Dynamic Campaign Discussion Thread

[doveman]

Admittedly I don't know how many AI DCS can currently cope with on an average PC but perhaps there are ways to increase that so that a bubble system isn't needed.

 

For example, ArmA3 is a single-threaded program like DCS that can only use one core for AI but it has something called 'headless client' which runs on separate cores to the server but is synced to it and offloads some AI, allowing for more AI than the server could otherwise support. Admittedly I find it all a bit convoluted and don't know why they can't just make it run this extra process and offload some of the AI to it automatically in both SP and MP (unless a command-line switch is used to disable it in cases where the server doesn't actually have any spare cores), rather than it having to be run manually and missions designed specifically to use it (that's actually not complicated and just involves setting a couple of parameters but it still means that missions that don't have those set won't make use of it) but it's better than nothing.

 

So perhaps DCS could do something similar, in SP running a background program that syncs to the main program and runs on a spare core not used by the main program to enable more AI. I think DCS only uses 2 cores at the moment, so on a quad-core system there could perhaps be two instances of this background program running on each of the unused cores to enable even more AI.

 

Likewise for MP, where the server has 4 cores available the same thing could be done. It might be more complicated in MP, as in SP it doesn't matter if the AI in areas where the single player isn't present is slightly out of sync with the main process as the player won't notice it or be affected but ArmA 3's AI running on the 'headless client' has to sync properly with the main server process and the clients, as players can be spread over the map and encounter AI running on either the main server process or the headless client one, so it doesn't seem to be impossible.

[Punisher74]

Dynamic Campaign Engine

 

http://forums.eagle.ru/showthread.php?t=153020&highlight=Dynamic+Campaign

[Big Nuts]

There used to be a game in 4 colour CGA, it used pyramids for mountains and the terrain was a solid lump of purple. It ran on a pc with a 25mhz cpu, 512k ram and 2mb hdd.

It had a campaign where you could plan attacks against any target to outcome the campaign.

OCA the airfields and deny them air power, then get the tanks? As a virtual F16 pilot you waged a war with your squadrons, and you quickly developed your own tactics. I would hit the fuel depots after a time and watch as the enemy war machine ground to a halt.

There is nothing like a mission that actually makes a difference to the war.

 

 

In the words of Brian Cox, it was 'Amazin'.

[Dappman]

What's the disadvantages to the bubble system? I always assumed DCS, and most games in general don't render objects that can't be seen, such as inside houses and stuff.

[GGTharos]

How would you process a trigger for a specific unit?

[Rangi]

What's the disadvantages to the bubble system? I always assumed DCS, and most games in general don't render objects that can't be seen, such as inside houses and stuff.

 

It's not the view rendering, the bubble is the AI calculations. In the other sim, units that contact each other outside the bubble don't really fight, the engagements are just a calculation/roll of the dice. I have seen Wags say a few times (years ago now) that ED don't want to go that way. I am hoping that after the poll, ED may put someone on to start seriously tackling a DC module, although even if they start now it will still be a few years before we hear about it, let alone get to use it.

Edited January 22, 2016 by Rangi
punctuation and autocorrect

[Jerkzilla]

It's not the view rendering, the bubble is the AI calculations. In the other sim, units that contact each other outside the bubble don't really fight, the engagements are just a calculation/roll of the dice. I have seen Wags say a few times (years ago now) that ED don't want to go that way. I am hoping that after the poll, ED may put someone on to start seriously tackling a DC module, although even if they start now it will still be a few years before we hear about it, let alone get to use it.

 

That's a very strange stance to take on the issue of dice rolling battles outside the bubble. AI battles either fully simulated or not are just calculations based on arbitrary characteristics and limitations anyway. You can account for various modifiers such as terrain like traditional wargames and basically tackle the issue backwards from the required result. Thus, you can decide that a T-72A has "x" chances to K-kill an Abrams in neutral conditions based on old fashion research, and then pile on contextual modifiers that account for unit training level, terrain, time of day and supporting units, such as recon, whose values are also determined based on old fashion research. The beauty of it is that if you do a decent job with the math, which presumably is easier than doing super capable AI, the result you get invariably is statistically consistent with your research.

