OFPEC Forum

Addons & Mods Depot => OFP - Addons & Mods General => Topic started by: bdfy85 on 20 Aug 2005, 20:18:16

Title: infinite time to live shells ?
Post by: bdfy85 on 20 Aug 2005, 20:18:16
What config parameter I have to adjust in order to get ammo that won't disappear in the air after flying several (at most several dozens ) sec ?
I know this parameter
timeToLive = 1e10; // infinite time to live
But it doesn't seem to be worky at all:(
Any suggestions ?
Title: Re:infinite time to live shells ?
Post by: Planck on 20 Aug 2005, 21:04:00
Not sure about it at all.

Have you tried something like:

timeToLive=10000000000.0;


This is used for class Thing


Planck
Title: Re:infinite time to live shells ?
Post by: bdfy85 on 20 Aug 2005, 21:37:30
No. It doesn't work :(
I tried any values to this parameter - no influence :(

PS  1e10 = 10000000000 ;)
Title: Re:infinite time to live shells ?
Post by: Planck on 20 Aug 2005, 21:54:47
Quote
PS  1e10 = 10000000000

Yes I know, but, I was working on the assumption that the scientific notation might not be working properly in OFP....was just a thought.


I always presumed that the game engine dealt with ammo timetolive and it wasn't changeable anyway.


Planck  8)
Title: Re:infinite time to live shells ?
Post by: bdfy85 on 20 Aug 2005, 22:00:29
Russian team  investigate OFP ballistic ( you willbe amazed - but it uses real physic model ;) to make artillery mod or smth ;)

Due to this issue with shell's lifetime I can fire gun only with small angles ( < 15 degrees or so ) to hit the ground. Of course this affects fire  range  :(
Look at the picture
ingame projectile's trajectories and it lifetime. I can't even determine  regularity in shell disappearance! :(
Title: Re:infinite time to live shells ?
Post by: Planck on 20 Aug 2005, 22:06:46
Looking in the games config.....cfgAmmo.....the only thing with a timetolive value is the smokeshell.

This is why I presumed the engine handles all other lifetime values for the rest of the ammo........unless of course there is another parameter that is used.


Planck
Title: Re:infinite time to live shells ?
Post by: UNN on 30 Aug 2005, 20:11:50
Quote
Russian team  investigate OFP ballistic ( you willbe amazed - but it uses real physic model  to make artillery mod or smth

Well perhaps not amazed, I used one for my own Mortars:

http://www.ofpec.com/yabbse/index.php?board=43;action=display;threadid=23490 (http://www.ofpec.com/yabbse/index.php?board=43;action=display;threadid=23490)

Coc used a Nueral network in OFP. Now, that I did find amazing :)

But do you mean. How long it lives for (exists as an object) ?

Cant remember the exact values now, but it goes something like:

Bullets 15 seconds ?
Shells 23 seconds
Missiles 30 ?
Bombs 52 seconds ?

AFAIK These values are hardcoded and cant be changed with configs.
Title: Re:infinite time to live shells ?
Post by: h- on 03 Sep 2005, 17:53:01
Quote
Bullets 15 seconds ?
Not even close...
2-3 secs...
Why else we would have had amazingly difficult time with some parts of MCAR dev ;)

Other ones seem quite close, although I think it would be more like:
Bullets 2-3 secs
Missiles 12 secs
Shells 20 (or is it really 23?) secs
Bombs 32 secs

Can't remember any others than the bullets for sure though...
Title: Re:infinite time to live shells ?
Post by: UNN on 03 Sep 2005, 23:11:11
Quote
Shells 20 (or is it really 23?) secs

Looking back at some of my scripts, 20 seconds sounds right. But I'm sure bombs last longer.

