Units spawned by the main AO

[Stanhope]

Hello everyone,

Recently we've noticed how on malden only 1 AO got completed in the 90 hours the server was up.  This sparked me to investigate whether or not it'd be possible to make the amount of units that spawn in an AO dependant on the amount of boots on the ground.  Let me explain in a bit more detail:

Let's say that there are 20 people on the server, of which 5 pilots.  When a new AO spawns the amount of units in the AO will be tweaked for 15 players instead of for 60 players.
Note that this tweaking only happens when an AO spawns.  If suddenly 40 more people join the server no additional units will be spawned in.  The AO will be left to be completed by more people than it was intended for.  Same is true the other way, if of those 15 people suddenly 10 leave the AO will be left as is.  No units will despawn.

How many units will spawn for how many players is still being discussed and will in all likelihood be subject to change after some real life testing.  

 

Let me give you a list of the pro's this method has over what we're doing now: (Now we're spawning a fixed amount of units regardless of the amount of players)

• It will take the AO significantly less time to spawn when there is a lower player count.
• Server performance will be better when there is a lower player count.
• An AO can be completed within an hour, even with low player counts.
• Enemy presence can be better tweaked to the amount of players on the server.
• Arma's engine does not have a garbage collector.  Meaning that overtime the RAM gets clogged with dead code, remnants of previously created missions/units.  By running less code (by reducing the amount of spawned-in units) this process will slow down.  Additionally improving server FPS.
• It'll make AO difficulty more consistent, as there will be the same amount of units to kill per player.  More people on the server won't make the mission easier anymore.

 

Possible downsides:

• It'll require some additional code to run, possibly lengthening AO spawn time.  However the effects described above should vastly outweigh this.  And the code to check for player count has been written with performance in mind.
• ...

 

So now we would like your opinion?  Do you want to change over to a system where the amount of spawned units is dependent on the amount of player or do you want to stay with the old system?

 

Stan

[TheScar]

Well,both ways have their up´s and down´s.

To me,mission/server performance is the no.1 priority as i want the mission to run stable as long as possible.

Currently,i´m still convinced the 4 FOBs + main base do have a large part in taking performance - but as heard in the meeting there will be mission versions with 1 main base only for test purposes which should drstically enlarge performance and it also eases up (regular) mission restarts as none will be bitching about loosing FOB(s).

 

On topic again,for performance issues my vote goes to fixed numbers of AI@AO.

It´ll be harder for low player crowds to conquer a AO but thats why we have side missions to go for.

[Stanhope]

You know that the fixed numbers have less performance than the variable numbers right?

[TheScar]

Nu!

[Tales]

I'd vote for the variable; I regularly tend to login on low headcount times due to work and it can feel a bit like one man versus the world when you've an AO full of enemies designed for 60. Feeling like you're grinding away against massive odds can suck.

 

Hopefully we'll see more AO's than the usual ones too if this occurs as they will cycle through quicker.

 

(I'm interested in the idea Scar mentioned above too, main base only but more central to avoid driving eons would be great)

 

 

[ansin11]

You might also want to decrease the AO diameter with less units.

[Ryko]

With a main, priority and side mission, that's three separate corps of AI sucking up CPU processing power. If the goal is to increase server FPS then every AI group that is removed will have a positive impact. The benefit for a variable / balanced algorithm for creating AI based on the amount of players is that for lower-player games, you'll create lower numbers of AI using a lesser amount of resources.

 

The problem is that if more players arrive, there will now be more players than anticipated for the mission and it will become easier ("why aren't there more enemy soldiers to fight? This is easy"). Imagine the server starts, 10 players join, 25 enemy soldiers are spawned in the first AO. Over the next 10 minutes another 50 players join, now you have 60 players fighting 25 soldiers.  The AO is quickly secured. Then the next mission is spawned and maybe it creates 100 enemy soldiers to fight. Players leave (maybe it's getting late and people are starting to go to bed, or maybe it's too hard) and now you've got 20 players fighting 100 enemy soldiers.

