LOS / LOF problems

[Martin Krejcirik]

Steve, one last thing to clarify - is there something more abstracted (less detailed) in CMx2 then in CMx1 ?

[Truppenfuhrung]

If we take the following screenshot:

1-Bug ?

2-Abstraction ?

3-The shot simply come over the wall from a unit on a higher ground ?

---------------

If we take the following screenshot:

1-Bug ?

2-Abstraction ?

---------------

If we take the following screenshot:

1-Bug ?

2-Abstraction ?

[Battlefront.com]

Bug

Bug

Bug

Steve

[Truppenfuhrung]

Originally posted by Battlefront.com:

Bug

Bug

Bug

Steve

Thanks for the answers !!!

And now I hope that all the whining about the abstraction will stop !

[dalem]

Originally posted by Truppenfuhrung:

</font><blockquote>quote:</font><hr />Originally posted by Battlefront.com:

Bug

Bug

Bug

Steve

Thanks for the answers !!!

And now I hope that all the whining about the abstraction will stop ! </font>

[Truppenfuhrung]

Originally posted by dalem:

Look sweetie, not all open discussion is "whining", hmm?

-dale

And blaming the abstraction when abstraction is not involve is a useless open discussion.

[Battlefront.com]

Dale,

First, thanks for that explanation - it helps me get my head around what is going on.

Second, my only point with the "hexes" comment was literally to draw a parallel to an abstraction mechanic. I guess where I'm having a hard time (and maybe it's none of my business) is how to reconcile where the LOS and LOF points originate and terminate.

You're welcome and I do understand. This is the problem with going with 1:1 for sure. And that is it looks like everything is simulating in 1:1 detail, but that is impossible to do so some things are abstracted.

Philistine,

AIUI, CM1's level of graphics abstraction was higher then the combat system. It seems with CMSF, it goes the other way.

Overall, no. A Sherman in CMBO was a point on a data grid that was linked to a table of data. When a shot was fired at it very detailed equations approximated that tank's 3D orientation and abstractly decided where the hit would strike, if at all. From that point on a fairly non-abstract system determined the hit's results.

In CM:SF a T-62 is actually there in the 3D environment, detailed in 1:1 resolution. The Line of Fire is not abstracted, so when the incoming round hits a particular part of the tank it goes right to the non-abstract system to determine the hit's results. There is no approximation where landed since that is already determined by the shell's non-abstracted flight path. Therefore the actual 3D representation is not only more exact than CMx1, but it enables the combat system to be more exact too.

If you think of soldiers, it's less abstract than CMx1 by far from a visual sense. Each guy is there in the 3D world, each guy is spotting, each guy is firing. True, the spotting is a bit more abstracted, though only partially so (see comments to Martin below), true we do have to make some allowances for the guys being better dispersed amongst subtle cover. It is also true that these abstractions are still fairly consistent with the 1:1 visuals and far more in synch with the degree of simulation than not. That running guy that makes it around the corner of a building has a chance of spotting something as an individual, not as a Squad. As more guys come around the corner the chance of spotting something increases since there are now more eyeballs looking around. All of that is 1:1 and in synch with the visuals.

What I'm taking from your comments here (and in other threads) is along the lines of: In CM1, what could be done with the terrain was much less--but it allowed for more of a WYSIWYG in relation to the combat engine--e.g. if you appeared to be covered by terrain or structures, you generally were.

Sure. To use Dale's point, a bit, is in CMx1 you had "squares" that were pretty distinguishable and very simplistic. So it was pretty easy to identify what terrain your unit was in because it was "head of a pin" and therefore could only be in one type of terrain at a time. Your entire Squad would be moving in Open and then suddenly, all at once, be on a Road. There was never 1 guy being on the road, 4 guys being pinned in a ditch, and 1 guy left behind wounded.

It seems this is less the case in CMSF--in that individuals (and possibly vehicles, too?) may appear to be either in or out of cover but in certain circumstances may actually not be for purposes of LOS/LOF.

