Jump to content
Bohemia Interactive Forums

Kerc Kasha

Member
 View Profile  See their activity

  • Content Count

    745

  • Joined

  • Last visited

  • Medals

  • Medals

Posts posted by Kerc Kasha

  5. Posted

    On 10/10/2019 at 1:25 PM, froggyluv said:

    So back to my unhealthy AI AT footman decision woes. It seems that because the PanzerSchreck soldier starts with his RPzB Anti-Tank launcher -unloaded. THIS seems to be a major contributor in them not firing when they have a wide open shot at enemy armor point blank in front of hem.

     

    I placed about 12 of them surrounding an enemy American Sherman (no fuel/no ammo) and NONE of them take the shot! That is disturbing. They all shoulder their launcher, have some moment of "erhm, maybe not", unshoulder their weapon and proceed to Chicken Dance. I then add 1 modern day NATO AT Lite soldier to their troop and he 100% of the times, shoulders his weapon and fires. the difference../ His weapon starts pre-loaded.

     

    Save the WW2 AT men kju

    I updated the ticket regarding this but i'll repeat it here too.

     

    The unloaded launcher issue is fixed.

    I've adjusted the minimum accuracy for AI using our launchers, they can fairly reliably hit targets now (but not 360 noscope unless you set them to accuracy 1 or whatever)

    AT grenades are... better. the ai seems really stupid with grenades in general, most of the time when I tested it the grenades killijng their group mates werent the AT grenades, it was frag grenades from other people in their group throwing them at the tank, landing near the guy who had to move up to throw an AT grenade.

     

    But apart from that it seems much better

    • Like 2
    • Thanks 1

  6. Posted

    OK to explain the SMG issue and why it only happens with iron front ones.

     

    We have an open bolt 'simulation' script. For those who don't understand firearm terminology - an open bolt firearm, to put it simply (so those who know better, please don't eat me alive), is when a round is only chambered after the trigger is pulled. most machineguns and smgs are open bolt.

    This means there's a very slight delay between pulling the trigger (clicking your mouse) and the gun firing, it's in the fractions of a second, something like 0.05-0.08 seconds, this allows us to also play a sound of the bolt 'closing' and chambering the round prior to firing.

     

    now this is all well and good but when the scheduler is overloaded a delay of 0.05 can increase dramatically, which would be the issue being posted. The script is fairly simple so it really shouldn't be getting delayed so much but apparently is.

     

    So for a solution:

    I just finished rewriting the script to run in an unscheduled environment so these delays should be dead and buried with the next update.

    • Like 1
    • Thanks 5

  7. Posted

    1 hour ago, pognivet said:

    what is this speed boost hack? should a move waypoint be given almost to the shore and then the amphibious assault waypoint only a couple meters ahead? is this "speed boost" intended to beach the craft? please elaborate. i am interested.

    in the amphibious assault waypoint, when you place it there's a check box for enabling/disabling speedboost.

     

    it just gives extra velocity to the boat (up to maximum) because the AI moves incredibly slow towards shore because for whatever reason they try to avoid beaching themselves when trying to land. the speed boost was to stop that from happening and have them just speed towards land.

     

    But if you have them too tight together, not facing directly towards their waypoint etc they'll start turning like idiots and this can interfere with the script. I couldn't work out a good solution for it so I simply allowed people to turn it off if they have issues and it will behave as per vanilla.

     

    As for the move waypoint - you want them to be fairly close to shore, a couple hundred meters at maximum before you issue it, and they need to be roughly facing the right direction for optimal placement.

  9. Posted

    #1 and #5 are a direct result of the scheduler being overloaded

    your mission must be running really awfully for it to come to that.

     

    #4 is the 'speed boost' hack I threw in with the sea assault waypoint, it works most of the time but if its causing you issues just switch it off

     

    we can fix 1 and 5 but if your game is running at really low fps you're gonna get some bugs and need to reduce complexity of your mission

    • Like 2
    • Thanks 1

  10. Posted

    33 minutes ago, mcnools said:

    Quick question about the Static Line Paradrop-Waypoint. When I play as group leader of the group that is about to paradrop only myself is ejected out of the plane at the waypoint, if I play as any other soldier in the group everyone disembarks properly.

    Is there something the group leader must do in order to get the others to disembark? I'm meaning to use this in a MP-scenario for my group later and I'm curious how this will work in MP.

     

    If you're planning on using it for players it should work fine, but it's usually intended for players to choose to jump out themselves. To get it to work with AI you'll have to make sure they're in 'cargo' rather than the 'standing up' position, just a little quirk with how the AI works when under a players command, which doesn't happen when the leader is AI.

     

    I've put a fix in for the next version

  11. Posted

    19 hours ago, Jono the Cutest said:

    This is a question that I don't questioning anyone yet but here it is now.
    Turning speed of every static weapon or vehicle turret are very slowly. Stupid slowly. Does anyone has this experience before?