 

The solution would be either an ambient AI system that dynamically spawns new enemy groups in response to players out in the battlefield (we use this on AWE and it does fill in the gaps nicely, and makes a less static mission) and/or a system that over time adjusts the number of enemy groups upwards or downwards in response to overall player count (maybe the enemy gets reinforcement groups if there is a sudden increase in player count, or enemy groups are called away as the player threat is not enough to justify the size of the enemy presence).

 

One particular problem with Arma is spawning units in places that make sense: finding the right number of enemy units to challenge a certain number of players is tricky by itself, but getting that number of units into positions where they can threaten the player is another matter entirely. If you have five groups on patrol around an objective, there's no guarantee the players will even see them because they might be out of position.  So tweaking the number of units upwards kind of addresses that issue... until suddenly all the enemy units are in position at the same time, and the players encounter 8 groups all at once.

 

It's not an easy fix, but for performance reasons alone, I think it's worth pursuing.

[Guest]

18 hours ago, Ryko said:

Imagine the server starts, 10 players join, 25 enemy soldiers are spawned in the first AO.

Fair enough, but then how are server restarts going to be handled? Normally on server boot up there are 15-20 people on when mission starts. Then the flood of players join another 5 minutes later. Now you have the problem of 25 v 50. Or what happens, since restart is now scheduled, when it is restarted and there is 0 players online? Does it default to if 1 players is on? What does it default to? If it goes back to acting like there are 60 players on then the issue of a full AO still remains. Also, will having this script not also fully extend the time that it takes for a new main AO to spawn in? Now that it has to calculate the number of players that are on the server will this not increase the delay? There are times when delay is quite long due to multiple scripts running at once. Yes this might be slightly negated due to the fixed daily restarts but it will still be a factor. 

 

Lastly, is it 100% needed? What is the difference between one AO and the next? The scenery? The terrain? What makes it so vital to increase the time of one AO? Are the units spawned completely different? No. All that this will do is decrease time to get FOBs, which appears that they might possibly be getting removed in the near future. I just don't understand the need to go from shooting things in one area for a prolonged time, to doing it in a different area for a shorter time. And as one other person has pointed out to you, Malden server is very quiet. There is a very low player count on Malden. Yes it may say there are 7 people online in one day but for how long? Those 7 people may have joined for one minute then left.

 

Just another opinion.

[Stanhope]

4 minutes ago, McKillen said:

how are server restarts going to be handled? Normally on server boot up there are 15-20 people on when mission starts. Then the flood of players join another 5 minutes later. Now you have the problem of 25 v 50. 

Server restarts will be handled like a normal AO spawning in.  The amount of people on the server will be counted and the AO spawned in with an amount of units based upon that number.  Do note that the script that starts the AO is only executed after everybody initially pushed through.

 

5 minutes ago, McKillen said:

Or what happens, since restart is now scheduled, when it is restarted and there is 0 players online? Does it default to if 1 players is on? What does it default to? If it goes back to acting like there are 60 players on then the issue of a full AO still remains.

Currently it's set up to spawn enemy units incremental to the amount of people on the server.  If there are between 0 and 10 people on x amount of units are spawned.  If there's between 10 and 20 Y amount of units are spawned and so on and so forth.

 

7 minutes ago, McKillen said:

Also, will having this script not also fully extend the time that it takes for a new main AO to spawn in? Now that it has to calculate the number of players that are on the server will this not increase the delay? There are times when delay is quite long due to multiple scripts running at once. Yes this might be slightly negated due to the fixed daily restarts but it will still be a factor. 

The "script" that will accomplish this is:
 1 line of code that reads an array (actually counts the amount of elements of this array to be exact) that is defined by arma regardless of the execution of this script.
1 switch case statement that defines 6 variables.  
 

