A Poor Man's Dynamic Campaign Engine

[ericinexile]

Is it possible to create a third party campaign generator for DCS? I ask because the lack of a real Head-to-Head element makes long-term playability questionable in my opinion, at least in multi-player. What if a server could, after shut-down, pass along unit position, strength, and supply level to the 3rd party generator, and then that generator could then "rewrite" the mission with new unit levels and tasks that the server could then start with at re-boot?

 

This may have been tried with LOMAC but LOMAC has the H2H element that makes it fun even in static missions that run the same day after day, month after month. DCS needs something different. The image I have in mind is the MultiVipers server running Falcon 4:AF. Campaigns run for days or weeks before one side claims victory. Players care about the outcome though their individual roles in that outcome are minimal. Players can either fly Packages generated by the campaign engine or pick their own targets and grab some friends to help take 'em out. The illusion that you are part of something bigger makes for a much better MP experience.

 

I'm not suggesting that ED change anything (...or...much)--DCS is the greatest sim yet made. I'm just asking if it is possible to have an evolving on-line battlefield.

 

Regards,

Smokin' Hole

Edited January 13, 2009 by ericinexile

[RichardG]

I was thinking the same thing. Or just a quick mission creator. I like flying sims, but I have no desire to program or build missions (thats work, not play). After a few solid weeks with blackshark, the same scripted missions are getting boring, so I find myself heading back for Falcon4 for something unexpected.

[MrReynolds]

....lol .... another thinking along the same lines.

 

Maybe there could be a way for some "demi-god" of a mission builder to build a huge structured mission. Units could become active after several hours, various fronts could be created.

 

Now....is there a way to save a mission? So when 1 flight is finished, instead of quitting the mission and that's the end.... you could just exit and the missions state would be saved. Now the user could open the mission and add a flight of ka50s and get back into the action. Certain units could be visible (intel) and others hidden.

That way anything left will remain, and everything yet to appear would be ready to spawn at the appropriate time. Airfields could get reinforcemets after a couple of hours.

I used to do that with my Falcon campaigns I edited. Airfields were always topped up with sams every so often.

Put that in your pip and smoke it!!! :P

Edited January 13, 2009 by MrReynolds

[H-street]

the best solution is just to add a true dynamic campaign into DCS, only ED would really be able to do it justice..

 

i'm hoping that with the A10 addon something like a dynamic campaign will be added. I mean this is a Digital Combat Simulator :) there is more to combat than just the aircraft..

 

its one of the reasons i have been taking it slow with BS.. i know once i fly all the missions and the campaigns it'll be harder to come back and replay it.

[Ferski]

I still don´t own the game but i have to disagree with you, i have read the GUI manual and with the given tools you could make a single mission that you will not run the same. Just put a 10 random trigger at mission start and you have 10 fresh missions, add flags to the combination and those random triggers could take several paths. With not so much effort you would need 30 or more runs of the same mission to "run the same" just playing with flags and triggers. So at the end it is on the hands of mission builder how much replayable can be a mission.

[RedTiger]

Here's a thought:

 

It would be interesting to hear from someone who has flown the campaigns in BS extensively (in other words, multiple times). Perhaps a beta tester? To what degree exactly is the campaign random? Are there enough combination of triggers, units, and missions to create enough permutations to truly seem random?

 

My assumption is that a person flying something like a flight sim campaign is only going to be able to perceive something as "random" to a certain degree and then every bit of randomness added on top of this is basically a waste since it won't be recognized. What's "inside the bubble"? What is outside the bubble? Should we care what is outside the bubble? From a realism stand point, tactically, should we realistically be venturing outside the bubble? If not, can you still criticize the campaign if nothing exists outside the bubble because you decided to play Dora the Explorer and find your way out of it?

 

Another way of putting this is to wrap your minds around this; if an F-16 shoots an AMRAAM over the Korean DMZ and no one is around to hear it, does it make a noise? ;)

Edited January 13, 2009 by RedTiger

[multiviper]

That Dynamic Campaign engine in Falcon 4.0 is what has kept Falcon Alive all these years, and still has not been duplicated to this day. also the leaked code helped to ;)

 

It would be nice to have a new Dynamic campaign engine in DCS.. :)

[ZaltysZ]

It would be perfect if there was possibility to add/remove/modify units at runtime via lua socket. What a "monster" virtual war could be created.

[LIONPRIDE]

Here's a thought: ... ;)

 

I think they're talking more along the lines of single mission multiplayer ideas.

 

 

.

[RedTiger]

I think they're talking more along the lines of single mission multiplayer ideas.

 

 

.

