Dynamic Campaign Discussion Thread

[GGTharos]

I'm not a programmer, and I will gladly admit my error if anyone wants to explain why making a dynamic battlefield simulation would not be a considerably more difficult task. And you seem unusually angry--no offense was intended. It was just a question.

 

I'm not angry - that just doesn't do anyone's nerves any good ;) No offense. It isn't the question, it is the use of the word 'trivial'. Or 'easy'. This happens a lot with this particular topic (and other tasks, actually) and it is annoying as all hell, especially when (some) people seem to imply that ED somehow hasn't thought things out or made up a plan about their pet feature.

 

The existing campaign system, at least as I understand it, already functions by pulling pre-made missions from a "folder" for each phase of the campaign. And the random mission generator makes missions. Is it really that crazy to consider the possibility of combining the two?

 

That is correct, and there is nothing crazy about it. It is a matter of time and effort in the never-ending task of adding other features also. ED isn't going to drop everything in favor of enhancing the campaign - like every other feature, it will be enhanced with each iteration of DCS.

[mic1184]

I guess GG's just tired of hearing that building a dynamic or semi-dynamic campaign would be (relatively) easy or trivial thing ;-) don't take it personally. we have to give him credit that he's still following the thread and takes part in the discussion.

 

Given the improvement from beta4 to final and the fact, that ED policy now tends to deliver surprises rather than promises (having had to cut battlefield commander, jtac mode and nevada map after feature-locking the beta), we might as well get a fully dynamic campaign exceeding any expectations in the next patch :thumbup:

 

i think they're going the right way in carefully improving all aspects of the game. they will continue improving the mission editor adding more random stuff and hopefully improve the campaign builder adding some inter-mission dependencies and the thing will slowly evolve into something with more re-playability

[HerrKaputt]

I'm not a programmer, and I will gladly admit my error if anyone wants to explain why making a dynamic battlefield simulation would not be a considerably more difficult task. And you seem unusually angry--no offense was intended. It was just a question.

 

The existing campaign system, at least as I understand it, already functions by pulling pre-made missions from a "folder" for each phase of the campaign. And the random mission generator makes missions. Is it really that crazy to consider the possibility of combining the two?

 

From my experience in reading the code in EECH and trying to improve the code in their DC (which is from what I've seen inferior to the one in Falcon 4), it's not trivial, but it's certainly not the Herculean task GGTharos has been implying it to be.

 

The biggest problem I've encountered is, as Speed mentioned, not generating the mission itself once you know what mission you want. For that, you start with generating an obviously poorly planned mission (say, you find a group of tanks that must be attacked, so you just find the nearest available Ka-50 or A-10C group, plot a direct course to the target, and voila, mission created), and you build from there (which is, like I said, not trivial, but not Herculean either).

 

The major problem is indeed: OK, you wrecked that group of tanks. Now what? You have to generate new missions that take that into account: perhaps some of your own ground forces can now advance, or perhaps you schedule a new air mission do do damage assessment? This type of decisions isn't easy at all to model (the coding isn't the hardest part, at least not for me -- it's knowing WHAT to code). Not to mention the fact that even in the middle of a mission things can change.

 

As GG and many others have said, it's a matter of priorities: ED's priorities are driven by their military contracts, and their military contracts do NOT include making random mission generators. It's fine that they don't want to do it -- it doesn't mean that it's incredibly hard, others have done it before and using computers that were equivalent to a modern iPad or something.

[FlyinJ]

Has anyone ever made an external shell dynamic campaign mod for any of the previous ED games, like the ones made for Rise of Flight and IL2? Has anyone said they were going to work on one for A-10?

[Ripcord]

Well, it may not be a dynamic campaign, but I have begun developing a dynamic tasking system that works within a mission. Basically, the idea is that it will be a lua script that will generate tasks for the players based on current battlefield conditions. These tasks are NOT preset, but they will have preset types... BAI, CAS, forward observer (for artillery, yes), maybe AFAC, strike, etc. The tasks would be communicated to the player through text to screen, (and quite possibly, work in multiplayer only- so a single player would just have to launch the mission as a multiplayer server). So, you might take off, and the script detects enemy units line of sight and within like 15km of a friendly ground unit, so you get a CAS mission. Or if the enemy is within range of an arty or MLRS piece, you might get a BAI/FO mission. Additionally, I'd like to work eventually it so that it gives SEAD and air to air missions to other aircraft. Maybe even carpet bombing missions to B52s. That kind of stuff is possible (you have to have the AI aircraft already airborne though).