The estimated amount of time this "script" will take is less than the time it takes the mission to spawn 1 enemy vehicle.

 

Also we mostly see longer spawn times of AOs when the server has been up for more than 24 hours.  The server will as of next update restart itself at 5 (if i'm not mistaken) in the morning.  Meaning that at its busiest time (between 5 and 10 in the afternoon/evening) it will have been up for 12 to 17 hours.  Which is a full 8 hours short of when the first signs of a long uptime show.  

 

11 minutes ago, McKillen said:

Lastly, is it 100% needed?

In the grand scheme of things, is anything necessary?  

 

12 minutes ago, McKillen said:

What is the difference between one AO and the next?

The difference will be the amount of units in the AO.

 

12 minutes ago, McKillen said:

What makes it so vital to increase the time of one AO?

I don't exactly know what you mean here.  Nothing is really vital.  And this change, if implemented, will decrease the amount of time needed to complete an AO with a lower player count and leave it mostly unchanged at peak times.  

 

13 minutes ago, McKillen said:

Are the units spawned completely different?

The actual spawning in of units will remain unchanged.  The only thing that's changed is the amount of units spawned.

 

14 minutes ago, McKillen said:

All that this will do is decrease time to get FOBs, which appears that they might possibly be getting removed in the near future. I just don't understand the need to go from shooting things in one area for a prolonged time, to doing it in a different area for a shorter time.

What's your definition of a successful night on EU1?  I'm guessing that for most people having completed at least 1 AO is a part of a successful night, it is for me at least.  Now on EU1 that's not really a problem, there are enough people to complete AOs.  On EU2 however completing an AO is a real issue.  Most nights not a single AO is completed.  That leads to people leaving the server with a feeling like they didn't achieve anything.  And now that's not a nice feeling.  If one gets it too often it isn't unthinkable that one might do whatever it takes to stop that feeling from reoccurring.  A very easy way to do this is not join the server that's making you feel this way.  

What the script does is make I&A playable with lower player counts.  Hopefully keeping those people on long enough to get other people to join.  It is basically an effort to repopulate EU2.
 

18 minutes ago, McKillen said:

And as one other person has pointed out to you, Malden server is very quiet. There is a very low player count on Malden. Yes it may say there are 7 people online in one day but for how long? Those 7 people may have joined for one minute then left.

 You're absolutely right.  Does this mean we can't provide services or make tweaks to the mission for these few people?  Can we not try to do something to get these numbers up?  To keep people on the server longer?  To get more people to join the server?

[Xwatt]

voted do not care as I dont mind either  

[Guest]

5 minutes ago, Stanhope said:

In the grand scheme of things, is anything necessary?  

wut? Bit off topic but okkk.... I dont get what you mean by this?

 

7 minutes ago, Stanhope said:

What's your definition of a successful night on EU1?

For me, I don't care if I even don't make it to an AO. A fun night for me can be me getting my arse kicked by AI for an hour flat, not being able to move most of the time because if I do I die. Its not about how much I do, its about the experience. Nothing can beat the feeling of being chewed up and spat out by an AO, it is fun because you realize you got your ass handed to you by a robot, then you get to think again. 

 

11 minutes ago, Stanhope said:

The actual spawning in of units will remain unchanged.  The only thing that's changed is the amount of units spawned.

In a way to justify this a bit more why not change up AOs a bit more. If this is done then I can get on board more with the idea. At the minute, which Im not sure if I got across clearly enough but will try again; every AO is the same. This means the same units, the same task etc. The ONLY difference is the area. Practically a carbon copy. Why not add variation. An example would be for in a town, not too have multiple T-100s but rather more MRAPs or maybe toy with making a mechanized infantry unit for OPFOR. Make it more realistic. This would make each AO unique to itself. It could also stop a lot of people sitting on hills with titans (cough) as there may be very little to no armor. But then the next has multiple armored units on a glass plain. It would also keep excitement going as no two AOs are the same which is what it feels like. 