 

There was a wee bit of side discussion about how random the DCS campaign is, but yes, the main discussion is about an on-going dynamic campaign for multiplayer.

 

However, if someone's asking about a 3rd-party DC, I don't see how my thoughts aren't on topic.

[d0ppler]

In the mission editor, I even don't find any "random weather" or "random time" function. It might be there somewhere since I haven't read the GUI manual though.. :)

 

 

But a dynamic campaign like we saw in Falcon AF was truly awesome, and adding this feature will make this game my favorite game for years to come.

[Nate--IRL--]

It would be perfect if there was possibility to add/remove/modify units at runtime via lua socket. What a "monster" virtual war could be created.

 

Also the ability to output events and object states from the Sim, and then you could have a completely independent piece of software controlling and monitoring the Sim environment in realtime.

 

With the ability to add/remove/modify units in realtime, we could get the bubble system similar to Falcons in order to get around the problem of the AI bogging down the sim. Only a certain amount of units in the Sim environment with the rest of the war being held in the campaign engine until needed by the Sim.

 

Well we can dream anyway.......

 

Nate

[uhoh7]

That Dynamic Campaign engine in Falcon 4.0 is what has kept Falcon Alive all these years, and still has not been duplicated to this day. also the leaked code helped to ;)

 

It would be nice to have a new Dynamic campaign engine in DCS.. :)

 

You got that right. I just wish we could get ED quality FMs and low level terrain in F4.

 

It's a cryin shame.

[tn_prvteye]

Take a look at what DoctorK is doing with LOMAC. Yes, it's not really dynamic (a la F4), but it is random, and in my opinion, very fun!

 

http://forums.eagle.ru/showthread.php?t=36455

 

He said he's not going to do one for BS, but it shows that it's possible.

[EvilBivol-1]

To create a third party campaign editor you really need two things from the simulation - ability to import the initial world state and export the ending (post-mission) world state. Black Shark missions (initial world state) are already exported into a text script by the ME (open any .miz mission file with winzip/winrar and see for yourself), so it's definitely possible to create mission generators/editors to import missions created externally into the sim. The more complicated question is of the export functions at mission-end, but looking at Tacview, I believe there is enough export data possible through LUA to accomplish at least the essentials. I'm not very knowledgeable on the subject though, so perhaps somebody else can provide more info.

 

In short, I believe that it should be possible to create third party mission generators/editors as well as campaign generators.

Edited January 14, 2009 by EvilBivol-1

[SuperKungFu]

this Random Mission Generator for FC should be a good start. I see a great futrue for this

http://forums.eagle.ru/showthread.php?t=36455

[A.S]

That Dynamic Campaign engine in Falcon 4.0 is what has kept Falcon Alive all these years,...........It would be nice to have a new Dynamic campaign engine in DCS.. :)

 

how true......its (in someones words) "the meat and the potato" of MP environment may it be FoF or not.

[Nate--IRL--]

To create a third party campaign editor you really need two things from the simulation - ability to import the initial world state and export the ending (post-mission) world state. Black Shark missions (initial world state) are already exported into a text script by the ME (open any .miz mission file with winzip/winrar and see for yourself), so it's definitely possible to create mission generators/editors to import missions created externally into the sim. The more complicated question is of the export functions at mission-end, but looking at Tacview, I believe there is enough export data possible through LUA to accomplish at least the essentials. I'm not very knowledgeable on the subject though, so perhaps somebody else can provide more info.

 

In short, I believe that it should be possible to create third party mission generators/editors as well as campaign generators.

 

Now that I think of it youre proabaly right, but the ulimate goal should be the export and import of this data on the fly into and out of the simulation as its running.

 

Nate

[MBot]

The big problem I see with a mission generators is the complexity of the ground war. Generating missions is easier for simple mission profiles such as air-air combat or strike missions. But realistically populating the ground war is incredible complex and difficult, even for a human mission designer. Without any serious AI development on multiple levels (chain of command), where ground units will dynamically behave realistically in a given envoirnment, I can't imagine that a generator creating a halfway decent mission.

[ZaltysZ]

Generator can create descent missions. Well, really it depends on how much work was put into developing that generator. Anyway, it is not a problem.

 

The problem is in dynamics (not in campaign level, but in mission level). Triggers are great of course, however they rather make mission to look more dynamic than to be more dynamic. For example: lets say there is package: 2 SEAD aircrafts and 4 ground attack aircrafts; SEAD fails; currently there is no way to make ground attack aircraft abort their mission and they will get shot down.

