume
how does the script work, fadeSound?
earplugs? yea i think basically fade sound
so what is the solution?
long term (2.12) i think piping it through radio was fine
short term, community will adapt
grabbing file path and duplicating the warning thru playsoundUI if the effects volume is low
If player is wearing helmet (or has radio) route it thru radio
Otherwise, effects 
i think it worked fine before, or at least it wasn't a common feature request or source of frustration. i disagree with changing source based on helmet/radio, i think its just adding complexity to something that just worked fine
hmm
on my end, I will just detect the "earplugs in" state and play the duplicate warning if so
(or has radio)
Dont think this is relevant, in vehicle radio has nothing to do with the radio
I like the helmet idea though
yea flight helmet
yes
not sure if helmet classes get that fine grain tho (is there anything in config which distinguishes?)
i might do the same for duplicate warning, play over radio if they have a flight helmet
how about we play it through effects, unless effect vol is lower than radio vol then we play it through radio?
whatever is easiest to implement 😄
that would be the easiest
sounds good (pun intended)
Now we need to test it
Yes that won't work.
Function editing works by (often, not always) copying the function, and putting the new one into a new place, and redirecting all references to point at the new function.
But, Arma only gets the address of the RVExtension function once when the plugin is loaded, it never notices if the plugin changes its own function address (because normally thats not possible)
Put all your code into a sub function that you call from inside your RVExtension function, and it should work
for me it doesn't work
public static int RvExtensionArgs(StringBuilder output, int outputSize,
[MarshalAs(UnmanagedType.LPStr)] string function,
[MarshalAs(UnmanagedType.LPArray, ArraySubType = UnmanagedType.LPStr, SizeParamIndex = 4)] string[] args, int argCount)
{
outSizes[1] = outputSize;
if (!hasConsole)
{
AllocConsole();
hasConsole = true;
}
foreach (var arg in args)
{
output.Append(arg);
}
Test();
return 0;
}
static public void Test()
{
Console.WriteLine("{0}, {1}", outSizes[0], outSizes[1]);
}
I update the Test method and change the output (e.g. "{0}, {1} update")but it doesn't work
Is it worth a ticket requestion a parameter for line thickness for https://community.bistudio.com/wiki/drawLine3D ?
And 2D 😀
https://docs.microsoft.com/en-us/visualstudio/debugger/supported-code-changes-csharp?view=vs-2022#unsupported-scenarios
"Mixed-mode (native/managed) debugging." shouldn't apply, that references the debugging mode not how the code was loaded 🤔
Maybe it doesn't work because VS doesn't just edit but also creates a copy of RvExtensionArgs, even though it should only edit a pointer for a call.
or your call to Test() is inlined, but in debug mode it shouldn't?
is there anything else to C# besides these?
Default settings should allow for it
yeah those are default. I just changed output type to "Class Library" and added WIN64 to the Conditional Compilation Symbols
I'll maybe test it in a few hours, I have some workaround ideas that may work
maybe it's the two DLLs that confuses the hot reload. having difficulties to get debugging working also
My visual studio doesn't even let me edit
Severity Code Description Project File Line Suppression State
Error ENC2018 Changes made in project 'ArmaExtensionTest' require restarting the application: ArmaExtensionTest O:\dev\ArmaExtensionTest\Class1.cs 18 Active
Error ENC0097 Applying source changes while the application is running is not supported by the runtime. ArmaExtensionTest O:\dev\ArmaExtensionTest\Class1.cs 1 Active
A proxy DLL should work?
Make your C# DLL load another dll that has RVExtension export, and let it forward all the things.
As the second one is not loaded from unmanaged code, that should be editable :harold:
I see the person who wrote the wiki makes the same typo's as I do
on the topic of audio @solid marten in the audio options there is a "Map factor" slider ... is there a way to access this value in sqf?
thinking that this would be a good "earplugs" factor for earplugs script
whoever made the thermal update thank you very much
you single-handedly fixed warlords
goat developer 🙏 🙏 🙏
what was wrong with it before? 
thermals were far too unskillful for warlords
basically esp
since theyre trash now warlords is pretty good
Isn't there a getaudiooptions command or smth like that?
getAudioOptionVolumes
https://community.bistudio.com/wiki/getAudioOptionVolumes
but nothing related to an opened map?
