Jump to content

4.268 Routes for vehicles - getting them to stick to the roads.


Gibsonm
 Share

Recommended Posts

Good morning,

 

I have an issue with a recent series of scenarios that are in draft form with scripted units remaining on roads.

 

This is hopefully being looked at based on correspondence with eSim but for the broader audience the issue in the Mission Editor "test mode" is that units appear to often deliberately miss turn offs (i.e. go beyond the way points) and double back, leave the road, etc. (I can post a video for those interested).

 

Given the size of the scenario (Only Div and Bde / Regt Recon plotted to date) it is not a simple issue:

 

723015022_220112Changepaths.thumb.PNG.11f2f5ac01f3106044a3aadf5cb97862.PNG

 

Given the release of 4.268 I thought I'd revisit the issue and it remains.

 

There are three things I thought I'd check while eSim do their research.

 

1. I've tried to confirm they use Navmesh routes throughout (on the basis that that would keep them on the road [given that the route is on the road - cross country travel would seem slower]), but this threw up a follow on question about the interface:

 

I've assumed that:

 

a. This means (option "unticked") that I do not need to hold down the "ALT" key (easier when you are already holding down "'Shift" already) to generate a "navmeshed" route:

 

581926911_220112NoAlt.png.af1447feaabb58f3b49bd9bcb2010ed3.png

 

b. This means (option "ticked") that I do need to hold down the "ALT" key:

 

865530165_220112Alt.png.0b667aab91bde6cba0b036a2f258518d.png

 

Question: Have I been correct in using option "a" on the basis that all routes will be "navmeshed", unless I do hold "ALT" and in those cases they aren't "navmeshed".

 

2. Assuming I haven't messed that up, the next possible area of concern was that the pathing wasn't locked to roads:

 

Individual routes are easy enough to check:

 

259409500_220112Roads.png.b3d6c2b935a180cfe2981890130e54f4.png

 

However I have noticed that I can select numerous routes, right click and choose "Follow roads":

 

1473119476_220112Roads2.png.dd335fa668421df25e31ef9e6ad0bbca.png

 

1018730885_220112Roads3.thumb.png.ddbb72b1e26da8f6c08491b68b163c39.png

 

But then on inspection of some routes within the group (still based on roads) this is not selected?

 

1761113639_220112Roads4.png.a338b2d8d41f95f482ea4108038603fd.png

 

Question: Is there some ability to globally select routes and tell them to "Follow roads" and then go back and adjust the limited number (so far) that aren't on roads?

 

3. The last possible issue is the route tactic selected.

 

Being reconnaissance, I've asked the units to "Scout" (which in turn means I'll need a second layer of routes for the Main Body, using "Engage" or similar). However the manual (p115) includes "Do NOT try to stay on roads" as a scouting unit's behaviour.

 

Question: If indeed the movement away from the road is due to the Scout route tactic (despite the route being locked to the road due to "shift click" and the pathfinding set to "Follow roads"), how do I achieve the required effect (Recce the high speed route, stop if fired upon / enemy spotted, look for minefields, etc.) if "Scout" is not viable?

 

I could use "Engage" with numerous "Retreat back if ..." routes followed by "Embark if ..." routes, effectively tripling the number of routes, but that doesn't help me with the minefields.

 

Note: It maybe a red herring but this is on the new "Bergen-Soltau-Munster" map (with the extra terrain detail) and I've not experienced this issue with earlier maps or scenarios with a similar playable area or force structures.

 

If the team can provide some insight it would be appreciated (perhaps Q1 could be something for the wish list to make this crystal clear?).

 

Thanks.

 

Edited by Gibsonm
Link to comment
Share on other sites

9 hours ago, Gibsonm said:

Question: Have I been correct in using option "a" on the basis that all routes will be "navmeshed", unless I do hold "ALT" and in those cases they aren't "navmeshed".

In case of routes bound to roads it is not relevant, if Alt key is pressed or not, since navmeshes are utilized for off-road routes. Hold Shift or apply 'Follow roads'.