Not exactly. For individuals if you have a Team in amongst trees there is some abstraction that presumes that the guys are making use of the trees for cover. So in that sense you are correct that a guy sitting in what appears to be the open may be treated with a little more protection than he in appears to have. But if the guy is prone, he is prone. If he is running, he is running. If the guy is on a part of an Action Spot that has a road under him, he is on the road and not with the other guys in the trees. So by and large you can trust the visuals to tell you what your guys are doing and where they are doing it.

One clarification... LOF is not abstracted, only LOS is. However, since you don't get one with the other that means that LOF is abstracted in the sense that if LOS abstraction affects something then to some degree it might (I stress might) affect LOF.

Abstraction is necessary, both graphically and in the mechanics, and it's really more or less a question of preferences, but I think I generally prefer a higher level of abstraction in the graphics then in the mechanics (if they can't be equal), as it can become frustrating when things are happening that look like they shouldn't.

It's far more than personal preferences when it comes down to accuracy of the simulation as a whole. There is no question about it... CM:SF is vastly superior to CMx1 when it comes to the mechanics, and it is largely because of 1:1 graphical representation.

Once some of the LOS/LOF issues get ironed out in a patch, it may turn out the mismatch isn't as great as it seems.

There are a couple of highly unfortunate bugs that are definitely making this discussion more confusing than it should be. We will get these issues fixed. However, remember that there are a lot of graphical abstractions in CMx1 that bothered people until they accepted them as necessarily artifacts of an otherwise realistic simulation. My example of a tank getting hit as it moves behind a house is the best one I can think of. Unrealistic visual result, unrealistic simulated result. But people weren't asking for us to go back to 2D top down graphics because of things like that Instead they asked that we refine the game so the visuals and the simulation were better in synch with each other and overall more realistic. That is what we have now for sure, though it is imperfect.

Martin,

Steve, one last thing to clarify - is there something more abstracted (less detailed) in CMx2 then in CMx1 ?

None that I can think of, except the point I made about LOS in some ways being more abstracted. What I mean by that is you draw LOS between Action Spots, which are 8x8m, so that is less precise than CMx1's system where you traced LOS from a specific point to another specific point. However, in CMx1 LOS was only abstractly blocked by things like trees, buildings, walls, etc. because those things were all abstractions in their own right. So in CMx2 it really does matter if your tank driver is 0.5m beyond the corner of a building because now he has LOS where in CMx1 he wouldn't because he wasn't simulated. The center of the vehicle had to pass beyond the building before LOS could be traced.

This is actually the reason for the LOS bugs you guys are experiencing. The LOS system is inherently far more detailed than the CMx1 system even if it is going from Action Spot to Action Spot instead of specific point to specific point.

But other than that... I can't think of anything that is simulated in less detail than CMx1. Which is why I get a chuckle when some people call CM:SF a RTS clickfest. If that is what CM:SF is, then CMx1 must be... I dunno, Pong with fancier looking paddles?

Steve

[metalbrew]

What factors affect how units aim at an enemy? I mean, if they have the whole tank profile to choose from what will they aim at?

[Battlefront.com]

They aim for center mass of what can be seen. However, this is difficult to see since most things fired don't go exactly where they are aimed. Heck, some don't even go close to where they are aimed

Steve

[Battlefront.com]

You know... there is actually one thing that can explain this screenshot that isn't a bug at all:

I don't know if this is the case here or not, but there is a setting that the Scenario designer can use that gives one or both sides some degree of "pre battle intelligence". This was a long requested feature for CMx1 that we never got around to implementing. Basically, what it does is "spot" some of the enemy units at the beginning of the game regardless of LOS. This simulates satellite, human, scout, etc. types of intelligence gathered prior to the benefitting forces going into battle. It's a great feature, but it is up to the scenario designer to alert the player to this in the Briefing.

Now I see this is a possible point of confusion since if you are NOT told this bit of info then you could only possibly conclude that seeing units like this is the result of a bug or design flaw. I'm not saying that this explains the above screenshot (it could be a bug), but it could very well be that it is the result of a setting. Easiest way to confirm this is to check in the Editor, but it's time for me to log off so I can't do that right now.

Steve

[jomni]

Steve,

Knowing about it's position is ok. But being able to kill it by shooting through 4 buildings? Take note the Technical is dismounted / abandoned(infantry icon) already and the tracers that go through.

