Lagging when firing rockets!

[Dark_P]

Hi guys first i like to say that DCS World is awesome! but i have some issues...

 

I have searched but i don't find anything that fixes this, i did the Performance Guide, here: http://forums.eagle.ru/showthread.php?t=107551 but it did not help me...

 

My problem is that the game is lagging when i am firing rockets, i don't know the name of them but it is the rocket pods on the SU-25 that are firing in a stream.

 

The game just turns into a slideshow. The game runs fine otherwise.

 

My Specs:

 

i5 @ 3,20 GHz

8 GB ram

HD 7850 2GB

 

Anyone that knows how to fix this, maybe there are something i can change in the ini file or something, i would be grateful for any help!

 

Regards!

[Milli]

Hello Dark P.

 

Its a well known problem in the world of dcs world. Unfortunately i think we are stuck with it. Same sort of thing happens when a cluster bomb goes off. Bit of a slideshow ensues. Explosions are not handled well in dcs world. Hopefully a fix will come one day. There is a small explosions mod over in the mods section which cuts down on the size of the pops and bangs, it might help to some extent.

 

Regards,

Milli

[Dark_P]

OK Thank you Milli, then i know!, i will take a look at the mod then and se if it gets a little better.

 

Hope this issue will get fixed when we get DCS WWII: Europe 1944!

 

Cheers mate!

[stavka]

"Hope this issue will get fixed when we get DCS WWII: Europe 1944!"

 

I not have so much hopes, if the real intention was to fix this problem, they already does that since along time before redirect developers time to new projects. After dcs ww2 goes out, even worst because then they need to take care of the new product, the dcs world is to fade out in developing effords. As any other company with multiple projects, the dcs world they considered stabilized (because already goes out many patches) the majority of devs time is to go for the new project.

[_Dredd]

Yes an annoying issue with rocket salvos firing. It doesn't seem to be as bad with A-10C rocket salvos however, maybe it's the model or effects, maybe one can be merged into the SU-25 to reduce the pain.

[WildBillKelsoe]

Oh, don't forget, the Mustang also is a slideshow, with 6 fifties firing all at once....

[AtaliaA1]

Oh, don't forget, the Mustang also is a slideshow, with 6 fifties firing all at once....

 

 

Do you have the i5 also?

[Xerxes17]

Go into "Eagle Dynamics\DCS World\Bazar\Effects\PFX"

 

Then open the folders High and Medium

 

cut and paste the files:

 

Ground

Ground1

Ground2

smokecloud

 

into another folder "disabled".

 

This removes the particle effects that cause the slow-down. It reduces the eye candy, but you can still tell where your rounds are falling.

 

I posted this earlier this year, I'm surprised it's not in any of the big guides.

[tase]

there are actually 2 problems here:

 

1) the well known lagging caused by particle effects (disable using xerxes17's method above)

 

AND

 

2) the event log.

 

yes, those events that appear in debriefing ("player fired kh-25 ml, player hit outhouse", etc) are the cause of the slowdown when firing rockets.

 

to "fix" open "Scripts\UI\DebriefingEventsData.lua" in a text editor, search for the function "addEvent" and disable it (comment out the function body).

 

it should look like this afterwards:

 

function addEvent(event)
--[[    local result = getEventUseful_(event)
   
   if result then
       addInitiatorMissionId_(event)
       addTargetMissionId_(event)
       addWeaponId_(event)
       addEventType_(event)
       processHitEvent_(event)
       processPlayerChangeUnit_(event)
       processDeadOrCrashEvents_(event)
       
       result = not checkEventIsNextHitOrShot_(event)
       
       if result then
           table.insert(events_, event)
       end    
   end
   
   return result]]
   return true
end

 

some well meaning folk at ED wanted to reduce the debriefing spam when firing hundreds of munitions and devised some clever LUA code that collates all firing and hitting events into one (like "player fired rockets(200)/player hit outhouse(2)")

 

unfortunately, calling the LUA interpreter every event to do some pretty complicated data processing has a few side effects...

[AtaliaA1]

Sounds logical. And it does seem a bit redundant when put like that.

[Flagrum]

2) the event log.

(...) reduce the debriefing spam when firing hundreds of munitions and devised some clever LUA code that collates all firing and hitting events into one (like "player fired rockets(200)/player hit outhouse(2)")

There happens more in that function. If you disable it like that, your kills will not be registred and no score will be calculated. I suspect, the collating part is only marginally relevant for any performance hit here.

[tase]

No.

 

If you disable it like that all individual firing and hitting events will be dumped in the post briefing log.

 

Your score WILL be registered and calculated.

 

The collating part is VERY relevant to this performance issue.

 

But you don't have to believe my word, just TRY the damn thing and see for yourself.

 

Later Edit:

 

Actually you're right, I double-checked and the score isn't updated. So, just disable checkEventIsNextHitOrShot() which is the culprit for rocket lag, instead of the whole function:

 

 

 

function addEvent(event)
   local result = getEventUseful_(event)
   
   if result then
       addInitiatorMissionId_(event)
       addTargetMissionId_(event)
       addWeaponId_(event)
       addEventType_(event)
       processHitEvent_(event)
       processPlayerChangeUnit_(event)
       processDeadOrCrashEvents_(event)
       
--      result = not checkEventIsNextHitOrShot_(event)
       
--      if result then
           table.insert(events_, event)
--      end    
   end
   
   return result
end

 

It fixes the lag and doesn't break the scoring information.

Edited January 27, 2014 by tase
updated information

[Serp Supreme]

there are actually 2 problems here:

 

1) the well known lagging caused by particle effects (disable using xerxes17's method above)

 

AND

 

2) the event log.

 

yes, those events that appear in debriefing ("player fired kh-25 ml, player hit outhouse", etc) are the cause of the slowdown when firing rockets.

 

to "fix" open "Scripts\UI\DebriefingEventsData.lua" in a text editor, search for the function "addEvent" and disable it (comment out the function body).

 

it should look like this afterwards:

 

function addEvent(event)
--[[    local result = getEventUseful_(event)
   
   if result then
       addInitiatorMissionId_(event)
       addTargetMissionId_(event)
       addWeaponId_(event)
       addEventType_(event)
       processHitEvent_(event)
       processPlayerChangeUnit_(event)
       processDeadOrCrashEvents_(event)
       
       result = not checkEventIsNextHitOrShot_(event)
       
       if result then
           table.insert(events_, event)
       end    
   end
   
   return result]]
   return true
end

some well meaning folk at ED wanted to reduce the debriefing spam when firing hundreds of munitions and devised some clever LUA code that collates all firing and hitting events into one (like "player fired rockets(200)/player hit outhouse(2)")

 

unfortunately, calling the LUA interpreter every event to do some pretty complicated data processing has a few side effects...

I've tried this but can't seem to get it to work, it crashes my game whenever I go to launch it. Can you perhaps upload the .lua file itself to make things easier for me?

[tase]

are you on 1.2.7?

[Serp Supreme]

Yessir

[tase]

here you go:

DebriefingEventsData.lua

[Serp Supreme]

here you go:

 

Thanks man, works great.

[tase]

I don't know, but I simply do not experience any lagging during rockets usage or cannons or any explosions, whether on the ground or in the air.

 

Load up your SU-25(T) with 8 UB-32 rocket pods, select salvo mode (Ctrl-V) and start firing. Keep firing until "rounds complete".

[AtaliaA1]

Load up your SU-25(T) with 8 UB-32 rocket pods, select salvo mode (Ctrl-V) and start firing. Keep firing until "rounds complete".

 

 

He is most likely talking about the A10c.