9 hours ago, Gibsonm said:

Question: Is there some ability to globally select routes and tell them to "Follow roads" and then go back and adjust the limited number (so far) that aren't on roads?

There is no  such global option. It is possible to use marquee tool, but it would select everything belonging to given party, and some options for selected objects might not be available. There is no global option to see which routes were edited, and which were not; routes need to be checked manually, one-by-one.

9 hours ago, Gibsonm said:

Question: If indeed the movement away from the road is due to the Scout route tactic (despite the route being locked to the road due to "shift click" and the pathfinding set to "Follow roads"), how do I achieve the required effect (Recce the high speed route, stop if fired upon / enemy spotted, look for minefields, etc.) if "Scout" is not viable?

'Follow routes' just makes route vertices  to snap to nearby road, essentially; it has zero relation to AI behavior defined by tactics. So if unit has scout tactics, it would never ever try to stay on road and if something made it to  deviate- it would just go to next route vertice or waypoint by shortest path. In this context 'Engage'  with  'Retreat back if...' and 'Open fire if...'(or 'Return fire') conditions defined is probably best substitute to a scout tactics, but unfortunately  it would not replace it completely.

Link to comment
Share on other sites

24 minutes ago, Jartsev said:

In case of routes bound to roads it is not relevant, if Alt key is pressed or not, since navmeshes are utilized for off-road routes. Hold Shift or apply 'Follow roads'.

 

Understood, but in the more general case, i.e. not following roads - is my understanding correct?

 

This means (option "unticked") that I do not need to hold down the "ALT" key (easier when you are already holding down "'Shift" already) to generate a "navmeshed" route:

 

581926911_220112NoAlt.png.af1447feaabb58f3b49bd9bcb2010ed3.png

 

Link to comment
Share on other sites

26 minutes ago, Jartsev said:

There is no  such global option. It is possible to use marquee tool, but it would select everything belonging to given party, and some options for selected objects might not be available. There is no global option to see which routes were edited, and which were not; routes need to be checked manually, one-by-one.

 

OK, so I need to check every route all over again????

Link to comment
Share on other sites

37 minutes ago, Jartsev said:

'Follow routes' just makes route vertices  to snap to nearby road, essentially; it has zero relation to AI behavior defined by tactics. So if unit has scout tactics, it would never ever try to stay on road and if something made it to  deviate- it would just go to next route vertice or waypoint by shortest path. In this context 'Engage'  with  'Retreat back if...' and 'Open fire if...'(or 'Return fire') conditions defined is probably best substitute to a scout tactics, but unfortunately  it would not replace it completely.

 

Thanks. :(

 

This scenario now goes in the bin and I'll forget about any large scale future ones with reconnaissance forces proving routes.

 

There is no point in plotting:

 

1. An initial "Engage" route.

2. Then a "Retreat if ..." route.

3. Finally an "Embark if ..." route for every 500m or so, esp. if they are just going to blunder into minefields anyway.

 

Or maybe just ignore "Bergen-Soltau-Munster" and any other new maps that may follow it - since the problem doesn't seem to arise with those (perhaps the AI can cope better with the lesser resolution).

 

@Jartsev I presume this means I should not expect an update to the email I received on 02 JUN where someone called "Vsevolod" was looking into this?

 

Edited by Gibsonm
Link to comment
Share on other sites

7 minutes ago, Gibsonm said:

 

This means (option "unticked") that I do not need to hold down the "ALT" key (easier when you are already holding down "'Shift" already) to generate a "navmeshed" route:

With this option unticked you would not need to hold ALT key at all, e.g.  for off-road routes, or portions of the routes extending outside of the roads navmeshes are going to be used by default

Link to comment
Share on other sites

10 minutes ago, Gibsonm said:

OK, so I need to check every route all over again????

