Dynamic Campaign Engine

Hey MBot! I'm looking right now at an ingame briefing with a table in it: maybe that could help with the alignment?

Great work, thanks for the screenshots showing development! Like Bad Habit said, I'm more excited for this than for DCSW 2! Lol

Agreed!

I have been using the DAWS save mission mod and having a lot of fun with that.

Subscribe

Only the dead have seen the end of war - Plato

So cool MBOT! another Falcon player that would love to see Dynamic campaign generation.

Just needed te register for this one!

First post: GREAT WORK MBot!!!

Didn't write an update for some time, but I have been very busy and made some good progress. The project becomes more complex day after day, but it is steadily getting closer to what I have in mind.

I don't want to write a big update, just wanted to share the following because I think this is pretty cool :)

Aviation units (squadrons/regiments) are now completely present with almost all of their aircraft in each mission. This means that also aircraft which are unassigned for the current mission sit idle and empty on their units airbase (I probably still have to do some sort of reduction factor for very large units).

At the same time, the ATO is able to plan strike missions against aircraft on airbases (Offensive Counter Air).

MiG-29 regiment at its airbase. Some of these aircraft might still go on a sortie later in the mission and some might be assigned to ground alert intercept duty, ready to launch when enemy aircraft are detected by EWR/AWACS. But here most are empty, perhaps doing maintenance or rest, waiting for a sortie at a mission later in the campaign.

A flight of Strike Eagles is about to do something about that "later" part.

2 MiGs were destroyed on the ground and 5 more damaged. The damaged ones might get repaired later in the campaign, but for now there are 7 MiGs less to worry about in the next missions.

Amazing work!

Nice. Keep it up.

Awesome!

MBot, that's so friggin' cool! Can't wait to get my grubby little hands on this once you're happy with it and release it to us poor slobs!!

Let’s talk a bit about intercept missions. The campaign designer can manually influence the amount of interception sorties in the ATO. This is done per airbase by adding an intercept target to the campaign target list. Target might be confusing term for defensive or support sorties, objective might be more appropriate in this case (but do not worry, this is all in the backend). So the campaign designer would define, let’s say: Airbase RAF Lakenheath should try to provide interceptors equaling a firepower amount of 50, with a priority of 10.

The campaign engine will now try to add to the ATO suitable and available aircraft for intercept sorties based at RAF Lakenheath. Perhaps 2 F-15C with a AMRAAM loadout with a firepower rating of 25 each, or 5 F-15 armed with Sparrows with a firepower rating of 10 each, or 20 F-16 armed with Sidewinders with a firepower rating of 5 each, or some combination. Of course, aircraft have to be available for this. Generally, higher capable aircraft/loadouts are assigned first. But perhaps all F-15 are already assigned to higher priority sorties and only 2 ready F-16 are available at that base, in which case only these two F-16 would be added to the ATO with intercept tasks.

All aircraft with intercept tasks are parked as uncontrolled aircraft at their base during the mission (so they can be destroyed on the ground). They are launched as targets are detected by the EWR network. All EWR stations and AWACS per side combine all the targets they detect. The detection logic runs every 15 seconds, so there is a chance that a short pop-up target might slip through. Once a target is detected by the network, interceptors from the nearest in-range base are launched on a 1-to-1 basis. As interceptors are destroyed in combat, more are launched to sustain a 1-to-1 ratio. Since the approaching targets usually consist of both attack aircraft and escorting fighters, this should generally result in a favorable interceptor-fighter ratio (of course only as long as enough interceptors are available).

As you can see, avoiding detection by EWR is paramount to avoiding interceptors during strike missions. This is especially true for scenarios without AWACS, where considerable gaps in early warning coverage exist. As discussed previously, strike missions are planned with routes to avoid early warning radar stations, both horizontally and vertically. Since version 1.5, radar horizon is modeled in DCS. Which means you can avoid detection by EWR down to a range of 40 km on flat terrain at very low level. Also, destroying EWR stations early in the campaign will significantly reduce your risk of being intercepted in later missions.

Players won’t have to spend time waiting on the ground for a possible intercept. Players are assigned to intercept tasks only if there is an enemy package in the ATO, which will enter EWR coverage and be in range for the player to intercept. If this is true, the enemy package’s time on target is modified so that it spawns at mission start right at the position where it enters EWR coverage. Which means for the player the mission starts at exactly that moment where he is supposed to launch. If the above conditions cannot be satisfied, the player will be assigned to some other available sortie instead.

Damn, that's some incredible work you're doing there! :thumbup:

MBot, maybe this has been already asked, but what's the technology for the back-end you're using?

Well, this is looking incredibly satisfying:)

MBot, maybe this has been already asked, but what's the technology for the back-end you're using?

LUA in an environment outside of DCS. In order to run DCE between the missions, a LUA script in the Simulation Script Engine will launch a bat file through os.execute() at mission end, which will in turn start DCE.

Awesome work!

Please adapt this technology to an "invasion defense" campaign for the Viggen when it's released.