[dalem]

Originally posted by Truppenfuhrung:

</font><blockquote>quote:</font><hr />Originally posted by dalem:

Look sweetie, not all open discussion is "whining", hmm?

-dale

And blaming the abstraction when abstraction is not involve is a useless open discussion. </font>

[Barleyman]

Originally posted by Battlefront.com:

So in CMx2 it really does matter if your tank driver is 0.5m beyond the corner of a building because now he has LOS where in CMx1 he wouldn't because he wasn't simulated. The center of the vehicle had to pass beyond the building before LOS could be traced.

This is actually the reason for the LOS bugs you guys are experiencing. The LOS system is inherently far more detailed than the CMx1 system even if it is going from Action Spot to Action Spot instead of specific point to specific point.

Wait a second.. First you say LOS is tracked on a 8x8 grid, fine. But then you go around and say it matters if you can see 0.5m of tank from beyond a house corner.

Wouldn't that be more that the front of the vehicle is in fact 0.5m inside an 8x8 spotting hex that the bad guy has LOS on but rest of the vehicle is inside another 8x8 spotting hex that is NOT visible?

</font><blockquote>code:</font><hr /><pre style="font-size:x-small; font-family: monospace;"> A B C D E

1 | |*| |

----------

2 | |h|h|

----------

3 | |h|h|

----------

4#| | | |

* good guy

# bad guy

h house

</pre>

[KNac]

Originally posted by Battlefront.com:

They aim for center mass of what can be seen. However, this is difficult to see since most things fired don't go exactly where they are aimed. Heck, some don't even go close to where they are aimed

Steve

shouldn't they aim the softest spot seen?

[Hertston]

It's terrain, rather than walls, that's really becoming the major issue for me. The 'plan' here ('Meeting at High Altitude') was to move the squad up to just short of the top of the ridge, the Syrians being on the other side, obviously. That's where, visually, they went - but ended up in a firefight through the hill with the results you can see.

[JonS]

Originally posted by KNac:

shouldn't they aim the softest spot seen?

Not if they want to hit. All training - that I'm aware of - emphasises COM aiming.

[thewood]

Hereston, that is exactly my complaint right now. Is this a bug? How can you ever do a hull down if the crest is see through to the AI? Action ponts, Smaction points, whats the point of all this visual granularity in a squad, if the terrain doesn't match the level of granularity.

[panzermartin]

Well, that "feature" is even described in the manual..I'm afraid we have to live with it till CMx3

[Truppenfuhrung]

Originally posted by Hertston:

It's terrain, rather than walls, that's really becoming the major issue for me. The 'plan' here ('Meeting at High Altitude') was to move the squad up to just short of the top of the ridge, the Syrians being on the other side, obviously. That's where, visually, they went - but ended up in a firefight through the hill with the results you can see.

I hope it's a bug otherwise...

[SlowMotion]

I don't know if this "shoot through hills" is the reason, but sometimes when I try to use Target command, the game says there's no line of sight. Still the game may shoot at this target on his own, but not because I could tell him so.

[Guest Guest]

deleted per user request

[Battlefront.com]

jomni ,

Knowing about it's position is ok. But being able to kill it by shooting through 4 buildings?

heh... obviously not! If you have that save game please email it to me and I'll take a look into it. I've not seen something like that before.

Guys, one of the difficult things about LOS/LOF (from a code standpoint) is that there are so many different combos of terrain, elevation, unit placement etc. These all have to be coded for in fairly rough ways and then refined when oddities pop up. Then again there is a lot more room for data errors. For example, we had one where a particular type of balcony was blocking LOS that shouldn't have been. Finding problems like this require rather specific battlefield conditions, starting with the inclusion of that type of balcony in the first place. So it will probably be a while before some of these oddities are found. This is the difference between having a couple dozen people pounding on the game compared to thousands of people pounding on it. Statistically the chance of finding these errors goes way up. They they have to be noted and properly communicated to us to fix them.

Barleyman,

Wait a second.. First you say LOS is tracked on a 8x8 grid, fine. But then you go around and say it matters if you can see 0.5m of tank from beyond a house corner.