Yes. Plus, it makes sense to inspect them individually for excess vertices which were placed automatically at wrong places(usually in proximity of intersections, bridges etc.). Also, bug #9802 still applies and  it  a good idea  to check if waypoints are not between tracks of  highway ID #5 (small 4x4 vehicle or motorbike  might not be able to reach such waypoint, if tactics forces unit to stay on road).

 

So there is still a big amount of manual labour needed.

Link to comment
Share on other sites

3 minutes ago, Jartsev said:

Plus, it makes sense to inspect them individually for excess vertices which were placed automatically at wrong places(usually in proximity of intersections, bridges etc.). Also, bug #9802 still applies and  it  a good idea  to check if waypoints are not between tracks of  highway ID #5 (small 4x4 vehicle or motorbike  might not be able to reach such waypoint, if tactics forces unit to stay on road).

 

Yes I already did this so that makes it 3 sets of checks and in any case its pointless since they wont use the road.

 

I have no idea what Bug #9802 is.

 

Edited by Gibsonm
Link to comment
Share on other sites

  • Members

Route segments can have both "follow road" and "Navmesh" route legs between vertices (e.g. you can hold Shift to make the unit use the road graph, and then release Shift for follow-on cross-country movement which would then be using the Navmesh mode (assuming that you removed the checkmark from "Hold Alt key to plot Navmesh routes") before right-clicking and thus creating the end waypoint.

 

"Navmesh mode" can be identified by creating a route across a village. If the route remains straight, Navmesh route plotting does not occur. If it's wiggly and avoids buildings or other obstacles not visible on the map, Navmesh logic is active.

 

We're currently working on a Youtube tutorial to cover the subject; a second one is in preparation covering the topic of task forces.

 

13 hours ago, Gibsonm said:

Question: Have I been correct in using option "a" on the basis that all routes will be "navmeshed", unless I do hold "ALT" and in those cases they aren't "navmeshed".

As far as I can see, yes (see above for the test to confirm it).

 

13 hours ago, Gibsonm said:

Question: Is there some ability to globally select routes and tell them to "Follow roads" and then go back and adjust the limited number (so far) that aren't on roads?

No - but to be honest, I don't think this would be necessary or make a difference to begin with. Generally, whenever a platoon is in column formation and the direction to the next route vertex does not deviate more than +/-30° from the direction of the road, the platoon will stay on the road anyway. So it is quite possible to use both Navmesh intelligence in combination with road-bound movement, but unfortunately you'd have to forego the convenience of holding the Shift key for long distance routes between waypoints (a practice that I have never recommended to begin with). The prerequisite is that the unit is in column formation (duh).

In version 4.1/4.2 units still don't have a way to anticipate sharp turns on a road. They won't slow down if moving at high speed before it's too late and are therefore alsmot guaranteed to overshoot a turn. We've been working on a number of improvements in the 4.3 branch, so coming June a scenario that might not work too well right now for this reason alone might actually behave better (but hard to say without looking at the specific case). So, one way to avoid trouble is to have a waypoint some 50m in front of a sharp turn, then make the turn with a slow route and narrow spacing, and resume normal speed and distances a while later. In an ideal world the units would do this automatically, I agree, but we have to work with how it is right now.

 

13 hours ago, Gibsonm said:

However the manual (p115) includes "Do NOT try to stay on roads" as a scouting unit's behaviour.

 

Question: If indeed the movement away from the road is due to the Scout route tactic (despite the route being locked to the road due to "shift click" and the pathfinding set to "Follow roads"), how do I achieve the required effect (Recce the high speed route, stop if fired upon / enemy spotted, look for minefields, etc.) if "Scout" is not viable?

The entry in the manual refers to the unit in column formation and following the direction of the road +/-30° rule that I mentioned above. If the chosen tactic is "scout", they would follow the route more directly. But if you "Shift-nail" the route to a road, then they won't attempt to drive off-road like water droplets repelled from a greased surface. It's just that they won't "feel attracted" to the tarmac.