[Rangi]

That's a very strange stance to take on the issue of dice rolling battles outside the bubble. AI battles either fully simulated or not are just calculations based on arbitrary characteristics and limitations anyway. You can account for various modifiers such as terrain like traditional wargames and basically tackle the issue backwards from the required result. Thus, you can decide that a T-72A has "x" chances to K-kill an Abrams in neutral conditions based on old fashion research, and then pile on contextual modifiers that account for unit training level, terrain, time of day and supporting units, such as recon, whose values are also determined based on old fashion research. The beauty of it is that if you do a decent job with the math, which presumably is easier than doing super capable AI, the result you get invariably is statistically consistent with your research.

 

It would certainly be better than the nothing we have now. I do trust ED to do it properly, whichever way they end up going. Just a matter of making like a hospital (plenty patients).....

[doveman]

Whilst a bubble system would seem sensible for a SP DC, it's not so straightforward for MP where you might have players all over the map and few or no areas where you can dumb down the AI without it being noticeable and affecting players.

[tintifaxl]

Falcon4 has a multiplayer 24/7 dynamic campaign, so Spectrum Holobyte/Microprose was able to solve that problem with hardware available in 1998.

 

Retrofitting such functionality into an existing engine might be a challenge, though.

[Dappman]

Any word from ED about the poll where DC scored top?

[mrfoxik]

New DCS 1.5 and still we don't have dynamic multi campagin? OMG

However, one key feature of the F4/EECH campaigns, is that it flows without slicing it to missions, you can join and leave anytime, there are missions generated continuously, AI or player pilots are accepting and flying them.

Would be great to have a dynamic MULTIPLAYER campaign.

That would require tools/methods to manipulate units on the fly which.

 

Im working personaly on something like that but I dropped works on it for a while because I am little frustrated, I hve paid for DCS not to sit houndreds of hours to make campagin but to play it. Never mind, I have idea how it can work to give good fun both for pilots and for ground unit players.

 

I have tested with friends the maximum number of objects AI who are moving on strong PC. Number is 4200 after this it is starting to lagging. I had problems with lags ealier with C.A 1500 objects on map. But problem was lieing in triggers and scripts. Too much complicated "continous" instructions, then I spend another day to find way to optimize it, it works fine up to over 4000. It means 2000 object per side of conflict. Is not much but ok.

 

My idea how to get it working:

Map is very big, so I have to divide map for 3 up to 4 stages just to move to concentration of units closer to battle front. Of course towns far behind are still covered by some units, bigger towns has SAM and other stuff, smaller AA guns etc. So if somebody would like to make guerilla tactic, he will be spoted and reported to players on other side.

Main conception is that both sides are fighting not for airports only but for towns (smaller and bigger) when is close to airport then its posible to capture airfield. Fighting for towns (capturing one opens the opportunities to preform attack to another one and in captured town its posible to spawn vehicles, Each city has limited unit spawn, if some will its posible to respawn them after 30 minutes) gives fun for tank drivers, because distance between cities are smaller that airfields, so they dont have to drive 100 km 45 min to die, hahaha. More over during moving to towns there is random function witch can generate and spawn some small forces like road blocks to stop tanks a little.

AWACS/EWR: I prefer to use EWR because its harder to find and to destroy it. Players are able to spawn transport in main base and send them somewhere where they want to put EWR. When transport will reach destination within 10 min EWR group will respawn and another one will be deactivated. When EWR group will be destroyed then new posibility of respawning EWR transport will apper after 30 min. When is not enought players CPU is taking control over units, when 0 players on server after while all units will stop because somebody has to use radio commands to show comtuper what to spawn and where it should go. There is also plenty CAP misions but you never know if you will find enemy in the air or not, another random function.

For optimalization, units far far away form players are deactivated (town covers etc) if enemy fores are close, then groups defending towns are activating again.

 

It is something like ARMA but DCS gives posibility to play it on simulation level what I love. I have hope that in the future we will paly something like this where both ground players will meet with pilots, and for both playing DCS on same server will give a lot of fun. I have tried tanks. With organized group attack or defend game looks amazing...and adding pilots on the sky its a hell on the ground. Battlfield in good simulate edition.

 

Greetings for players and programers

 

P.S I will buy all modules for DCS when something like this we will get it working.

[mrfoxik]

Its a pity that Digital Combat Simulator will not simulate military manoovers, but considering the name of DCS it should. For what do we have tanks artilery, vehicles, AA, posibility to comand the groups large map, EWRs, AWACS, ships and many many more if ED does not offer multi campagin. Its nice looking game with good phisics but still something is missing in this chain. Have bought combined arms because had hope that in the future I will take tank regiment and will go to battlefield of online dynamic campagin. Now I understood that Flaming Clifs and combined arms is enought for me, rather will not buy more.

