[Logged] Design Question Regarding Weapon Firing/Sequencing

Post bug reports and ask for game support here.

Moderator: MOD_Command

Post Reply
DWReese
Posts: 1879
Joined: Fri Mar 21, 2014 11:40 am
Location: Miami, Florida

[Logged] Design Question Regarding Weapon Firing/Sequencing

Post by DWReese »

This is an observation. It is not a criticism. It may be helpful to the devs to have this info, if they aren't aware of it, or haven't thought about this topic recently.

Often, game players wonder WHY a unit won't fire. Usually it can be traced back to some legit reason.

Sometimes, however, something appears strange. A game player may be able to manually fire a weapon, but the AI-controlled side will not fire, even though a quick look at their weapons through the Editor reveals that they are GREEN and ready to go. So, I asked myself, why? Why would a game player be able to shoot, but not the AI?

I researched this further, and found that often the AI-controlled unit WOULD eventually fire, but it was delayed and not nearly as fast as a human could, even when both were GREEN. So, it occurred to me that the programming must go through cycles, and those cycles are probably time-related. When a certain time on the clock occurs (probably at the top of each minute) the firing sequence (or detection sequence, or any other sequence) is evaluated. If the criteria is met, then the weapons are fired by the AI. If not, then they aren't. At most, this would probably delay normal weapons from being fired by up to one minute slower than the human player.

But, here is the icing. For many months I have tried to figure out why some ABM SAMs would be fired by the AI-controlled side, and then others wouldn't. Even with v21, the missiles can often be fired manually, but the AI-controlled unit doesn't always do it.

Then it occurred to me, if a player monitors it, and fires as soon as he can, then it works. A few minutes (maybe seconds) later (depending on the missile) the target now no longer appears within the DLZ, so the SAM can't fire.

But, what about the AI-controlled unit? Well, if the AI side has to wait for a sequence to roll around, then because the incoming ballistic missile is travelling so fast, it could literally enter the DLZ and then EXIT the DLZ before the sequencing window comes around. Therefore, the missile would never be AI-fired (even though it was GREN) if that was the only opportunity within the DLZ.

This would explain why it fires sometimes, and other times it doesn't. There may not have been much wrong with the missile logic. It may have been a sequencing programming issue that was having trouble dealing with the speed of the incoming missile.

That sequencing module has worked great for normal missiles for years, but when you are now dealing with supersonic missiles, or ballistic missiles, the DLZ isn't open very long, and if you don't catch it just right, then the SAM will never fire at them.

This is just a theory. It may be completely wrong. But, if it's correct, or partially correct, then a minor tweak or so, and this just might solve any or all of these mysteries.

BTW, v21 is much better, but there are still some of these sequencing issues that I cited above. If this is true, this might be the cause. Obviously, I don't have access to the source code so if this isn't how you have programmed the game, then disregard this message. I just offer this as an observation. Perhaps it can be helpful.

Anyway, It's just a thought. Thanks for taking the time to read this. Hopefully, it can be of some use to you.

Doug
User avatar
ultradave
Posts: 1368
Joined: Tue Aug 20, 2013 7:01 pm

RE: Design Question Regarding Weapon Firing/Sequencing

Post by ultradave »

Good stuff puzzling this theory out. I'll be really interested to see what the developers comments on this are. It does make logical sense.

Dave
----------------
Dave A.
"When the Boogeyman goes to sleep he checks his closet for paratroopers"
thewood1
Posts: 6835
Joined: Sun Nov 27, 2005 6:24 pm

RE: Design Question Regarding Weapon Firing/Sequencing

Post by thewood1 »

I think there is some possible truth to that hypothesis. I actually think it was discussed years ago when there were some issues with units not firing during high time accelerations. That was back in CMNAO when we had the high fidelity option for game execution. Either way, its a good deduction on ABM firing sequences.
You are like puss filled boil on nice of ass of bikini model. You are nasty to everybody but then try to sweeten things up with a nice post somewhere else. That's nice but you're still a boil on a beautiful thing! - BDukes
Rain08
Posts: 93
Joined: Mon Jul 18, 2016 9:54 am

RE: Design Question Regarding Weapon Firing/Sequencing

Post by Rain08 »

I'll take a guess and say what if the Targeting sequence of the OODA cycle is doubled? Can you try testing units with varying proficiencies (or even just Novice vs Ace)?
thewood1
Posts: 6835
Joined: Sun Nov 27, 2005 6:24 pm

RE: Design Question Regarding Weapon Firing/Sequencing

Post by thewood1 »

I did a test a couple weeks ago with OODA and it has an impact on when the batteries start firing, but once engaged, it means nothing. Not sure if it plays into the OP's post. The weirdness I saw was a THAAD battery with six launchers and a WRA set for max fire only launching from the first two launchers until they were exhausted. That meant only a max of six missiles ever made it into the air, vs. eight guidance channels. It was only after the first launcher exhausted it missiles that the third launcher started firing, and so on until all launchers were exhausted. Because of ROF, that type of sequence takes a lot longer than if a missile was fired from each launcher. That issue might be mixed in here as well.