Uff i think it should've been added there
i would be very happy if my feedback gets taken int account 😃
taken into account yes
answer:
no
why 😠
This one, at this specific time: https://youtu.be/a7FLyLSqhN8?t=99
TCs plays in the Uncle Games against each other. All hail Uncle. Enjoy this video of this event made by Zapper and Stormhammer.
The Cooler Server is an Arma 3 group. Apply here:
https://www.thecoolerserver.com/
TWITTER: https://twitter.com/darkliberator
FACEBOOK: https://www.facebook.com/DarkLiberatorZone
TWITCH LIVESTREAM: http://www.twitch...
I can't shake the feeling this suggestion was not made in good faith, but just in case:
- it's not realistic; 60 rounds of 7.62 would be huge and I've never heard of such a thing for a DMR
- it would require a new model, which is highly unlikely at this stage of development
- it doesn't fill a necessary gameplay role
- it would require conversion of the Mk-I to the magwell technology, which is blocked by the fact that it shares a magazine with several other guns, a problem which is too complex and too non-backward-compatible to solve now
- in summary: too much work required for little to no benefit
Never had/saw this bug in 15k hours of dealing with arma.
So i blame their mods/scripts.
I've had that issue many times in MP. Can't say that it wasn't the mods or not but I don't use many. Your 15K hours are in MP or SP? Because I believe it is a MP specific problem... and also might be related on how the mission maker places those weapons perhaps... can't tell but I've experienced it many many many times
Never played SP 😆
I think that can happen if you disable simulation on crates (if i remember right).
https://feedback.bistudio.com/T61518 the issue has been here since the beginning. In the video you can see that players could not pick up from weaponholders and crates while I have experienced the issue when picking items from deadbodies. I remember some dev saying it was a locality issue so perhaps the container/body could be made local to the player who used the addaction first. The downside is that no other players could look at the container while the first one is still there... Don't know how more modern games such as reforger solves the "locality" issue but there must be a solution
https://feedback.bistudio.com/T126030 👈 this illustrates the issue better, but not better than the MP video proof I linked before
"little to no benefit" best gun in the game
"- it doesn't fill a necessary gameplay role"
literally a razor
best gun in the game should get justice by getting a 60 rounder
" I've never heard of such a thing for a DMR" bro you live in 2035?
if you want it that badly just make it yourself 
yeah then run my diy 60 mag in an official warlords server?
no. can't do that one
"I really like this gun" is not typically good enough reason for the devs to give it a massive unrealistic buff, especially at a stage of development that means the required changes are not practical. Nothing gets new models at this time and getting existing models changed is a rarity.
I can now confirm that it is not a modding issue... Tried it in vanilla on a hosted mission with a client. Host could pick up items but the client couldn't until I changed to locality of the container to be that of the client...
That might explain why I never had it. Never played a game mode with arsenal allowed (FBT ticket)
And containers were always created by the server (dedicated)
Maybe you are like me and have 14k hours in the EDEN editor and 1k actually playing it in MP.
Indeed it happens less often on dedicated servers
use the 556 60 rounder ez
i would buy the same model just reconfigured for mk-1 emr
20 dollar ldc
dlc
I don't have anything I can say in response to that. Please don't misunderstand that as you having convinced me it's a reasonable suggestion; it's just that I can't find a way to express how stupid this conversation is.
Lou's already told you the answer, I'm just trying to explain why that is the answer. If you don't want to accept it, well, that's unfortunate, but it's very unlikely to change.