I'm sure the confusion is my fault. This is a difficult thing to explain, especially for someone who has been using it for years and not using CMx1. It is also difficult because I didn't code the thing and some of the very fine points aren't known to me unless I check with Charles first. Since he's busy I'm going to try and wing an answer to your question.

As I understand it LOS is drawn from the location of the unit spotting to the Action Spot of the unit that it is trying to target. So if the Bradley is mostly behind a building it doesn't get to spot something that is around the corner. However, LOF is exact so the enemy can "see" that portion of the Bradley that is poking out and take a shot at it even though the Bradley doesn't have LOS.

KNac

shouldn't they aim the softest spot seen?

Yeah... I was thinking of that specific screenshot of the Bradley and RPG when I said that. I'll get back to you about what a more precision weapon, like a tank's main gun, aims at. However, as JonS said it is probably center mass. When given a chance between having a better chance of hitting and a smaller chance of hitting the right spot, always go with the chance of hitting. Same is true for small arms training. Sure, a head shot will kill a man, or take him out of action, with a single hit. However, the chances of hitting a head is far less than hitting the torso. Since a round to the torso is likely to at least take the man out of action, the torso is the better target to aim for.

Truppenfuhrung

It's terrain, rather than walls, that's really becoming the major issue for me. The 'plan' here ('Meeting at High Altitude') was to move the squad up to just short of the top of the ridge, the Syrians being on the other side, obviously. That's where, visually, they went - but ended up in a firefight through the hill with the results you can see.

Yup, there is something to be fixed there for sure. There was something about this earlier in this thread or elsewhere. While a LITTLE bit of this is expected when there is a slight ridge in the middle of an Action Spot (say 2 guys shooting through the terrain and 7 not), the screenshot you have up is clearly not such an event. Charles is looking into such things now.

AdamL,

Does the LOS/LOF mechanic you describe above (thanks btw) account for why sometimes I cannot target an enemy unit but can target the ground underneath him?

As for gameplay, by doing so I can still kill him. Since that is the case, it would be nice to add some kind of a check where if the unit can target the ground underneath (LOF) then it should automatically grant LOS to the enemy unit on that terrain. Or you can perhaps there is some better idea.

I'm a bit confused here about the situation you are describing. Are you talking about an individual soldier or something like a whole vehicle?

Steve

[metalbrew]

Originally posted by Battlefront.com:

So if the Bradley is mostly behind a building it doesn't get to spot something that is around the corner. However, LOF is exact so the enemy can "see" that portion of the Bradley that is poking out and take a shot at it even though the Bradley doesn't have LOS.

Going with the Bradley example, does the location of the imager, vision block, binocs, etc... matter? I mean, is the "Bradley" just a single point in space or do the locations of the sensors matter?

In the Bradley vs. RPG situation, the imager was above the roof of the building and the turret could have rotated and fired at the RPG team (the RPG team was on the roof of a 2 story building) but didn't. Judging by dead reckoning the Brad should've been able to fire at the RPG team. It got me wondering about whether the location of the vehicle's eyes matters in the equation.

Other angles

[Truppenfuhrung]

Originally posted by Battlefront.com:

Yup, there is something to be fixed there for sure. There was something about this earlier in this thread or elsewhere. While a LITTLE bit of this is expected when there is a slight ridge in the middle of an Action Spot (say 2 guys shooting through the terrain and 7 not), the screenshot you have up is clearly not such an event. Charles is looking into such things now.

Thanks for your answer !

I little bit of "shooting through the terrain" of a slight ridge will be fine. I think that we can all accept that. It's not the kind of thing that will ruin a strategy.

[Thomm]

Originally posted by Truppenfuhrung:

It's not the kind of thing that will ruin a strategy.

Yesterday, I played the campaign scenario where you have to hunt down fleeing SF while sustaining only 10% losses of your own (my punishment for not getting them all in the scenario before).

In that case, where I wanted to go over "Guitar Hill" it was almost impossible to determine at what point to switch from 'Move' to 'Slow' to delay detection. Lots of good guys died on that hill!

And I think I moved dangerously close to the 10% mark as a result ...

Best regards,

Thomm