 

So anyway, that's my current intention. It's SORTA half a dynamic campaign system- the harder part is making a program that automatically builds a mission that's based off the results of the previous.

 

I can't promise anything, but that's what I'm working on. My intention is to build it up slowly over the next few months. The first basic elements should be released as part of a mission within, maybe a month or so. Who knows how long it will take me to build it up.

 

It's no dynamic campaign, but it's a mission that is not preset, and is based on battlefield conditions. What you do, what other aircraft do, what ground forces do would determine your next task.

 

Speed, I am very interested in this project. I think it has a lot of merit and it is very similar to the concept that I tried to describe earlier in this thread - except that your are attempting to build the mission 'on the fly'. I would think that there should exist a program of some sort that analyzes the tactical / strategic situation, runs a series of calculations and returns with a certain TYPE of mission - eg, CAS, BAI, AFAC, all the things you mentioned.

 

Mission templates would need to track the advance of friendly and enemy troops, using a series of zones inside mission, located at various places along the front line. Units would need offensive and defensive objects. The program running in the background (or running before and after the missions) could also keep track of the Order Of Battle for all combat units in theatre, and keep track of their combat condition/status in order to determine which units should be placed on the map and where -- not unlike what we see in the mission generator tool.

 

A lot of the work would be just sitting down and deciding what data to track and store and analyze. I wish I knew more about programming -- I'd join you in your quest. Keep giving us updates

 

Ripcord

[Eddie]

How hard was it to do Falcon's DC?

 

From the man himself........

 

http://www.cleared-to-engage.com/2011/03/12/interview-with-kevin-klemmick-lead-software-engineer-for-falcon-4-0/

[Distiler]

How hard was it to do Falcon's DC?

 

From the man himself........

 

http://www.cleared-to-engage.com/2011/03/12/interview-with-kevin-klemmick-lead-software-engineer-for-falcon-4-0/

 

It was a very nice read!

[asparagin]

How hard was it to do Falcon's DC?

 

From the man himself........

 

http://www.cleared-to-engage.com/2011/03/12/interview-with-kevin-klemmick-lead-software-engineer-for-falcon-4-0/

 

Thanks for sharing.

 

Someone get this guy out of Facebook games and put him back on the sim market!

[HerrKaputt]

Excellent read.

 

I found it really interesting that he used three layers for the F4 DC.

[CyBerkut]

If E.D. were so inclined (yeah, yeah, I know), I wonder if they could lure Kevin Klemmick away from Electronics Arts to do a dynamic / scripted campaign system for the DCS series? He still seems interested in that sort of a project, and look at what he accomplished with approx. a 5% slice of CPU time all those years ago. Imagine what could be done with today's multi-core processors.

 

It seems like that could have positive implications for the The Battle Simulator as well (ie. an economical war gaming option built into TBS).

[GGTharos]

If E.D. were so inclined (yeah, yeah, I know), I wonder if they could lure Kevin Klemmick away from Electronics Arts to do a dynamic / scripted campaign system for the DCS series? He still seems interested in that sort of a project, and look at what he accomplished with approx. a 5% slice of CPU time all those years ago.

 

 

What makes you think ED isn't already working on such things? Did you somehow miss the ME developments so far?

 

 

Imagine what could be done with today's multi-core processors.

 

It seems like that could have positive implications for the The Battle Simulator as well (ie. an economical war gaming option built into TBS).

 

The most interesting thing that he said was that he wouldn't try doing a DC again given the difficulty in terms of technical obstacles, and especially time and money.

 

In any case, the military likes to set up their own war games, and they probably do it a little better than any DC ever could. Falcon's DC isn't realistic - it's there to entertain.

[asparagin]

The most interesting thing that he said was that he wouldn't try doing a DC again given the difficulty in terms of technical obstacles, and especially time and money.

 

Yeah, but I like the way he put it better: ;)

 

I think if I were to do it over I would do a mix of scripted/generated missions, so that the player still feels like they’re involved in the world, but there is also some variety thrown in to keep things interesting.

[GGTharos]

Yep. That's where DCS looks like it's heading towards now - maybe in a module or two? Just my opinion, of course I'm an optimist.