So, choosing "Scout" tactic is perfectly viable even with road-bound movement. Turning around on a narrow road to break contact with the enemy may of course not work too well, nor is there a guarantee that minefields will be discovered while traveling at elevated speeds (other than by blowing up, the traditional and reliable alternative method). But of course you're well aware of the trade-offs.

 

13 hours ago, Gibsonm said:

Note: It maybe a red herring but this is on the new "Bergen-Soltau-Munster" map (with the extra terrain detail) and I've not experienced this issue with earlier maps or scenarios with a similar playable area or force structures.

It may still be worth our programmers' time checking selected situations with a modified version of your scenario. If you can identify specific locations and have units start right in their vicinity, and delete everything else from the scenario, then this would make an excellent test case for our pathfinding specialist to see what else he can do to improve behavior. We would then use that scenario also as an automated test case that gets routinely tested with every new build that has the potential to change AI behavior, so we detect unwelcome changes well before even the beta testers get to see the changes.

Link to comment
Share on other sites

7 hours ago, Ssnake said:

The entry in the manual refers to the unit in column formation and following the direction of the road +/-30° rule that I mentioned above. If the chosen tactic is "scout", they would follow the route more directly. But if you "Shift-nail" the route to a road, then they won't attempt to drive off-road like water droplets repelled from a greased surface. It's just that they won't "feel attracted" to the tarmac.

So, choosing "Scout" tactic is perfectly viable even with road-bound movement. Turning around on a narrow road to break contact with the enemy may of course not work too well, nor is there a guarantee that minefields will be discovered while traveling at elevated speeds (other than by blowing up, the traditional and reliable alternative method). But of course you're well aware of the trade-offs.

 

Ah you taunt me. ;)

 

Just when I resign myself to giving up, you dangle the possibility of them staying on the roads after all. :)

 

They are travelling at the default Scout movement rate of "slow" so I'm happy that the AI will do its best to react appropriately.

 

I'm also happy that "likely" does not mean "guaranteed to" and as in real life, they may not spot the minefields (esp. if they are not fenced, etc.).

 

Edited by Gibsonm
Link to comment
Share on other sites

7 hours ago, Ssnake said:

It may still be worth our programmers' time checking selected situations with a modified version of your scenario. If you can identify specific locations and have units start right in their vicinity, and delete everything else from the scenario, then this would make an excellent test case for our pathfinding specialist to see what else he can do to improve behavior. We would then use that scenario also as an automated test case that gets routinely tested with every new build that has the potential to change AI behavior, so we detect unwelcome changes well before even the beta testers get to see the changes.

 

Ah I thought I'd done that back in June but I'll recheck and if not I'll attach it here.

Link to comment
Share on other sites

I had a look and there is an email sent by me on 06 JUN 21 entitled "Request for help with routes in 4.2X - Part 1 of 2"

 

That included links to:

 
1. The original scenario:
 
2. The original scenario with all the units in the margin (on blue and red) removed. That is only the recon units on map will spawn. I've left one Blue unit on to facilitate testing but it is well away from the routes:
 
3. A third version with only that specific motorcycle (no certainty it will take that route though).
 
4.  A video of Div Recon / 033's progress.
 
The good news is that it makes it all the way. The bad news is it quite often does loops and wanders off the roads (seems to be worse on the autobahn).
 
At one point it appears stuck in a loop so I went to F5 to work out where it was on the map, when I returned it had resolved itself:
 
5: AAR file.
 
I'm pretty sure that one had the unit using a "March" route tactic throughout.
 
I'll send new links to the scenario using a "Scout" route tactic, since that's what we are aiming for.
 
Edited by Gibsonm
Link to comment
Share on other sites

  • Members

OK, thanks, I'm back on track with this one and will check what, specifically, has been done about your case in the last months (and what else we can do in the coming months). Anything that we can do to improve AI behavior in your case is likely to help everybody else, too, so it's great to have specific cases of failure that we can use as a reference case.

Link to comment
Share on other sites

Just a bit of an update.

 

400+ routes checked, about 150 to go.

