Question about cover

[jonp]

All,

How do we know what cover our men are in? I know in the original CM you could tell what type of terrain you were in (at least I think you could, but I haven't played it in ages).

Thx

jonpfl

[Secondbrooks]

umm... By looking what cover is there when men are lying down? Single squad can have various types of cover where it's men are lying in (3 guys in trench, 3 in crater, 2 in open and 1 got stuck into house 200 meters behind). Only way to know where they are is to look at each men.

[YankeeDog]

Ya, it's pretty much WYSIWYG. Unlike CMx1, where terrain was abstracted and each point on the map had a "cover rating", in CMx2 one bush = one bush. A soldier is either behind said bush, or he's not.

In general, in CMx2 it's more important to look for terrain folds that provide defilade and fighting crests. This is far better cover than bushes and trees, which don't stop bullets.

Cheers,

YD

[Dirtweasle]

Ummm, the problem is it's not always apparent.

Setting aside the issues with 1:1 over the unseen (hidden?) action point grid it's still at times very difficult to tell visually just where the tree line actually ends, and where to place the final waypoint to have your squad / teams stop and form up in, (and add uncertainty surrounding if the they will stop and not expose themselves.)

Resolving cover and - really concealment might be the better word - could stand some sort of enhanced user feedback.

[Childress]

Great minds think alike. BFC and/or Scenario designers might consider adding indicative underlying terrain to forested areas. Just like CM1. Yes, I realize that flora are not abstracted in CMSF, still.... Better yet would be diaphanous trees a la MS's Age of Empires. But that's asking a lot of a small team.

[Dirtweasle]

Yea using an underlying terrain type is a decent work around, so long as the map maker / scenario designer take the time.

A permanent solution though, maybe something like where when you hold the mouse on a soldier or it's name it provides that information. You know what I mean? Maybe a little pop up balloon or box with that detail?

[YankeeDog]

Originally posted by Dirtweasle:

Ummm, the problem is it's not always apparent.

Setting aside the issues with 1:1 over the unseen (hidden?) action point grid it's still at times very difficult to tell visually just where the tree line actually ends, and where to place the final waypoint to have your squad / teams stop and form up in, (and add uncertainty surrounding if the they will stop and not expose themselves.)

Resolving cover and - really concealment might be the better word - could stand some sort of enhanced user feedback.

Well, the OP asked how to tell what cover your units *are* in, not what cover they *will be* in if you put a movement waypoint at a certain location.

The former is very simple; if a soldier is behind a certain tree, he has cover from that tree. If not, he doesn't.

But as you note, it can be hard to tell where best to place your waypoints, and how to use orders such as Face and Cover Arc, so that a squad takes best advantage of certain types of cover, such as a small stand of trees. What looks like a reasonable location for a squad can end up with half the squad hanging out in the open sometimes, and I suspect you're right that the reason has something to do with the "Action Spot" grid, and how the AI organizes the exact location of teams and individual soldiers based on this grid.

In short, the problem is not with being able to see what is, but rather being able to make a reasonably inference as to what will be. . .

Cheers,

YD

[Dirtweasle]

It's both actually.

[YankeeDog]

How so?

Personally, I haven't noticed any trouble telling what kind of cover my soldiers are *in*. When they're behind a tree or whatever, by all appearances and experience they seem to have good cover. When they're out in the open, they don't.

But I do sometimes have trouble figuring out what kind of cover they *will* be in, if I order them to move to a certain spot. Sometimes, they use the trees, bushes, barrels etc. as I expect, sometimes not.

How does your experience differ?

Cheers,

YD

[thewood]

There was a long thread on cover back around a month after release started by adam1, IIRC. I am pretty sure Steve said that some cover is abstracted still. So it is not completely WYSIWYG, IMO.

[Secondbrooks]

Trees and brushes (and something which i maybe have forgotten) aren't drawn to distance... My kinda guys with poor computers can see trees just at very close, 200-300 meters, distance if not having trees disabled. I've wondered few times why my unit in firesupport doesn't seem to fire...

Thewood Cover issues have been tweaked after that thread... If it's the same thread where i "published" some my limited tests about LOS blocking and cover from bullets which certain types of terrain seemed to offer (in bad english naturally), i think CMSF was yet in 1.04 state back then... 1.05 changed cover things quite much if i remeber correctly.

[YankeeDog]

Originally posted by thewood:

There was a long thread on cover back around a month after release started by adam1, IIRC. I am pretty sure Steve said that some cover is abstracted still. So it is not completely WYSIWYG, IMO.

Ah. Yeah, I see your point. I mean, I've always assumed that a soldier lying out in the middle of a perfectly flat asphalt road is worse off than a soldier in the middle of a flat, grassy field or a rocky waste. In the latter two cases, at least there would be some "micro cover".

But I've never had much of a problem eyeballing it, and the results seem realistic so I have no complaints in this regard.

Where I see problems is when a few members of a squad decide to park themselves out on naked asphalt, when I expected the entire squad to be able to find cover behind a group of barrels and shipping pallets where I placed the waypoint.

Doesn't happen often enough to be a game-killer for me, but happens often enough to be annoying.

Cheers,

YD

[Dirtweasle]

Originally posted by YankeeDog:

How so?

Personally, I haven't noticed any trouble telling what kind of cover my soldiers are *in*.

...

How does your experience differ?

Cheers,

YD

I've had several occasions where I have a unit at or near the edge of some trees that I cannot tell for sure if a few of the soldiers are actually in the trees or sticking out. Trees mainly. ...er, rather predominantly or maybe even exclusively now that I think about it.