[Distiler]

A il2 SEOW like dynamic campaign with commanding UI, etc. would be very nice too. This software is independent from il2, it just builds mission files for it. Dunno how, but il2 community created some cool apps out there, including two or three awesome dynamic campaign generators (BADC/BELLUM and il2DCG, way simpler than SEOW), hard to do, absolutelly, doable even by one guy, true! if I had the knowledge I would probably try to build one myself, don't know programming tho :(

 

Scorched Earth Online War (SEOW) is a persistent coop campaign system for the IL-2 series of flight simulations. SEOW allows users to download a variety of pre-built campaign templates, or to build their own campaign templates using the IL-2 Full Mission Builder (FMB). A SEOW campaign, basically a set of opposing forces placed on a map, is initialized with a starting template which includes the order of battle (on both sides) that will take part in a given theatre of operations. The system will then allow for sequential hourly mission generation based on direct input from at least two opposing commanders, one Allied and one Axis. The system will also allow sub-commanders of individual units.

 

All air, sea and ground operations are planned using a web-based mission planner. As a commander, you classify which flights will attack available targets, and when. Similarly, you command your own tanks, guns, vehicles, trains and ships. Fog of war persists throughout, with reconnaissance levels affecting the amount of information that opposing commanders can use to plan missions. The mission plans for any given hour in the day cycle are merged based upon the plans of the opposing sides. One essential difference between SEOW and other IL-2 online wars is that there is no computer-generated movement or opposition. SEOW is a head-to-head, human-vs-human strategic contest in terms of planning, PLUS you also get to fly each mission as a tactically oriented pilot - the best of both worlds.

 

The mission is generated local to the Host computer, then hosted to the squadrons taking part in the online campaign using standard IL-2 coop hosting methods. The best part of all of this is that the combat outcomes in each mission are persistent. If you take out a bridge early in the day, it will still be destroyed several hours (missions) later. A tank destroyed at a junction will leave a burning wreck in the precise location. Bombing of industrial area will destroy buildings and cause smoke visible for miles. As you can see, SEOW expands upon elements featured in previous online campaigns, but a major aspect of SEOW is that it offers persistent damage templating and full command over all units, as well as an open source software format. If you feel like a part of SEOW is not to your liking, code a new one. We hope to make the process of providing modifications to the system as smooth as possible and centralized via http://www.sourceforge.net/

 

http://seow.sourceforge.net/

 

Perhaps, with more DCS modules (and popular like A-10) we build enough community to get to the point of someone embarking in a similar project (I think it has been tried before, dunno if right now somebody is doing it, user "Speed" perhaps?)

Edited March 15, 2011 by Distiler

[TheCabal]

Falcon's DC isn't realistic - it's there to entertain.

 

Dude, whether Falcon's DC is realistic or not, it doesn't actually matter. Fact is, its unique since 11 years in a sim. You got the Weapon System named F-16 and you are to use this under different situations. You feel extremly involved in one of the DC's, since everything makes you thinking is actually moving in Korea & Co. . It's not about entertaining as you try to keep F4's DC Image low, it's more a challenge which is changing all the time.

 

I feel like you're trying to let Falcon's DC look smaller than it actually is. It is still (thanks to the modders) the "overall" god in flight simulations.

Edited March 16, 2011 by TheCabal

[empeck]

This software is independent from il2, it just builds mission files for it.

 

Over Flanders Fields is a CFS3 (yes, I know) mod with external campaign engine and it's really something special.

[CyBerkut]

What makes you think ED isn't already working on such things? Did you somehow miss the ME developments so far?

 

The most interesting thing that he said was that he wouldn't try doing a DC again given the difficulty in terms of technical obstacles, and especially time and money.

 

In any case, the military likes to set up their own war games, and they probably do it a little better than any DC ever could. Falcon's DC isn't realistic - it's there to entertain.

 

Hi GG,

 

I don't know if they are working on getting a dynamic element into the campaign system, or not. I am admittedly not up to date on the latest ME developments. In truth, a dynamic campaign isn't a really big deal to *me*, but I have seen where a segment of the community has hammered E.D. over it repeatedly. I don't presume to know what is best for E.D., but I do want to see them continue to prosper. Kicking ideas / concepts around is a "good thing".

 

I agree Falcon's DC isn't realistic. But then, I wasn't talking about adding Falcon's DC, but rather it's programmer... and imagining what could be done with the much better processing power that is available now.