And just to reiterate the OP's point, in most engagements, that firing sequence issue doesn't have any meaningful impact. But in high-speed ABM engagements, the engagement window is so small, it has an impact.
You are like puss filled boil on nice of ass of bikini model. You are nasty to everybody but then try to sweeten things up with a nice post somewhere else. That's nice but you're still a boil on a beautiful thing! - BDukes
DWReese
Posts: 1879
Joined: Fri Mar 21, 2014 11:40 am
Location: Miami, Florida

RE: Design Question Regarding Weapon Firing/Sequencing

Post by DWReese »

I don't believe that the OODA restrictions have anything to do with it. As I said, it's already GREEN and can be fired.

Doug
guanotwozero
Posts: 651
Joined: Fri Dec 27, 2013 1:53 am

RE: Design Question Regarding Weapon Firing/Sequencing

Post by guanotwozero »

This is indeed an interesting hypothesis. If you think about it, the vast majority of weapons & aircraft travel at speeds where CMO simulates their interactions well, i.e. the necessary approximations are within acceptable limits to produce an intended performance. It could be that the high speeds involved with RVs and ABMs mean that those standard methods can't handle the interactions quite as well.

Think of polynomial graphs being approximated by many small sections, like a histogram - if they represent velocity, location, etc, those slices are small enough to model their interactions. However if very high velocities are involved then maybe those slices are too large (wrt time) and the intended interactions are more prone to error. Think of a graph with many peaks and troughs being approximated by slices too large to do justice to the actual shape. Which means that maybe an RV's next location computation 'misses' the DLZ because the computation 'time slice' is too large.

Think also how the game is divided up into 1-second 'ticks'; to be able to run on anything other than IBM's latest chess computer, the game processing can't all be done within a single tick. While we players have no specific knowledge of what happens under the hood, I'll speculate much of the game processing is done as threads/tasks running parallel to the main game clock. In other words, there are ongoing tasks that are sequenced/processed over many such ticks, and some processes may only come around every so often. Normally that 'so often' is fine for calculating game interactions, but for high-velocity objects it may be too infrequently to handle them as intended. It might also be the case that CPU type/speed could make a difference, where more effective thread processing (more calcs per tick) is handled better by more powerful CPUs. Please note that's all speculation on my part!
DWReese
Posts: 1879
Joined: Fri Mar 21, 2014 11:40 am
Location: Miami, Florida

RE: Design Question Regarding Weapon Firing/Sequencing

Post by DWReese »

I agree with you agreeing with me. <lol>

Maybe we are correct, and maybe then again, maybe it has nothing to do with this at all. But, it just might give the devs something to think about when trying to solve these anomalies.

I'm fairly certain that the speed of these new weapons, and the firing opportunity window are much different than the logic that was needed to calculate weapon firings from before. The programming may need to be tweaked to handle these new, fast moving weapons with smaller firing/detection/reaction segments.

Doug
guanotwozero
Posts: 651
Joined: Fri Dec 27, 2013 1:53 am

RE: Design Question Regarding Weapon Firing/Sequencing

Post by guanotwozero »

Yep, if that's the case, then for an acceptable fix those of us who play on wimpy laptops may just have to accept game slowdowns while RVs/ABMs are in flight!
DWReese
Posts: 1879
Joined: Fri Mar 21, 2014 11:40 am
Location: Miami, Florida

RE: Design Question Regarding Weapon Firing/Sequencing

Post by DWReese »

I guess that we will have to wait and see.

Perhaps the windows of opportunity can be adjusted to allow for a quicker response. Of course, this is all speculation as we don't really know what the cause is at this point. It would make sense, though. There has to be a point where the game says, okay, time to check and see what can see what, and what can shoot at what. They couldn't possibly have that occurring every second, or every 5 seconds, so it could be anything after that. It would seem as though setting it at the top of the minute would be logical, since Missions are sort of set up to work like that as well. That way, it could be 59 seconds at the longest. But, with incoming ballistic missiles, you may never get a window that long, unless the missile is intercepted far away like in the OVERSHOOT scenario that I provided. So, it may be time for the game devs to update how fast a unit can detect and fire at things, assuming that this is the issue at hand.

Things traveling at 6500 or 10000 kts are much faster than a Harpoon traveling at 500. <G> So, what works well with the older weapons may nit work as well with the faster stuff, and the window is MUCH smaller.

Doug

Doug
WSBot
Posts: 181
Joined: Sun Jan 17, 2021 9:31 pm

RE: Design Question Regarding Weapon Firing/Sequencing

Post by WSBot »

0014519
Post Reply

Return to “Tech Support”