 

20 minutes ago, Stanhope said:

What is the difference between one AO and the next?

Sorry, meant currently.

 

21 minutes ago, Stanhope said:

I don't exactly know what you mean here.  Nothing is really vital.

Well considering what has been said in the past, it seems vital to you. I just don't get why it is so needed. I personally haven't heard many complaints of this issue. Yes, you shouldn't wait for complaints to fix or improve something, but on the counter, don't fix something that is working well. In the past on my times on the server, I didn't hear one complaint other than a spartan moaning about it.

 

I personally feel over the past two times we have argued about this you have yet to answer my question. What does it change? Not in terms of mechanics. In terms of gameplay. To make you understand my point read this: 

In the near future, from what has been said it appears you are debating removing FOBs, which many people support. If this is to happen then AOs have no huge significance other than a place to fire your weapons at the bad guys. That is fine. But what makes it that it needs to be shorter? The core gameplay is the exact same. The players will still point their guns, and fire. With or without this change. The only change it makes is that the specified area shifts slightly to one side. I just don't see the point in it, but I guess you will still implement it regardless which is fair... I guess.

[Stanhope]

12 minutes ago, McKillen said:

What does it change? Not in terms of mechanics. In terms of gameplay.

It allows a smaller player base to also effectively complete AOs in less than a few hours per AO.

 

13 minutes ago, McKillen said:

In a way to justify this a bit more why not change up AOs a bit more. If this is done then I can get on board more with the idea. At the minute, which Im not sure if I got across clearly enough but will try again; every AO is the same. This means the same units, the same task etc. The ONLY difference is the area. Practically a carbon copy. Why not add variation. An example would be for in a town, not too have multiple T-100s but rather more MRAPs or maybe toy with making a mechanized infantry unit for OPFOR. Make it more realistic. This would make each AO unique to itself. It could also stop a lot of people sitting on hills with titans (cough) as there may be very little to no armor. But then the next has multiple armored units on a glass plain. It would also keep excitement going as no two AOs are the same which is what it feels like. 

Except that the sub objective is always different, different houses and different amount of houses will be garrisoned, different units will spawn because of the randomization that is currently already in the script and that will remain there.  
I already added selfmade units in the form of static weapons strapped to trucks.  Both regular spawns and as a sub objective.
Nobody complained to me about there not being more variety in the units spawned before I added them.  And the AOs were working perfectly fine.  Yet I changed something and people liked it.

There are multiple reasons why you'd change something that's working and nobody is complaining about.  For example you know something has more potential or people just don't know that something else is possible hence they don't complain.

 

17 minutes ago, McKillen said:

In the near future, from what has been said it appears you are debating removing FOBs, which many people support.

I'm not actively participating in that debate.  I only offer my technical knowledge of SQF in that debate.  I won't be the primary dev designing it, at best I'll be actively helping out with the development of it.  

 

19 minutes ago, McKillen said:

That is fine. But what makes it that it needs to be shorter? The core gameplay is the exact same. The players will still point their guns, and fire. With or without this change. The only change it makes is that the specified area shifts slightly to one side. I just don't see the point in it, but I guess you will still implement it regardless which is fair... I guess.

Even better, when there are 60 players on practically nothing will have changed.  The time it took the AO to spawn will have ever so slightly lengthened, that's it.  But when you've got a server with a notoriously low population like EU2 everything will have changed.  You'll now have AOs that are completable within an hour even if you've got only 5-10 people on.  This will hopefully encourage people to actually stay on the server, attracting more people in the process.  

 

Does everything we do to I&A have to be only directed at EU1 and for conditions of 60 players?  Because this is change will among other things also hopefully bring more people onto EU2.