 

And yes, I understand that the military likes to setup their own war games. However, financially, times are getting tough all over, even where the big eagle writes the checks... so even the military looks for ways to save money. My reference to "an economical war gaming option built into TBS" wasn't so much aimed at entirely replacing the big production war games so much as it was aimed at giving the military a lower cost means (ie. fewer people needed) to have supplementary sessions. They may be sessions that are more limited in scope (almost certainly so), but that can still have value. If that can help them somehow, and E.D. can sell them the means, so much the better. :)

 

E.D. will do whatever is best. I don't second guess them.

[GGTharos]

I really doubt that any dynamic campaign as implemented in entertainment products is of any use whatsoever in military applications.

 

And I really doubt you'd enjoy (or even get your hands on) on anything that resembles a DC simulation that the military does in fact use. ;)

 

They're not interested in DCs in flight sims AFAIK. They're useless for military training.

[GGTharos]

It's not about entertaining as you try to keep F4's DC Image low, it's more a challenge which is changing all the time.

 

It is all about entertaining. It isn't even hard to figure out how and why - the persistent world idea is something that is desireable both in terms of continuity and in terms of not having to do much to set up your next missions.

 

I feel like you're trying to let Falcon's DC look smaller than it actually is. It is still (thanks to the modders) the "overall" god in flight simulations.

 

It is smaller than you think it is. If it is the 'overall' god in flight sims, why aren't all flight simmers playing F4 exclusively?

[Haukka81]

It is all about entertaining. It isn't even hard to figure out how and why - the persistent world idea is something that is desireable both in terms of continuity and in terms of not having to do much to set up your next missions.

 

 

 

It is smaller than you think it is. If it is the 'overall' god in flight sims, why aren't all flight simmers playing F4 exclusively?

 

 

 

Can you tell us how falcon dc it's not realisticand how it's small?? I don't care for dc but im just curious :book:

 

I like janes style semi dynamic mission/campaign system :thumbup:

[GGTharos]

Those two go hand in hand.

 

The ATO generates silly missions or silly mission flight paths. In the end, it's suicidal for the flights involved. Sure, you can involve yourself and replan flights, but then the 'red side' is suddenly at a huge disadvantage.

 

The order of battle is wrong usually - the opposition has much more equipment than they should have. If they didn't, it would make for a boring game ;)

 

Aircraft in F4 often equip capabilities that don't exist in real life: R-27ETs with datalink, R-77M's with ramjets, etc.

 

Can you tell us how falcon dc it's not realisticand how it's small?? I don't care for dc but im just curious :book:

[Distiler]

But the man says the hard stuff was making it work with just 5% of the cpu! and that is because the campaign is like in real time. Can't be that hard to make a non real time mission generator that links missions. Like il2 default dynamic campaign, or bellum, or il2dcg.

I mean, it would be very hard for me that I ignore how to even start, but seems the guy had no problems except for cpu resources. Peraps the A-10 quick mission generation is a first step? I don't even know, don't have A-10 to try.

 

There was even a little dynamic campaign generator ala il2 for lockon, named RamGen: http://forums.eagle.ru/showthread.php?t=36455&highlight=ramgen

 

Can't ED allocate some resources to do a similar stuff, even if it's for the sake of entertaining? or even just giving special help to some ppl that tries to do it by their own (I know, some tried in the past, and perhaps it is trying right now). Making missions manually takes a lot of time, and debriefing, etc. then making a linked mission takes even more time :(

Edited March 16, 2011 by Distiler

[GGTharos]

DCS: A-10C comes with a dynamic mission generator. I'm sure they'll work it into the campaign engine at some point. In this iteration they allocated resources to making the mission generator happen.

[Distiler]

DCS: A-10C comes with a dynamic mission generator. I'm sure they'll work it into the campaign engine at some point. In this iteration they allocated resources to making the mission generator happen.

 

I don't have A-10, but from reads I think you mean dynamic as automatically generated right? so the next thing would be to link those missions with some persistency between them right? and, perhaps later, put some resource management system and career system.

 

If that's I'm for it!

[GGTharos]

Whenever 'the next thing' is scheduled to happen, I'm sure it will. It's on their schedule, not yours - that's all I'm saying ;)

 

I don't have A-10, but from reads I think you mean dynamic as automatically generated right? so the next thing would be to link those missions with some persistency between them right? and, perhaps later, put some resource management system and career system.

 

If that's I'm for it!