Cheers
Title: Re:infinite time to live shells ?
Post by: bdfy85 on 14 Sep 2005, 23:16:01
HateR_Kint
UNN
Look at the picture I posted - ther's no definite time value for shell's time to live ;) Also different shells/missles lives for differenr t amount of time ;)
But anyway we have to use CoC way - setvelocity bombs , in this way they klive forever. But trajectory in this way ( uincluding CoC artillery ) is not ballistic  - it seems  just like in vacuum and has to be somehow corrected.
Title: Re:infinite time to live shells ?
Post by: Kyle Sarnik on 15 Sep 2005, 01:22:05
You can't set time to live to infinate anyways, EVERYTHING must have a limited lifetime.
Title: Re:infinite time to live shells ?
Post by: bdfy85 on 15 Sep 2005, 02:26:33
http://www.thechainofcommand.net/cgi-bin/ikonboard/ikonboard.cgi?s=5a32884242d2fcb87539bdb7ea390c87;act=ST;f=20;t=648;st=0;r=1;&#entry6299
Quote
1) OFP Projectiles have hardwired TTLs: shotShell = 20 seconds, shotBullet = 3 seconds, shotMissile = 10 seconds. None of these are acceptable to artillery.
2) BIS hacked their own system to get an object with a TTL of 120 seconds (=acceptable for artillery). You need a shotMissile with a thrustTime of 0. That's how they did the LaserGuidedBomb
3) But doing that means that you can't have the engine automatically fling the shot -- you have to do it manually, and write your own system for firing shells.
4) Ballistics is a tricky problem. I've figured out a couple ways to do it. In UA, we trained neural networks to do it: basically we decided it wasn't a problem worth solving, so we had someone write up a neural net, we fed it a bunch of data, and we generated a "black box" system. We give it the range, difference in altitude, and trajectory, and it spits out the elevation.
Title: Re:infinite time to live shells ?
Post by: UNN on 15 Sep 2005, 09:50:20
Quote
Look at the picture I posted - ther's no definite time value for shell's time to live


Well it's definite enough for BIS, CoC, MCar and myself (mine is ballistic) to use in game. But without knowing how you came across those values, there is not much I can say. Although this quote from Suma may explain why you get confliciting values?

http://www.flashpoint1985.com/cgi-bin/ikonboard311/ikonboard.cgi?s=db3645b33a926c66f69de59f56aaa25f;act=ST;f=7;t=47930 (http://www.flashpoint1985.com/cgi-bin/ikonboard311/ikonboard.cgi?s=db3645b33a926c66f69de59f56aaa25f;act=ST;f=7;t=47930)

Quote
Timing and profiling is always right, but only when set up properly.

There problems with your setup which make timing wrong:

You cannot use _time for profiling, as this gives you how much time elapsed in the game world, not in the real world, even less how much time was spent in the script itself. The game does not  execute the whole script "in one go". It executes several lines, then continues simulating and rendering, and goes back to the script in the next frame. The time spent you are measuring therefore covers mostly OFP engine simulation and rendering.

But thanks for the heads up on the timimg for bombs, I was sure it was considerably more than regular shells.
Title: Re:infinite time to live shells ?
Post by: bdfy85 on 15 Sep 2005, 12:04:25
Quote
1) OFP Projectiles have hardwired TTLs: shotShell = 20 seconds, shotBullet = 3 seconds, shotMissile = 10 seconds. None of these are acceptable to artillery.
I'm a bit unsure in this fact...
Enjoy the test mission I attached ( fwatch needed ) . I can see 2 sec for bullets , but no exact value for shells,  rockets... There can be a bit fault in my test mission (I start measuring time only after  to nearestobject command... it takes a bit of time ) But this small fault can't explain results...
Title: Re:infinite time to live shells ?
Post by: Dinger on 15 Sep 2005, 14:56:47
Hi -- okay, I'll confuse things further by putting differents posts here and in the CoC forums.

Here:

This mission (also uses EHs) will give you a TTL measurement that is roughly accurate enough for you to see the actual settings. Just set your unit type, fire a round, and wait for it to timeout. repeat if necessary.

Title: Re:infinite time to live shells ?
Post by: UNN on 16 Sep 2005, 11:55:44
Using a slight variation to Dinger's example mission, that allows the shell to follow it's default path. I get exactly the same results. Shells inparticular seem to be consistent. Bullets can vary a bit more, but they are always within the times posted above, give or take 0.002 seconds.

I think FWatch will cause the problems you have seen? If you want to grap the XYZ position of rounds to use in a graph, output the results to a dialog in OFP. Then you can just cut and paste them into the likes of Excel.
Title: Re:infinite time to live shells ?
Post by: h- on 16 Sep 2005, 13:25:40
I concur with UNN.

I have run (in early MCAR dev) a lot of tests with these and shells etc. have steady lifetimes (as steady as OFP happens to be at the given moment)...

Anyway, the LGB thingy was news to me... :)
Title: Re:infinite time to live shells ?
Post by: bdfy85 on 16 Sep 2005, 19:19:59
First - You are right - time values are constant. Seems like problem with my dome was - loop time value ( OFP engine can't process corectly small  time space (like 0.01 I used )
Full post is at CoC forum.