[Ryko]

1 hour ago, McKillen said:

Fair enough, but then how are server restarts going to be handled? Normally on server boot up there are 15-20 people on when mission starts. Then the flood of players join another 5 minutes later. Now you have the problem of 25 v 50. Or what happens, since restart is now scheduled, when it is restarted and there is 0 players online? Does it default to if 1 players is on? What does it default to? If it goes back to acting like there are 60 players on then the issue of a full AO still remains. Also, will having this script not also fully extend the time that it takes for a new main AO to spawn in? Now that it has to calculate the number of players that are on the server will this not increase the delay? There are times when delay is quite long due to multiple scripts running at once. Yes this might be slightly negated due to the fixed daily restarts but it will still be a factor. 

 

Lastly, is it 100% needed? What is the difference between one AO and the next? The scenery? The terrain? What makes it so vital to increase the time of one AO? Are the units spawned completely different? No. All that this will do is decrease time to get FOBs, which appears that they might possibly be getting removed in the near future. I just don't understand the need to go from shooting things in one area for a prolonged time, to doing it in a different area for a shorter time. And as one other person has pointed out to you, Malden server is very quiet. There is a very low player count on Malden. Yes it may say there are 7 people online in one day but for how long? Those 7 people may have joined for one minute then left.

 

Just another opinion.

 

How would you propose this be handled?

[Guest]

1 minute ago, Ryko said:

How would you propose this be handled?

Well as Stan just said, this is more largely focused on EU2. With the new 77th format coming to 2 I don't see any need for it at all. In my opinion the sooner the 'Project whatever it was' gets done the better. This will go a bit off topic but I will bring it back. Essentially when the format gets created, we will have a tiering system. Chaos -> Organized mayhem -> 'The Elite'. And no I can't believe I just called EU3 the elite either. But this way, people will get a taste of more organised gameplay and will hunger for more, leading them to join 3 and 6 and stop the whole problem of them dying, which many people have said.

 

My proposal is to leave it. Others have agreed it isn't a huge problem. In my eyes it does not need handled. Players go to the same location, to do the same thing, the same gameplay regardless of the number of EI there is. All this does is increase the number of times the big circle moves.

[Stanhope]

6 minutes ago, McKillen said:

My proposal is to leave it. Others have agreed it isn't a huge problem. In my eyes it does not need handled. Players go to the same location, to do the same thing, the same gameplay regardless of the number of EI there is. All this does is increase the number of times the big circle moves.

I don't agree with your conclusion but is it a problem that it isn't a huge issue we're tackling?  Does it matter to you how many times the AOs circle moves?  Does it matter to you that we're dealing with an issue that in your eyes doesn't need handling?  

 

If you're wondering, I've already written the code so it's not like i'm gonna be wasting my time on it.

[Ryko]

1 hour ago, McKillen said:

Lastly, is it 100% needed? What is the difference between one AO and the next?

I guess I don't understand this comment. Are you proposing we just have a continuous battle in one AO, with new enemy units continuously spawning in?

[Guest]

Just now, Ryko said:

I guess I don't understand this comment. Are you proposing we just have a continuous battle in one AO, with new enemy units continuously spawning in?

No, but the need to accelerate an AO seems pointless to me since they are similar

[Guest]

13 minutes ago, Stanhope said:

I don't agree with your conclusion but is it a problem that it isn't a huge issue we're tackling?  Does it matter to you how many times the AOs circle moves?  Does it matter to you that we're dealing with an issue that in your eyes doesn't need handling?

When there are larger issues at hand, yes it is. I don't really care what you do with your time. But there are other things that could be getting worked on, bigger problems that every time they get brought up 'Its too much to do'. You know there are larger issues and a lot more work to be done, hence why you need more devs I assume. But alas here we are getting dragged off topic again...

 

13 minutes ago, Stanhope said:

If you're wondering, I've already written the code so it's not like i'm gonna be wasting my time on it.

