Odd Ground Movement Routes

[BS8th_Bulau]

In another thread, I mentioned some problems I was having in Hour 1 of a campaign, with fuel trucks running out of gas after only a short distance, and appearing to go the wrong way, or the long way, on the Ostfriesland map.

I had an opportunity to view the Hour 2 mission file in the FMB, and also run it from there to observe vehicle movement. Some of my vehicles are moving pretty much as I intended, however, many are taking very odd routes, going the long way around, taking huge detours, or going opposite direction, then back again. Also, if there are trees in the way, they mill around forever trying to find a way around or through, and there are a LOT of trees on this map. I can see now, why my trucks ran out of gas in first mission, as they must have tried to find a way out of the airport and onto the nearby road, and ended up going on some sort of "Walkabout".

I am wondering if this is a problem with the "Relaxed" mode of movement we are using, and perhaps we should have stuck with "Normal". Can these problems be fixed by just placing LOTS of waypoints very close together for my ground movements?

I also noticed, that those trucks which are out of gas, and "In Progress" in the MP, do not appear in the subsequent coop mis file. Not even a static version appears to mark their location. Is that normal?

Yes, we are still using the older v4.360 MP.

[IV/JG7_4Shades]

Hi Bulau,

I take these sorts of reports pretty seriously, because ground movement is of central importance to running an SEOW campaign. I also have a report of some ground movement anomalies from 22GCT on the Crimea map.

For background, here is how SEOW handles ground movement. The commander specifies waypoints in the MP. The MP simply collects these and stores them in the DB. The MP does not do ANY sanity checking of commander-entered ground waypoints at all, regardless of the CGM moed being used in the campaign. At mission Build time, the DCS reads the waypoints from the DB and also reads the entire Highways and Highway_Intersections data for the sector. It is the job of the DCS to "map" the commander's intended waypoints onto the actual Highway network, using the CGM Mode as a kind of ruleset. The "DCS Road Code" is not perfect, it can still generate wierd routes on occasion (just as the FMB does!).

In my view, the "Normal" CGM Mode is best, because it limits the amount of off-road excursions, allowing an excursion only at the LAST waypoint of the movement order. But that is not to say that it always finds the best route. Every mode, except "Simple" can produce poor or incomplete route to the destination. Commanders can assist the Road Code by judicious choice of waypoints along the intended route, i.e. just before and just after junctions.

If places are found on a map that always produce poor results, please report the location in detail. That's the only way we can fix them up. Putting the Highways table together is a massive and complicated effort by many people - there are bound to be mistakes in there.

Now, about trucks running out of gas after 1 hour of a campaign. That is most likely nothing to do with the generated road route, since the truck can only move at most 22 km in an hour, which is around 10-15% of its total range. If you are using DCS 3.1.3 (prior versions had a bug in supply consumption rates), the main things that determine supply consumption rates are the Initial Fuel Load of your truck, the Sector_Movement_Cost for your map and season, and the Global Fuel Consumption rate.

I also noticed, that those trucks which are out of gas, and "In Progress" in the MP, do not appear in the subsequent coop mis file. Not even a static version appears to mark their location. Is that normal?

Hmm, could be a bug - definitely not intended to be like that!

Cheers,
4Shades

[BS8th_Bulau]

I think the problems I'm having are due to a combination of several factors:

1. Mistakes on my part, or ignorance of how the route mapping works. I think on the second try, still under the Relaxed mode, I was placing too many waypoints. Idealy, I would like to place very few points, start and end, and have the route follow the roads in the most direct way. I would be nice to see some confirmation in the MP of the route which will be generated, but I understand now it doesn't work that way at this time.

2. Coarse granularity of the MP. Placing a waypoint with the MP can result in errors of 3km or more. With the relaxed mode, any unnecesaary waypoint I place resulted in an excursion off road to the waypoint and back.

3. The usual idiosyncrasies of the game with regard to route mapping, as you have alluded to. The "milling around" of ground pieces in response to barriers is an IL-2 thing and nothing to do with SEOW. I'm curious, then, how SEOW works with that. If, in the game, a tank mills around for the entire hour, and never gets out of Dodge, does it end up there in the next coop, or does the DCS assume it went one hours worth along it's intended route, and place it accordingly?

4. Quite possibly, some errors in the road network on this map. I will try to test this in the weeks ahead see if that is the case. In the first try, trucks from Enshede had only the start and end waypoints. The routed they took apparently involved some very large detour. Similarly the trucks from Groningen only went about 5 km and were shown as out of fuel, and they were missing in the subsequent coop generated. Some other ground movements in other areas of this map (Ostfriesland) also resulted in wrong way excursions on the order of 2 map grids (20km).

[IV/JG7_4Shades]

or does the DCS assume it went one hours worth along it's intended route, and place it accordingly?

Oleg has refused several requests over the years to place in the log file all unit location coordinates at the end of the mission. This means that SEOW has no way of knowing where units actually are at the end of the mission. So SEOW just *assumes* that units follow their courses. Therefore those instances where the units repeatedly bump into trees or buildings really have no effect on movement from mission to mission. That's a good thing.

The bad thing is that units are "teleported" from where they actually end the mission to where they should have ended the mission, sometimes avoiding combat. However, if units spend most of their time on roads (e.g. "Normal" and "Strict" CGM Modes) then this issue doesn't really impact the campaign to any great extent. Traffic jams on roads are similar - too hard to avoid by coding and the units will achieve their destinations at the end of the mission anyway.

Cheers,
4Shades

[BS8th_Bulau]

Aaaah! A faint glimmer of light at the end of the tunnel!

Yes, the "teleporting" is a two-edged sword. Knowing that, however, means I can plan movements with more confidence now. It also means, some tactics I employed on my template, to block roads with an object placed squarely on it, wouldn't actually work in SEOW, at least not exactly so.

Your mention of "avoiding combat" raises another question: If RED and BLUE movements that pass each other enroute are "teleported" in SEOW, but in the actual coop that plays out, they meet enroute, duke it out, and suffer some losses, are the teleported units still penalized with those losses? Or does SEOW assume they never met, and reach their final waypoints intact?

I'm also wondering about this because certain objects, like artillery pieces, for example, which ordinarily could shoot, are replaced by a truck during the movement phase in the generated coop. A truck, in IL-2, is essentially a sitting duck without any defenses.

[EJGr.Ost_Chamel]

Your mention of "avoiding combat" raises another question: If RED and BLUE movements that pass each other enroute are "teleported" in SEOW, but in the actual coop that plays out, they meet enroute, duke it out, and suffer some losses, are the teleported units still penalized with those losses? Or does SEOW assume they never met, and reach their final waypoints intact?

"Teleportation" only occurs, when the SEOW-system has no other information at hand, about the state of a unit at the end of a mission. It then has no other choice but to estimate the position according to the orders.
But as soon as a unit is destroyed, IL-2 writes this info into the logfile and the DCS can then correctly delete the unit from the database.

I'm also wondering about this because certain objects, like artillery pieces, for example, which ordinarily could shoot, are replaced by a truck during the movement phase in the generated coop. A truck, in IL-2, is essentially a sitting duck without any defenses.

Yes, and that is what artillery in transit was/is in reality. As soon as a battery of guns is towed by a truck/halftrack, it can't shoot and it takes quite a while, to bring it back into operational status.

Greetings

Chamel