 

All that are meant to be on a road / autobahn have been "shift-clicked", with column formation (although the first units are individuals) and their route pathfinding set to "follow roads".

 

Once checked I'll remove all the extraneous Blue/Green and Red units and run it again (just in case by some miracle it works this time) and either way post:

 

1. The scenario

2. An AAR

3. A video if I see this going weird, or

4 Screenshots from the AAR if I don't witness it (there are 18 M/cycles on various routes so this should provide a good sample size - although not every route with be traversed).

 

Link to comment
Share on other sites

  • Members

We think that it's particularly the light wheeled vehicles that are affected (partly because they have no pivot steer; once that they are off the road and the sides are lined with trees, they rarely make it back on the road). Then of course there's the issue with two (near) parallel roads very close together, which of the two should be taken. The road graph contains no height information, so when choosing (even just for a moment) a point of a track running at the foot of an embankment, off you go and never come back.

I suppose, if we can mitigate the two issues it might eliminate a large portion of failure cases.

Link to comment
Share on other sites

11 minutes ago, Ssnake said:

We think that it's particularly the light wheeled vehicles that are affected (partly because they have no pivot steer; once that they are off the road and the sides are lined with trees, they rarely make it back on the road). Then of course there's the issue with two (near) parallel roads very close together, which of the two should be taken. The road graph contains no height information, so when choosing (even just for a moment) a point of a track running at the foot of an embankment, off you go and never come back.

I suppose, if we can mitigate the two issues it might eliminate a large portion of failure cases.

 

If you like I can include some BRDM 2s from the Regt Recon (they turn up a while after the Div Recon) if you want to look at 4x4 vehicles (at 7t, unsure if they still sit in the "light wheeled vehicle" category)?

 

Edited by Gibsonm
Link to comment
Share on other sites

On 1/14/2022 at 7:56 PM, Ssnake said:

Well, it's wheeled, comparatively fast on roads (so tends to overshoot sharp turns if not slowed down), and can't pivot steer - so it still falls into that category.

 

Ah OK.

 

All of the "Scout" routes are using the default behaviour (apart from Column formation) including slow movement so hopefully overshooting will be limited.

 

The route pathfinding varies between "Follow-roads"" and "Cross country (vehicles)" as appropriate.

 

Where there are Bridge crossings, I've reverted to "March" (column, slow) with "Retreat back if ..." and "Embark if ..." just to avoid any issues with the road/bridge areas.

 

I might run a small number of BRDMs (say 4).

Edited by Gibsonm
Link to comment
Share on other sites

Start state:

 

1303351015_220116RoutetestStartState.thumb.png.4ed30816e25f0b7886379b5d3cbdc0b4.png

 

Div Recon consists of 6 Motor Cycles and 12 BRDM2 (18 in total), spread across the various routes.

 

95 min behind them is the Regt Recon (M/Cycles, BRDM2, PT-76 and a couple of BMP2) which spawn at 95mins, 100mins and 105mins. I can reduce this gap if we need to do additional testing on the "heavier" vehicles.

 

1996540472_220116RoutetestStartStatedetailed.thumb.png.9a51363cae73daac372fcc7d35112559.png

 

Will post an end state screen shoot, ARR etc. once the first run is complete.

 

Edited by Gibsonm
Link to comment
Share on other sites

Just had a quick look (I haven't checked every unit on every route - yet) at the 50 minute mark.

 

BRDM decides on a more adventurous way to "cross" a bridge.

 

Routed with a March command to remain on the autobahn and cross the bridge:

 

1567378751_220116BridgeRoute.thumb.png.d68dec76c89f7a0b2b950b174086458a.png

 

Decides "nope, orders are for losers" and for sh*ts and giggles determines the best way is to go under the bridge and climb the abutment (with a few I guess of eventually smashing up through the road bed?

 

SS_09_29_53.thumb.jpg.2478a494aa0424e7aeec614cbb69c550.jpg

 

Until I see the AAR I wont know if he left the road before the bridge or crossed the bridge and then doubled back.