I know you have. Honestly this seems a little uncalled for and rude but w/e I see you won't properly listen to me so fair enough. I'm sure all things that get added into I&A get accepted well. But I feel an increase in FPS will get accepted more strongly than a minor change, that is why Im complaining strongly. There are bigger fish in the sea yet you choose to stick to the shallows and ignore them. This clogging up the forums though, but if any of you actually care enough to follow this up let me know, but I doubt you will

[Stanhope]

2 minutes ago, McKillen said:

When there are larger issues at hand, yes it is. I don't really care what you do with your time. But there are other things that could be getting worked on, bigger problems that every time they get brought up 'Its too much to do'.

3 minutes ago, McKillen said:

Honestly this seems a little uncalled for and rude but w/e I see you won't properly listen to me so fair enough. I'm sure all things that get added into I&A get accepted well. But I feel an increase in FPS will get accepted more strongly than a minor change, that is why Im complaining strongly. There are bigger fish in the sea yet you choose to stick to the shallows and ignore them. This clogging up the forums though, but if any of you actually care enough to follow this up let me know, but I doubt you will

By all means make a separate topic discussing these bigger things and things that can increase FPS that I'm not already doing.  

 

3 minutes ago, McKillen said:

You know there are larger issues and a lot more work to be done, hence why you need more devs I assume.

The need for devs hasn't dramastically increased but we thought it was about time we give people a clear set of requirements and what they need to do to sign up.  

 

5 minutes ago, McKillen said:

But alas here we are getting dragged off topic again...

Yes, yes we are.  Let's do this: if you want to continue to talk about any of the issues you've raised that don't directly relate to this, make a separate topic for them.

[Ryko]

4 minutes ago, McKillen said:

But I feel an increase in FPS will get accepted more strongly

Scaling enemy units so that there is an appropriate number of units compared with the number of players will increase FPS, so I don't understand your premise.  Did you read my earlier post about this issue?

 

6 minutes ago, McKillen said:

No, but the need to accelerate an AO

If I am understanding this comment, you feel that players are just fine with the idea that it may take a long time to win an AO because there are many more enemies than the current number of players can effectively handle, and that our proposition to scale the number of EI to balance with the player base, in order to complete AOs, is unnecessary. Correct me if I'm wrong.

 

I don't think there's evidence to support this, however, there's no evidence to support the other argument, too.  I would submit that there is a natural tendency to want to win an AO, and the current reward system also requires an AO to be completed for additional vehicles to spawn. So I think there are plenty of good reasons to want to make AOs "winnable" from a player perspective.

 

8 minutes ago, McKillen said:

You know there are larger issues and a lot more work to be done

Please itemize.

 

8 minutes ago, McKillen said:

There are bigger fish in the sea yet you choose to stick to the shallows and ignore them.

Itemize, and stop dragging this down to a personal level.

[Mad Eye Moody]

Itemize, and stop dragging this down to a personal level.

To reply to the OP I think it’s a brilliant idea to have proportional EI to players. As a player with lots of spare time I often play on Malden and am the only or at most 1 of 5 players. I use the space to practice flying, loadouts etc. It would be nice to have fewer enemy.


Sent from my iPad using Tapatalk

[Stanhope]

13 minutes ago, Mad Eye Moody said:

To reply to the OP I think it’s a brilliant idea to have proportional EI to players. As a player with lots of spare time I often play on Malden and am the only or at most 1 of 5 players. I use the space to practice flying, loadouts etc. It would be nice to have fewer enemy.

As the poll at the top suggest this got passed, it's been implemented in I&A 3.3.7 and has been live on the server for nearly a week now.

[Mad Eye Moody]

As the poll at the top suggest this got passed, it's been implemented in I&A 3.3.7 and has been live on the server for nearly a week now.

I thought so when I played last week.


Sent from my iPad using Tapatalk

[Mad Eye Moody]

I like the new feature but can I make a request. If I transport myself to the AO and leave the vehicle in a safe place, when I am more than a certain distance away it respawns back to base leaving me with no way out except to respawn. Is it possible to limit that when the server is sparsely populated?


Sent from my iPad using Tapatalk