 

If there was a way to add/remove aircrafts and change their waypoints at runtime, above mentioned problem wouldn't arise. 3rd party could write campaign generator with complex logics, which would query situation regularly and adapt; the best part would be that it could evolve to online war without need to change missions hourly. Imagine how it would be nice, if attacked convoy changed its waypoints and tried hide in forest or city.

[Teapot]

... i'm hoping that with the A10 addon something like a dynamic campaign will be added. I mean this is a Digital Combat Simulator :) there is more to combat than just the aircraft..

 

When addressing the subject of simulation, IMO there are many particulars that can be specifically treated.

Any simulation needs to be clear what its goal is. A clearly defined goal of what we wish to achieve via the simulation package will produce the template and guidelines in producing/choosing that simulation in the first place.

There are studies *out there* arguing that a fully featured high fidelity simulation (particularly one that simulates an air combat environment) is not (counter intuitively) necessarily the best approach for honing particular skills in the fluid/dynamic air combat environment.

In other words, we start from what skills we're trying to acquire, then having a simulated environment that focuses only on that aspect. Having more than this becomes a distraction. To this end a fully featured simulator that models a *dynamic campaign* doesn't usefully target a specific set for skill acquisition, because the parameters within the campaign are *dynamic* and thus not able to be controlled to produce any measurable outcomes ... unless the metric used is the degree of pleasure we gain through *playing* make believe :P.

I think that DCS straddles that area between a simulation tool and an entertainment package; and IMO (again) I find the TE (OF4.7) and ME (Lock On FC) to be much more *useful* tools than Falcon's Dynamic campaign ...

 

My 2 cents,

 

Cheers.

Edited January 14, 2009 by Teapot

[d0ppler]

You have some good points Teapot. I may be speaking for myself, but I believe most DCS gamers wants to be entertained rather than using DCS as a tool to practice lifelike 'real' scenarios/situations to become a REAL Ka-50 pilot.

 

It's nothing wrong with the scenarios/missions being realistic, and I believe a well written dynamic campaign engine is able to create such missions/events.

[Teapot]

... I believe most DCS gamers wants to be entertained rather than using DCS as a tool to practice lifelike 'real' scenarios/situations to become a REAL Ka-50 pilot ...

 

!!BLASPHEMER!! :D

 

Heheh ... you're absolutely correct, I too have no wish to become a REAL Ka-50 pilot; but that perspective will likely change once we get a REAL fast jet module (NOT including the A-10) ;).

 

Cheers.

Edited January 14, 2009 by Teapot

[ericinexile]

When addressing the subject of simulation, IMO there are many particulars that can be specifically treated.

Any simulation needs to be clear what its goal is. A clearly defined goal of what we wish to achieve via the simulation package will produce the template and guidelines in producing/choosing that simulation in the first place.

There are studies *out there* arguing that a fully featured high fidelity simulation (particularly one that simulates an air combat environment) is not (counter intuitively) necessarily the best approach for honing particular skills in the fluid/dynamic air combat environment.

In other words, we start from what skills we're trying to acquire, then having a simulated environment that focuses only on that aspect. Having more than this becomes a distraction. To this end a fully featured simulator that models a *dynamic campaign* doesn't usefully target a specific set for skill acquisition, because the parameters within the campaign are *dynamic* and thus not able to be controlled to produce any measurable outcomes ... unless the metric used is the degree of pleasure we gain through *playing* make believe :P.

I think that DCS straddles that area between a simulation tool and an entertainment package; and IMO (again) I find the TE (OF4.7) and ME (Lock On FC) to be much more *useful* tools than Falcon's Dynamic campaign ...

 

My 2 cents,

 

Cheers.

 

Well sure. When I get in a 737-800 sim there are just three people: Me, the FO, and the sim instructor. It's all set pieces. No ATC. No other aircraft. No Pax. No FAs. Just us and a tight sim schedule. It trains us very well, but it's not immersive...it's training. DCS isn't training for me, it's entertainment but I want as much of the illusion of realism as possible.

 

Smokin' Hole

[ericinexile]

The big problem I see with a mission generators is the complexity of the ground war. Generating missions is easier for simple mission profiles such as air-air combat or strike missions. But realistically populating the ground war is incredible complex and difficult, even for a human mission designer. Without any serious AI development on multiple levels (chain of command), where ground units will dynamically behave realistically in a given environment, I can't imagine that a generator creating a halfway decent mission.

 

There is no way to do it with substantial realism. But anything is an improvement over flying down the same canyon day after day. A DC generator would need to look at the battle field in much the same way that the old SSI turn-based war simulations. Every nuance of the battlefield will be elusive but a good generator would provide enough to give most of us the sense that there is a war going on out there.