I am curious if this is an issue that will ever be addressed?
im joking btw
the mk1 emr is the best weapon in the game a 60 rounder would unbalance the game even more
Don't know but inventory management is one of the cornerstones of Arma so one would think issues with it would be a priority. Additionally we not only have several feedback tracker reports but we now have video evidence as well so...
I think you guys need to see comments about the new "Thermal Imagery Enhancements" and fix that issue please
https://store.steampowered.com/news/app/107410/view/3338876362833761030
People sad about TI nerf. TI was OP and far from reality. Now it is much better. You need some attention and effort to use it now.
TI fix for winter maps should be already in perf branch.
no issue bro
its just a thermal nerf
skill issue if you cant play without it
@lean mica @vocal plover should I remind you again not to flamebait or is it going to be ok from now on?
it looks alot better, just needs some tinkering, but it's hard because you just have to experiment with values randomly 😄
good, thanks
I don't think any of you play as infantry , you mostly play in tanks or jets
i literally main infantry in warlords
Flaming is when you bait for response. My message does not require for any response. It is just statement as it is.
Rly🤣
#videos_arma even uploaded a infantry compilation today after the thermal patch... no offense tho
you calling people babies is flame-baiting, or whatever you want to call it.
don't do it.
I'm not waiting for you to comment about it, I raised an issue here and really need a dev response not yours
either they can say we will keep it or we will change something
if this place is not for giving feedback , just letus know and we will leave to somewhere else
it will be kept, most likely tweaked
Fixed
for me no more play 🙂
and this channel is mostly about the feedback tracker, even though some feedback conversations can happen (as here)
that wasn't conversation that was a way further than that
don't want to make it arguing now
There is no issue. People always sad when something OP was nerfed.
Wheaton's Law as usual
mkay mkay
so ragdolling people with ifrits then finishing them off when theyre in the animation is okay
Well I am sure the devs will always apretiate constructive feedback and grounded suggestions. The new system is to stay because it is far more realistic but that doesn't mean it cannot be further tweaked and refined. I am sure lots of dev time went into it and for good reason. Now you can have balanced operations without needing to ban the usage of thermals during missions.
climate change night is hot bro
Now that is great feedback. I am 100% sure that the new thermals will still get some love from the devs because they are kinda obsessive perfectionists. It is pretty hard to get them to start but once they do then you can be sure it is going to be as best as the engine can support
Also don't throw jokes randomly at the devs because they take everything very seriously!
My sarcasm filter is active 😝
Do you know if there is a variable in profile namespace for it
or some displayCtrl that can be got without opening the options menu
It also appears that various terrain objects are hotter than or as hot as a human body in the middle of the night.
That is mainly because Arma thermals don't work with temperatures, they never did.
Everything is just a +/- offset from ambient.
In winter body temperature is like 20°C, in very hot summer body temperature is like 50°C. Its all a mess. And because the contrast was fixed with histogram equalization (which basically every thermal imaging device has, but Arma didn't have) it made all the old bad things obvious and visible now
Another thing, I noticed the head mounted uncooled thermal/nvg goggles produce the exact same image and clarity as a 65+ pound
We added new options to control thermal noise and sensor resolution, but we are not using them in vanilla content, partly to not nerf/change them too much, and also because they are 2035 year futuristic equipment
Yes the grass and foliage are definitely too hot.
Problem with toning them down is that it makes the thermals too OP and unrealistic again because humans and vehicles will stand out too much
We tried to improve what we could, but there are also huge limitations because we couldn't just fully rewrite the thermal system, it still has a majority of its flaws, just the contrast improved (which mainly makes it not be totally broken in winter times where previously everything was pure black with just white dots in it. The brightness previously was smashed in a 5-20% range of your total brightness. Now its spread out over the full 100% black to white range) which made these flaws more obvious
my question is, is a hotfix planned to address the most obvious issues?
ye
We have thermals broken in negative temperatures, linux not detecting cpuCount correctly, and epilepsy and game crashes on intel integrated graphics
I can reproduce the strobing on my laptops integrated graphics, so I can probably get it fixed soon 
the more nerfed thermals, the better
long ago i experimented with lowering view distance when in thermal
we didn't have the tools we do now though, like "getOpticsMode"
what a chad opinion
Seems we have a fix for the epilepsy, prof v2 next week probably
btw, has the audio tails system been touched in this patch?