     

    Statics might need to be checked but tank turrets are accurate to the actual rotation speed.  I highly suggest binding/using keyboard keys for rotation it feels a lot less sluggish when you're not dragging your mouse out the door

  13. Posted

    On 4/24/2019 at 1:36 AM, Gunter Severloh said:

    Kju how goes the research into why the AI are driving tanks at full speed, any ideas yet?

     i was thinking, and im sure you guys do but do you guys have the previous version of the mod?

     

    I know that in the previous version not the latest hotfix that the tanks were driving normal, so something in the update changed the AI's behavior, i recall this as i built a mission

    that i released and it was playing how i wanted, after the update, the AI response changed, the tanks came sooner and kinda of ruined the atmosphere that worked previously.

     

    I was thinking if i may do some testing for you with the previous version, if there was a way to list what files (not new additions) but what files in IFA3_AIO_Lite that were updated.

    I like to compare notes with the previous working version with the new version, of the files, while using the changelog as a reference.

     

       My idea and i've done this in my WarMod series for locating bugs, or conflicts was to basically do a process of elimination where you take the previous version while

    using the changelog (v34 - 2019-03-31) and basically apply 1 file based on what was done to the old version in terms of updating that file to the current version, then test and see

    if it did something to the tanks in terms of the AI driving them fast.

     

    Is this possible?

    Or have you guys done this?

    As theres something in that massive list of changes and fixes that made the AI tank drivers now drive the tanks at full speed and then also make

    them drive in circles, like these guys had 2 cups more of coffee then they should, lol

     

    What does this mean or fix do btw?

    Fixed: standardized terrainCoef across all vehicles.

     

    we believe the tank driving issue has been fixed, 99% sure.

    • Like 1
    • Thanks 2

  14. Posted

    On 4/2/2019 at 3:13 AM, Gunter Severloh said:

    Will do kju, my theory on why they are flipping is they are driving at full speed and hitting obstacles, yet to see it.

    Another thought is reviewing the changelog from the latest update, these listings i wonder:

     

    Features

    • Changed: disabled anti flip running by default for now until improved. (Tanks are flipping on just about everything, or getting stuck, i seen a tank drive somehow hit and run over top a stone wall 2ft high and blow up)

    Terrains

    • Changed: defined base surface in order to standardize terrains surfaces. ( Whats this mean? is it possible that tanks have less resistance on terrain now enabling them to be able to drive faster or full speed?)

    Assets

    • Added: ejectDeadCargo = true to all tanks. (something i noticed is when a tank is about to blow up, crew bails, they run and right when the tank blows, crew somehow vanishes, think this has a bug)

    Tech details

    • Changed: removed non working allowDamage in anti flip system.

     

    EjectDeadCargo is a config parameter, all it does is if a unit is riding outside the tank and they're killed they'll fall off. Previously the animation was bugged and they'd just sit there 'undead'. What you're reporting isn't anything we're doing.

     

    On 4/1/2019 at 7:09 AM, Wiki said:

    Kudos for the update!

     

    One thing still  there though:

    when we switch the shell type in a tank, the "loading" green / red doesn't change, it stays green.


    This can't be fixed until the next arma version, limitation im afraid.

    • Thanks 1

  16. Posted

    13 hours ago, Bish91 said:

    Just watched the video all looks excellent! Do have a bit of a query regarding the static weapons though... with the 3CB style of static weapons the AI was nearly useless on static weapons after they had expended all their ammo, i.e they couldn't reload. Not sure if there was a module that disabled that for AI , but just wondering if an AI MG42 position becomes irrelevant after its fired off its belt and cant reload anymore? Just a bit concerned as I mainly play with and against AI. Keep up the good work though gents , I'm extremely excited! 

    Yeah we took this into account when setting up the system.

     

    When the weapon is placed in the editor it will have a bunch of extra magazines so the AI can use it, if the weapon is assembled from launchers though it starts empty.

    • Like 1

  22. Posted

    1 hour ago, road runner said:

    I have read the OP, all I've said is that no matter what changes BIS Devs make, there's always going be complaints, they're damned if they do, damned if they don't, :shrug:

     

    I would, if I was still using dev branch, however I never had issues with the splendid upgrade in the first place, to me it was a step in the right direction, to many guys like yourself, it's a step backwards. It's always going to be a case of you can please some of the people some of the time, you can't please all of the people all of the time :f:

    Dev branch broke too many scripts for me for what I was trying to achieve, however, I'm offering an observation that the devs are doomed to fail regardless of what they do to some members.

    As I also pointed out, there's very few games still offering updates 4 years post release.

     

     

    Baermitumlaut gave very constructive feedback, the only complaints I'm seeing is yours about his post.

    • Like 6

  25. Posted

    5 hours ago, shukari said:

    Nice mod from an paranoid author, this is my opinion.

    With this "rules" and "hidden" endMission sqfs you force more and more guys to rePbo your work to unblock it.

    Every mod cost days or years of development but i hate authors who are so limited in a so open minded and open source community like arma.

     

    I hope you change your mind a little bit. But thats only my two cents worth.

     

    Pretty much this.

     

    It's a very good way to encourage people to violate your license more

    • Like 1
×