ED will produce more and more modules but my question is for what? I am feed up of clasic PVPing and formation flying. I think that will instal RAGE eurofighter typhoon :D

Edited March 16, 2016 by Fair_Player[PL]

[paulca]

Imagine a world where software companies actually worked together instead of Golum'ing their precious IP.

 

Imagine a world where different games could interact online.

 

You could have someone (or many) people playing CMANO creating strategic and tactical strike plans and posting the results to a tasking order board. People playing DCS (or other sims) pick up tasks which generate missions which they fly. Imagine the results go back to CMANO where you can watch realtime or send new orders down the chain of command. Flights not taken by players in any one of dozens of sims get AI to conduct them.

 

Call it DUCT... Digital Unified Combat Theatre.

 

But... unfortunately we don't live in that world we live in a world of Golum's..."My precious... My precious..."

 

Out of interest, one of the greatest exercises that forced software companies to work towards a common goal is.... The Internet. Didn't do too badly, but in fairness it was started as a military project and conducted initially by Universities. If companies where left to their own way of doing thing (Golum) we would still have Microsoft Network, Compuserve, AOL and Prestel.

Edited April 7, 2016 by paulca

[Alwandy]

Just curious, has DCS released any statement about future in regards of dynamic campaign?

As I recently bought M-2000, F15, A10C and MIG 21.. I see a insane amount of fun with a dynamic campaign than keep re-making missions per statistics.

 

Or is this still a no go to happen? If so, anyone know reason?

[StandingCow]

I could def be wrong, but so far the only thing I have seen about it was it's inclusion of a poll... and nothing else has been said about it.

[Alwandy]

I could def be wrong, but so far the only thing I have seen about it was it's inclusion of a poll... and nothing else has been said about it.

 

Okey, what was the result?

[StandingCow]

Okey, what was the result?

 

It won by a longshot against everything else. But we haven't (again, that I recall) heard anything else about it. They were probably just gauging interest.

[J. Heller]

Dynamic Campaign

 

All this time... Where is it? Put this game down two or three years ago due to the boring, repetitive, and scripted missions. Nothing has changed, will it ever change? I am still using software from 1998 that provides more replayability than this "modern" sim. Thats my one wish, a dynamic campaign. Everything else is just... repetitive.

[QuiGon]

All this time... Where is it? Put this game down two or three years ago due to the boring, repetitive, and scripted missions. Nothing has changed, will it ever change? I am still using software from 1998 that provides more replayability than this "modern" sim. Thats my one wish, a dynamic campaign. Everything else is just... repetitive.

 

I feel kinda the same, so I can totally understand you. Unfortunately there is nothing to be expected in the foreseeable future in this regard. :(

[Rangi]

I feel kinda the same, so I can totally understand you. Unfortunately there is nothing to be expected in the foreseeable future in this regard. :(

 

I foresee a future DCS with dynamic campaign, however that future is very, very distant. I am in 2 minds about a good dynamic campaign, on the one hand it would be great and I would play dcs a lot more and spend a lot more money on modules and peripherals, but on the other hand, I would play dcs a lot more and spend a lot more money on modules and peripherals and my family may not appreciate it as much as I would. ;)

[boopidoo]

Yes, true dynamic campaigns would be really great. I too remember TAW and have been looking for something of that kind since then.

 

I would gladly pay for true dynamic campaigns on a grander scale (like TAW) in DCS.

[bkthunder]

ED has a plan for dynamic mission generation which was not entirely put into implementation in A-10C due to resource constraints. That's all I'll say and hope the powers that be won't zap me.

 

As you can see, a host of new features are added with each iteration DCS. While the OP complains that it's all graphics, some people like Ripcord have seen that there are significant improvements in the ME as well - and I don't mean little bug fixes, I mean real, useable, new features.

 

So, this was back in 2011 (quoting post #10 http://forums.eagle.ru/showpost.php?p=1071158&postcount=10)

5 years later, what are the significant improvements to the ME, that bring us closer to a dynamic campaign?

[GGTharos]

I believe that the idea is always in ED's view, but probably a lot of other things are of higher priority right now.

[Angelthunder]

I hope we get more realistic campaigns in the future.I'm getting tired of seeing campaigns using fictional nations like the one ED just released yesterday for the F-86 Sabre and the Mig-15.I would like to see real or what if conflicts between real nations or organizations like NATO or the Warsaw Pact.I'm sure someone could make a realistic scenario for the Caucasus,SoH and the Normandy maps coming up.