I feel like assault rifles are not "blasting" my ears indoors (but handguns still do)
it might be just me misremembering though 🙂
there were audio changes but none that should be noticable
Go back to 2.08 and test ;P
but I'm lazy!
would it be feasible to have https://community.bistudio.com/wiki/Arma_3:_Event_Handlers#SoundPlayed either also return the sound references (to be able to delete it) or to have a parameter to modify the volume? (aka worth creating a FT ticket for that?)
you want to override it?
What’s going on today?
It's the international spammers day. Didn't you get the memo?
yep
convince me
here or via FT ticket?
@untold sky Should I keep my hopes up on a fix for this? https://feedback.bistudio.com/T136844
Will need a ticket so might as well make one
roger. will do
Maybe hotfix for stable right away?
Who, if not you?
ur mum, HAH !
flamebait
Sweet thanks for the follow up. Assuming vanilla is the stock platform, would one be able to obtain the version with extended options for private use? Or is it a developer only kind of thing
The extended options are config things that mods can use, so if you want to make vanilla NVGs crunchier you can make a mod that tweaks those config options. They've chosen to not do anything with those options for vanilla items in the vanilla game, but they still exist and can be applied through config mods. I'm sure ACE is turning down the resolution on everything as we speak.
I've yet to figure out how to use them because my config fu with PP effects is not up to par yet but I'm sure once someone utilizes it or an example is posted I'll figure it out
I'm sure ACE is turning down the resolution on everything as we speak.
Not really.
Contributions welcome 😉
the reference you want so that you can delete sound?
wouldn’t it be better to override and never play it?
I suppose override could include volume but apart from setting it to 0 to mute what other useful things can you do with it? might as well just make it bool, true don’t play otherwise play as normal
if we are asking willing dev for script commands ... "nearestSimpleObjects"
so we dont have to (allSimpleObjects []) inAreaArray [_pos,50,50,0,false]
sure. on/off might be most simple and flexible
I would like to request a new command that gets all triggers present in a scenario. Right now it seems allMissionObjects "EmptyDetector" is the only way, but this command is really slow.
Result:
0.772201 ms
Cycles:
1295/10000
Code:
allMissionObjects "EmptyDetector"
For comparison
Result:
0.0008 ms
Cycles:
10000/10000
Code:
allUnits
so allTriggerswould be appreciated. Let me know if you want me to create a ticket.
if you only have a few objects you can also do:
63 allObjects 0 select {_x isKindOf "EmptyDetector"}
but it's still slow 
lemmie see if triggers have collection of their own, if so allObjects could be improved
speaking of allObjects, are there "mini-collections" for particles?
like instead of returning thousands of smoke puffs belonging to 1 smoke source, it just gives that 1 smoke source 
dunno need to look
I noticed something weird. it seems that wind is "lagging behind" on actual wind value 
anywayyy...
so here's the problem. when I throw a smoke grenade I manually measure the velocity of the particles (because velocity returns [0,0,0]) using their positional difference.