This is again once they are situated at or near the tree-line. As separate matter from the issue of precicting how exactly or the squad will fan out and maybe extend beyond the cover / concealment you intend them to use.

Both are aggravations.

[Dirtweasle]

Originally posted by YankeeDog:

...Doesn't happen often enough to be a game-killer for me, but happens often enough to be annoying.

Cheers,

YD

Aye, it's an annoyance and not nearly as much of a issue as the pathfinding thing, though I suspect they are somewhat related matters.

[YankeeDog]

You can't just look to see whether they're behind a tree trunk and/or in the shadows of the branches or not? So far, this has worked fine for me: if the soldier isn't behind a tree trunk, he doesn't have cover. If he's not under the shadow of the branches, he doesn't have concealment.

Worked fine for me so far. But I must admit I haven't done any extensive testing on the issue. :confused:

Cheers,

YD

[Dirtweasle]

Well, it depends on the angle.

Let's back up though, as maybe I'm operating under a false assumption or somefink.

Are you sure the trees, and I mean each and every branch and trunk, are modeled both for purposes of cover and concealment?

Or is it just from the action point? So that if there is a tree between (or not) one action point or another it doesn't matter that PFC Jones has his fanny hanging out in the breeze?

[thewood]

Originally posted by YankeeDog:

You can't just look to see whether they're behind a tree trunk and/or in the shadows of the branches or not? So far, this has worked fine for me: if the soldier isn't behind a tree trunk, he doesn't have cover. If he's not under the shadow of the branches, he doesn't have concealment.

Worked fine for me so far. But I must admit I haven't done any extensive testing on the issue. :confused:

Cheers,

YD

Maybe the question should be if he is not behind a tree trunk, but in woods, does he get any cover from undulations in the ground, etc.

[slug88]

Dirtweasle, my observations, which I have made by scanning the targeting line across trees and various cover, lead me to believe that the individual tree trunks are what provide cover/concealment, rather than the overal terrain composition of the action-spot.

[slug88]

Furthermore, if you try firing ATGM's or RPG's with intervening trees, you'll probably note that about half the time they fail to reach their target, it's because they fly right into an obstructing tree; otherwise they continue unimpeded. I'd imagine it's the same for all projectiles, be they bullets, guided missiles, or anything in between.

[ April 22, 2008, 11:49 AM: Message edited by: slug88 ]

[slug88]

Also, tangentially, the above reminds me of another issue that I think should be looked at, which is the durability of trees. I know that they are fully destructable, but each time I've seen it happen it took an overwhelming amount of firepower. I once had a Bradley expend at least a couple dozen cannon rounds on a tree, trying to hit a target far behind it, before the tree finally disintegrated. I'd think that for a medium sized tree, just one or two 25mm rounds through the trunk would do the job. Am I wrong?

[Dirtweasle]

Right, but that's LOF not LOS and I recall that LOF, trajectory and ballistics and such are handled different than sighting.

If I understand it sighting or LOS checks are done from action spot to action spot, (on this what is it 6m x 6m grid?) not from PFC Jones eyeballs to any given point on the map, correct? So you have cases where PFC Jone may look to the player to be in cover (or not) when looking over PFC Syrian National eyes at him or vis-a-versus at his position but since the LOS check is not done that way, it leads to certain anomalies, things happening that from the players eye should not.

This disconnect between what is being observed (1:1 modeling) and what is happening under the hood (the grid) a sort of reality on top of an abstraction leads to some odd situations. Like the Bradley thing.

[slug88]

Well, if you agree that LOF is modelled as I described above, then *cover* would work in the most optimal way; that is, if it looks like PFC Syrian can hit PFC Jones, then he will be able to every time. (This disregards the awknowleded abstraction of 'micro-cover', which I suppose could be refined to give the player more feedback.) If indeed the system goes by action spots for LOS but true trajectories for LOF, then it is only *concealment* which could lead to discrepencies. Cover would always be WYSIWYG. And in my experience, for the most part it is.

[YankeeDog]

Originally posted by Dirtweasle:

Right, but that's LOF not LOS and I recall that LOF, trajectory and ballistics and such are handled different than sighting.

If I understand it sighting or LOS checks are done from action spot to action spot, (on this what is it 6m x 6m grid?) not from PFC Jones eyeballs to any given point on the map, correct? So you have cases where PFC Jone may look to the player to be in cover (or not) when looking over PFC Syrian National eyes at him or vis-a-versus at his position but since the LOS check is not done that way, it leads to certain anomalies, things happening that from the players eye should not.

This disconnect between what is being observed (1:1 modeling) and what is happening under the hood (the grid) a sort of reality on top of an abstraction leads to some odd situations. Like the Bradley thing.

I don't think this is quite accurate; with the new "sub-action grid" added with 1.07, I believe LOS checks are actually done on a more refined basis than simply action spot to action spot; individual soldiers do spot things separately, and are spotted as individuals. However, I think communication of spotting info is instantaneous within a unit. IOW, there is still "borg spotting" within an individual squad or team.

I must admit I don't really understand all the details, but I think Action Spot LOS checks are used more as a pre-check to cut down on computing load. IOW, the game engine first checks whether it is theoretically possible for something in Action Spot A to see something in Action Spot B. If the answer is "no", then it stops the process and doesn't waste any more computing cycles. If the answer is "yes", then it gets down to the nitty gritty of checking who sees what.

But I could be wrong about this. . . just my layperson's attempt to grok what's going on under the hood.

Cheers,

YD

[Dirtweasle]

. . just my layperson's attempt to grok what's going on under the hood.

Hey, mine too.

[Combatintman]

I admit it isn't a showstopper for me but agree that if it could be implemented CMX1 style and for waypointing without too much effort it would make me happier.