 

Note: This is not using a "Scout" command with the discussions about staying on/near the road. This is a "March" (stay on the road) command.

 

Perhaps there is an issue with where the bridge and the road "join". I also don't know if a different unit used this same route successfully or not. I'll have to check in the AAR.

 

Edited by Gibsonm
Link to comment
Share on other sites

First run through is complete (a mere 4 hours and 50 mins)

 

Link to AAR: https://u.pcloud.link/publink/show?code=XZM9HGXZKMAPn9QLeWbhms1yE7xInke1rGgk

 

Known issue areas (i.e. where vehicles ran off the road or did not for some reason proceed - numbered 1 - 7):

 

1427466888_220116Issues.thumb.png.66d51c78accbdfda547333ff34a96060.png

 

General Observations:

 

1. It appears (I need to confirm this by reviewing each unit) that running the session in normal time, does not replicate the issues seen with the "Scout" command when running it in accelerated mode.

 

A video of what happened with a M/Cycle when the test was run in accelerated mode (admittedly this test was recorded prior to the 100% review, but I doubt I messed up so many routes / waypoints): https://u.pcloud.link/publink/show?code=XZIYK9XZtnF7gj39DEQX8CmOXyOr2h3mCI0k

 

2. It seems from the points identified above that all are units on March routes, none on Scout?

 

Detailed Observations (prior to reviewing each unit's progress to lack thereof):

 

1. Waypoint 308 - 309: March Route, pinned (shift click) to road, slow speed, column formation, follow road pathfinding, It has a Retreat Back route if under either direct or indirect fire. It has an Embark route if not under either direct or indirect fire and a 3 minute delay:

 

 

1390823468_220116Pt1routeonmap.thumb.jpg.6b927b7b0473ba43038e02f36a2ef180.jpg

 

 

2005910950_1.220116BridgeRoute.thumb.png.8059f0f30634c2dc37046fb3fc2fc4fd.png

 

 

488463280_1.220116Pt1whereitendsup.thumb.jpg.36275530bef951151bc77c1840e5cf5d.jpg

 

Link to relevant AAR video segment:  https://u.pcloud.link/publink/show?code=XZDdHGXZdRuTJyr339XS3oFCJ805yB7dRrdy

 

No other vehicle takes this route.

 

2. Waypoint 387 - 388: March Route, pinned (shift click) to road, slow speed, column formation, follow road pathfinding, It has a Retreat Back route if under either direct or indirect fire. It has an Embark route if not under either direct or indirect fire and a 3 minute delay:

 

1525311680_2.220116Pt2routeonmap.thumb.png.27c9dbb5cc9bc77effeaa32dd656fd1d.png

 

1954923804_2.220116Pt23D.thumb.jpg.c125d320402d2758e19f60e95a875308.jpg

 

Some vehicles continued on after being delayed on this route.

 

The PT-76's were happy to proceed after doing a 360 spin around the waypoint (unsure if their better gap crossing ability helped keep them "on" the bridge):

 

https://u.pcloud.link/publink/show?code=XZbyzGXZblfnlIaD9LB9wfA4ci5u6kdVda7y

 

3. Waypoint 390 - 391: March Route, pinned (shift click) to road, slow speed, column formation, follow road pathfinding, It has a Retreat Back route if under either direct or indirect fire. It has an Embark route if not under either direct or indirect fire and a 3 minute delay:

 

Some units survived location 2, only to be caught here.