but it doesn't match the wind velocity. the particle velocity is "ahead of time" (or maybe wind is behind)
so either smoke is broken or wind is
or your PC
here's a video of the problem
initially the smoke and wind are both going at +10 m/s in +x direction
then I use setWind to change the wind to -10 m/s. but the smoke particles are going "ahead" of the wind and quickly reverse direction before even the wind does 
the 3rd watch field shows: [smokeVelocity#0, wind#0, smokeScale]. notice the discrepancy between the 1st and 2nd values
the smoke has already reached ~ -10 m/s when the wind is still at ~ +7 m/s
TMI on that screen, really >.>
only this part matters 😅
even values don't matter. just the signs. in this picture both are + (wind and smoke both going at 10 m/s)
now one is - the other + (wind is at 7 m/s, smoke is at -10 m/s)
to repro:
lastPP = [];
lastPos = [0,0,0];
particle = objNull;
createVehicle ["SmokeShell_Infinite", getPosATL player];
setWind [10,0,true];
onEachFrame {
lastPP = pp; // backup last particles
pp = 63 allObjects 3; // new particles
if (isNull particle) then {particle = (pp - lastPP) param [0, objNull]}; // if particle is deleted, pick a new one that was created in this frame (shortest age)
_pos = getPosWorld particle;
_vel = (_pos vectorDiff lastPos vectorMultiply diag_fps); // calculate particle velocity
if (!isGamePaused) then {lastPos = _pos}; // update last velocity
hintSilent format ["V_par: %1, V_wind: %2", _vel#0 toFixed 1, wind#0 toFixed 1]
};
I'm gonna guess the particle is broken
it seems that the equations are not accurate. I think they just increment the value and then cap it by max wind.
if so there's no need to fix it if you guys don't want. but at least give us a getter command for maxWind
and maybe other commands for "weather forecast"
or just one command for all weather forecast stuff 
btw @solid marten I noticed something weird. 63 allObjects 0 returns terrain objects as well (lamp posts)
can they be removed or at least be moved to another flag please?
e.g. on Altis I get ~2814 terrain objects
code: (warning: it's slow)
_t = nearestTerrainObjects [[worldSize/2, worldSize/2], [], worldSize/sqrt 2, false, true];
_m = 63 allObjects 0;
count _m - count (_m - _t)
I had noticed that before but it didn't seem like a huge problem, because I was testing on Stratis. I didn't realize it could return ~3000 bogus objects 
but right now on Altis it takes ~2 ms which is not acceptable for my use case
what did you expect?
0 - Slow (and very slow) entities: houses, rubbles, street lamps, churches, etc
getterrainobjects?
they are
it shouldn't return terrain objects
if you run this code you'll see that allObjects objs are in nearestTerrainObjects (only lamp posts)
around 2800 objs on Altis, ~200 on Stratis
examples:
After getting Projectile Event Handlers will it be possible to create a solution for "jumping 40mm GL smoke" on the engine side?
see tickets for problem info:
https://feedback.bistudio.com/T158405
https://feedback.bistudio.com/T157121
https://feedback.bistudio.com/T82608
Well it is what it is, I told you, the command takes in game array and gives it back to you, what’s in it is what you get, if terrain objects are in it then they are returned
remind me next month
setTerrainHeight changes players map on execution, is it possible to extend the hideObject command when terrain object is removed so its marker is also be removed?
setTerrainHeight changes players map on execution
how?
i mean this:
https://ibb.co/q9gQQrM
or is it 'other' and doesn't apply to terrain objects markers?
objects != heightmap
can you add a filter to that one? 😬
(only for allObjects 0)
it would be faster than checking that myself in SQF 
what kind of filter, show me sqf
params ["_flags", "_type"];
_ret = [];
if (_type == 0) then { // filter out terrain objects from slow objects
{
if (_flags findAny (_x get "flags") >= 0 && // flags match
{!(_x get "isTerrain")}) then //not a terrain object
{_ret pushBack _x};
} forEach getGameObjects _type;
} else {
// other object types (normal, fast, etc.) is like before
};
that's incorrect but I guess you understand it 
so you want allObjects - allnearedtterrainObjects?
yeah
but I think it only needs to be checked for slow objects 0
other types seem to work fine 
I’ll have a look how it works can’t remember everything off the top of my head
you mean terrain objects do not exist in >0 what about left arg?
left arg is 63
didn't check others
there is no flag for terrain object as far as I see:
1 - Primary - Normal object placed in Visitor, part of landscape 2 - Network - Road placed in Visitor, part of landscape 4 - Temporary - Temporary object (like tracks) 8 - TypeVehicle - Some entity added by game 16 - TypeTempVehicle - Temporary entity 32 - LandDecal - Land decal
oh wait. I guess visitor is terrain only? 
there's a -1? 
you owe me a crate of beer for emotional distress
Visitor is the name of the ooooold terrain editor
yeah but I always thought that in the game, Visitor means the part of the game that creates objects 
because nearestObject also says it needs visitorID but it works with mission objects too
Biki broken atm ?