@Andrei; I think he said somewhere that he was using Lua for the back-end, as there already is a Lua interpreter shipped with DCS = No need for additional runtimes or configuration

Edit: Sniped by MBot himself

Awesome work!

Please adapt this technology to an "invasion defense" campaign for the Viggen when it's released.

I am eagerly waiting for LNS's Viggen :)

It might also be worth to mention that I meanwhile managed to integrate my system with the standard campaign system of DCS. This means that DCE campaigns can be started, played and reset just like any other campaign in DCS.

I am eagerly waiting for LNS's Viggen :)

 

 

It might also be worth to mention that I meanwhile managed to integrate my system with the standard campaign system of DCS. This means that DCE campaigns can be started, played and reset just like any other campaign in DCS.

Man, seriously, that's just plain awesome! :thumbup:

Is there much left to do before you can release the first version to public? I can't wait for this! :joystick:

<---- Praying for this engine to work on massive multiplayer PVP.

That's amazing! Fantastic work, look forward to testing it out

Great! Looking forward for this :)

Is there much left to do before you can release the first version to public? I can't wait for this! :joystick:

The big items are slowly coming to completion. Next, I will work on more detailed debriefings. Currently this is just an overall stats evaluation but I also want a section specific to the player’s package and target. Everything should be wrapped in some nice text, so to set this up for all the possible task and target combinations will be a lot of tedious work.

Then I will do a major element that I am looking forward to write: Weather model. I plan to utilize the Dynamic Weather system by DCS instead of the standard weather, which should not only provide diversity but will also greatly simplify any campaign weather model. Dynamic Weather actually consists of very few variables, which will be easy to write into the mission file. I will then have to write some model which controls strength and movement of pressure systems as campaign time progresses, as well as some means for the campaign designer to select on overall weather picture. I will also have to refresh my meteorology knowledge to get a better idea of realistic properties of pressure systems. In addition, as everyone that has used Dynamic Weather knows, getting good results with it is not easy (especially preventing excessive winds). I think this will be a big challenge but I look forward to get this working.

Then of course, there are still a million of small bits and pieces, which have the habit of taking up a lot of time.

I have been thinking about my planned release strategy. Previously I just wanted to get DCE out with some barebone campaign template for people to start working with it, while I develop my own campaigns. I am now getting away from this. For once, DCE has become very complex. There are a lot of variables to set up campaigns which have complex interactions that influence the results you are getting. This will require detailed documentation, which does not exist yet (and which I am not looking forward to write). And even then, campaign designers would probably still need considerable support. On the other hand, I no longer believe that it will be possible to thoroughly test and debug DCE without a “serious” attempt to build a complete campaign. What this means is that after DCE is feature complete, I will build my first campaign and in the process test and polish DCE. When this campaign is released, other campaign designers will have a working example to study and adapt.

It might also be worth to mention that I meanwhile managed to integrate my system with the standard campaign system of DCS. This means that DCE campaigns can be started, played and reset just like any other campaign in DCS.

:shocking::cheer3nc::beer: Man, this will be so awesome to use!!! We're unworthy...:worthy::worthy::worthy:

The big items are slowly coming to completion. Next, I will work on more detailed debriefings. Currently this is just an overall stats evaluation but I also want a section specific to the player’s package and target. Everything should be wrapped in some nice text, so to set this up for all the possible task and target combinations will be a lot of tedious work.

 

Then I will do a major element that I am looking forward to write: Weather model. I plan to utilize the Dynamic Weather system by DCS instead of the standard weather, which should not only provide diversity but will also greatly simplify any campaign weather model. Dynamic Weather actually consists of very few variables, which will be easy to write into the mission file. I will then have to write some model which controls strength and movement of pressure systems as campaign time progresses, as well as some means for the campaign designer to select on overall weather picture. I will also have to refresh my meteorology knowledge to get a better idea of realistic properties of pressure systems. In addition, as everyone that has used Dynamic Weather knows, getting good results with it is not easy (especially preventing excessive winds). I think this will be a big challenge but I look forward to get this working.

 

Then of course, there are still a million of small bits and pieces, which have the habit of taking up a lot of time.

 

 

I have been thinking about my planned release strategy. Previously I just wanted to get DCE out with some barebone campaign template for people to start working with it, while I develop my own campaigns. I am now getting away from this. For once, DCE has become very complex. There are a lot of variables to set up campaigns which have complex interactions that influence the results you are getting. This will require detailed documentation, which does not exist yet (and which I am not looking forward to write). And even then, campaign designers would probably still need considerable support. On the other hand, I no longer believe that it will be possible to thoroughly test and debug DCE without a “serious” attempt to build a complete campaign. What this means is that after DCE is feature complete, I will build my first campaign and in the process test and polish DCE. When this campaign is released, other campaign designers will have a working example to study and adapt.

The amount of work you've to put in there is incredible! I will feel bad to just use this for free. Ever thought of turning this into an official 3rd party DLC of some kind? It might not be the usual campaign DLC, but what you're doing is some really needed pioneering and should be rewarded in some way IMHO. :)