 

Again it seems the vehicles do 360 degree laps around the initial waypoint prior to embarking on the route, and again at the end and it seems that for some these circuits are where they get caught:

 

https://u.pcloud.link/publink/show?code=XZWLzGXZLEbwC18DdcjXQpKFkiUoLH2avikX

 

4. and 5. Waypoints 413 - 414 and Waypoint 491 - 415: March Route, pinned (shift click) to road, slow speed, column formation, follow road pathfinding, It has a Retreat Back route if under either direct or indirect fire. It has an Embark route if not under either direct or indirect fire and a 3 minute delay:

 

Route on Map:

 

974120259_4.Pathing.thumb.jpg.de352ce3c513f00120a41d58086786ca.jpg

 

Vehicles doing laps on Northern Bridge and numerous stuck on Southern one:

 

935571216_4.Vehiclesdoingloops.thumb.jpg.9158c1f4f6aad48f14c7d14d33cb656b.jpg

 

End of AAR only one now "'stuck":

 

1937102614_4.Onestuckatend.thumb.jpg.517a1b94bc5c914dffd6fc3e1fa0d661.jpg

 

The video covers from time 06:47:06 until 08:12:20 (recommend you FF as required) shows the "360 degree"' lap at both bridges (conducted by BRDM2s, BMP2s and PT-76s) and some vehicles freeing themselves.

 

Video: https://u.pcloud.link/publink/show?code=XZ9TYGXZEz9FAhVQDbbsx6h4XDIcBYLeAkMy

 

6. Waypoint 433: March Route. BRDM2 stops for no apparent reason and fails to proceed (nothing in either the Embark if ... or Retreat back if ... conditions should prevent it from continuing on).

 

7. Waypoint 439: March Route. BRDM2 stops for no apparent reason and fails to proceed (nothing in either the Embark if ... or Retreat back if ... conditions should prevent it from continuing on).

 

First thoughts:

 

One thing I did think of is that I habitually remove "rough spots" on a route using "Control + click" in order to straighten out the route:

 

For the first location (remainder yet to be checked):

 

Original "shift click" March route (note the 90 degree turns on both sides of the bridge and using the "wrong" bridge span compared to the other routes):

 

282068329_1.220116Rawroute.thumb.jpg.b012d0f1b387edb39246c860414fe111.jpg

 

Cleaned up with "Control click" (single route on "correct" side of bridge, but when test conducted, unit leaves road):

 

485296990_1.220116cleanedroute.thumb.jpg.abf7d8bbdab346857b1a2db26ac7e5d4.jpg

 

Question: Whilst counter intuitive, to me at least, should I just not clean up these routes?

 

Edited by Gibsonm
Link to comment
Share on other sites

  • Members
5 hours ago, Gibsonm said:

1. It appears (I need to confirm this by reviewing each unit) that running the session in normal time, does not replicate the issues seen with the "Scout" command when running it in accelerated mode.

YES.

Time acceleration can lead to abnormal results with units traveling on roads in restrictive or complicated terrain (yes, overpasses count as "complicated" in this context). Off-road travel is less affected.

Link to comment
Share on other sites

  • Members
5 hours ago, Gibsonm said:

Question: Whilst counter intuitive, to me at least, should I just not clean up these routes?

I'm also in the "clean up" camp but to be perfectly honest, I really can't say with certainty if this makes a difference other than looking cleaner to the human eye. Units traveling on routes created with the Shift+click method will clear the route vertices before they even reach them (once that they are "near") and then follow the direction towards the next vertex (and stay on the road, if the directions roughly align and the unit is in column formation and not in Sout or Breach mode). So, small "kinks" that go 90° left or right to the general direction of travel will get removed even before the unit comes close enough to a 30° direction deviation, and the vertex after that is typically straight ahead again, so "in theory" this shouldn't result in erratic behavior (except that in practice, it does).

This is one of the things that need to be investigated in the debugger to see where, exactly, things go wrong. Once that the unit is off the road and/or on a steep slope/inside forested area (especially in time acceleration) a recovery from is very unlikely.

Link to comment
Share on other sites

Hi Gibsonm and Ssnake,

 First of all: happy new year!

Now to the problem:

In Germany and thus on the used map "Bergen-Soltau-Munster" is legal traffic on the right side!

On this map I usually did not use the briges #448 (210 tonnes) for highways, but two bridges #447 (105 tonnes) each.

Therefore you should always take the right-hand one in the direction of travel when using the bridges. This also applies when delaying!

Otherwise, the vehicles fall off!

best regards

Abraxas

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...