Yup, already on it. Thanks for the answer 🙂
Will add trigger collection to allObjects. Also another collection @sinful kettle is going to love it
not particles
setPosed objects?
https://community.bistudio.com/wiki/getObjectType
btw what objects have type = 4?
wiki says like tracks. if it means tire marks #track, they are type = 16 (so are footsteps #mark)
I tried every object I could think of (projectiles, particles, sounds, etc.). nothing is 4. there's nothing on any terrain either
almost but no
while on topic of collections ... is there a collection for vehicles belonging to a group
wanting to get all vehicles belonging to a group
Dunno, experiment, all I know it was not always empty, I've seen it returning something
Yes, make a ticket
the "getVehicles" ticket already exists 😬
allVehicles <group> with possibility to add other arguments later?
smells like an array :p
but it is group can have a bunch of them
i dont like assignedVehicles because it is misleading
allVehicles <side> in the future?
allVehicles - transport obviously
well, we have assignedVehicle for a unit
vehiclePool <group> then?
I suggested
assignedVehicles <group>
groupVehicles <group>
getVehicles <group>
in https://feedback.bistudio.com/T166543
note that we have allUnits, allCurators, allGroups… but allVehicles is actually just vehicles e_e
now if allVehicles is only "crewable/getin-able vehicle", I can go with it (vehicles returns weapon holders as well 😬)
I think allVehicles would be misleading. It and vehicles would have basically the opposite roles to what you would expect.
assignedVehicles is fine imo, even with a side expansion (or at least the alternatives are mostly worse). Vehicles sort of have to be "assigned" to a side - by having someone in them - otherwise they're all just civilian by default, so it kinda makes sense
I believe nullary commands cannot be unary
unary & binary is possible iirc
why we don't have allUnits and allUnits west for example
Vehicles sort of have to be "assigned" to a side
no, they dont and this is mighty confusing
they called assigned because unit is assigned to them AS role. so assigned vehicle is not vehicle assigned to unit but vehicle unit assigned to
addedVehicles 😅
vehiclesThatThisSpecificGroupCanUse
yes better but this is waste of command, if we have to add one might as well be expandible
also, a vehicle can be shared among multiple groups, right?
dunno, added vehicle exist in group collection, need to test if several groups can reference the same vehicle
AFTER we decide on the name so I can test the getter
Btw is the opposite possible?
As in getting the groups a vehicle belongs to?
allGroups _veh
And then allGroups _side for the future 😉
Looks like we already have a nullary allGroups 
I thought it was groups
there was an issue in A1 iirc where a vehicle that had a crew member of a certain side would "keep" this side, and enemies would fire at it, even empty and them not knowing the original unit
allTransport?
better
Just checked vehicle does keep track of the group it is added to but only last group, so it can be added to several groups but will only relate to the last group
what a mess
assuming it wont include static weapons?
I would be ok with a command that just iterates over all groups and finds which group it belongs to 
It's slow but faster than doing that in sqf...
don't hesitate to cleanup while you're in there hey!
Almost dont wanna touch it
I get the feeling
smells rabbits
this @west onyx
wat, but why? 
nular commands don't even exist in parse time either (afaik?) it's just an ident that is checked later at runtime if it's a nular cmd or not (why not parse time? :D)
imagine a command like cmd1 that can be both unary and nular
now imagine another command cmd2 that can be both unary and binary
now let's say you write:
cmd1 cmd2 _var
now what is this?
is it ncmd1 bcmd2 _var?
or ucmd1 ucmd2 _var?
you can't resolve that expression hence nulars can only be nulars