Bizarre Spotting behaviour is ruining the experience for us

[ChappyCanuck]

Hey guys, I would like to share two experiences that a buddy (retired armour officer) and I (retired infantry officer) observed while playing the awesome battle, "Huzzar!", and both occurred on consecutive turns. We thoroughly enjoy applying the tactics we learned through our careers to this game, so perhaps the occurrences described below may seem more bewildering to us.

1. My three panzergrenadiers were in ambush behind bocage near a gap. A Sherman drove slowly through the gap and they engaged it with 2 grenades. The Sherman stopped on our side of the gap in view (8 meters away). The dust cloud then caught up to the Sherman and obscured it for about 20 seconds - totally realistic. Once the cloud dissipated, the Sherman "reappeared", again 8 meters away, same spot. My "vigilant" soldiers, still on one knee and ready to pounce with their remaining grenades, did nothing. They stared at it. And stared at it...and when the turn came to an end the leader put his binoculars to his eyes. "Hey guys, do you see that tank?" "No sir." "Well then let me have a look with my binoculars". OMG we laughed, it was so ridiculous. End result - all 3 were gunned down next turn. Sure this could be a "lack of will to engage" problem instead of a spotting one (although they were keen on grenading the tank 20 seconds earlier), but putting the binos to the leader's head indicated to us that they couldn't see the tank.

2. My panther, on one side of a bocage line, sound spotted an approaching Sherman on the other side of the bocage line, travelling down the road. My panther was near a gap in the bocage. The crew was opened up, armour arc sighted through the gap, and non-moving. Good bye Sherman, right? Well the high silhouetted, noisy Sherman that was moving across the gap, stopped (invisible to my waiting crew) engaged, and destroyed my panther. WTF?

There have been other such occurrences in other battles (my buddy has experienced them also) that just don't make sense to us. And I am sure you folks are having similar issues.

We thoroughly enjoy CMx2 for its graphic and certain play-ability improvements (arty system for one), but these spotting irregularities are far too common. In all of the years I have played CMx1, I don't remember these kinds of things happening. And the result is that we are not enjoying CMx2 as much as we should - and its downright frustrating.

I am sure you guys have experienced similar things and I may be beating a dead horse here, but why can't spotting be improved to more closely resemble reality? We are not techies in any way, and maybe itis a technical problem, but can't it be solved? And please, no "fog of war" rebuttals. We get that.

Cheers!

[womble]

Once the cloud dissipated, the Sherman "reappeared", again 8 meters away, same spot. My "vigilant" soldiers, still on one knee and ready to pounce with their remaining grenades, did nothing. They stared at it. And stared at it...and when the turn came to an end the leader put his binoculars to his eyes...Sure this could be a "lack of will to engage" problem instead of a spotting one (although they were keen on grenading the tank 20 seconds earlier), but putting the binos to the leader's head indicated to us that they couldn't see the tank.

The "raising his bins" is a default animation for the "Spotting" action for the pTruppe with the binos. It, by itself, is only indicative of the action the pTruppe is taking. It's entirely clear-cut as to whether the team in question could see the tank: if you selected them did it show up as spotted? If it did, then them not close assaulting it is nothing whatsoever to do with spotting, and you're blaming entirely the wrong mechanism. If it didn't, then there may be some wonkiness going on.

2. My panther, on one side of a bocage line, sound spotted an approaching Sherman on the other side of the bocage line, travelling down the road. My panther was near a gap in the bocage. The crew was opened up, armour arc sighted through the gap, and non-moving. Good bye Sherman, right? Well the high silhouetted, noisy Sherman that was moving across the gap, stopped (invisible to my waiting crew) engaged, and destroyed my panther. WTF?

Are you certain there were no trees involved? Was the turret slewed to create the armour arc, or was the Panther glacis-on to the threat? I ask these two because if there was a tree in the bocage blinding the Panther's turret, and the driver/bow gunner weren't pointing at the road, it's possible the germans couldn't see the Sherman, whereas the Sherman could see the side of the hull of the Panther. I have often been stymied by trees in Bocage even in "Show trunks nearby" mode, because there are lots of trees placed in Bocage whose trunks don't stick out the top, and aren't visible in that mode.

I'm only reaching for these explanations, because every time something like this has happened to me there's actually been an at least plausible reason (except for the snap shots through tree canopies that seem to have died out in later versions, although that might be because of the types of scenarios I've been playing).

...CMx1, I don't remember these kinds of things happening. And the result is that we are not enjoying CMx2 as much as we should - and its downright frustrating.

It's pretty much not worth comparing what happened in CMx1 to CMx2. The technical demands of x2 are orders of magnitude greater than those in x1. And yet, for every time there's a spotting FUBAR (and they do happen, I'm sure, in that what happens isn't what you expected to happen, and the effect was negative), think of the number of times your units don't "Borg-spot". The latter outweigh the former by several orders of magnitude, since that's the default behaviour and happens all the time, every turn, for every unit. If you're a glass half empty kinda guy, you'll see the time the mechanisms produce a result you don't like and focus on those, but if you're a glass half full, or an engineer, you'll appreciate the times it "works". And still, in most of the "bad" occurrences, there will be an explanation, other than "it sucks". LOS rules are absolutely reciprocal, as Steve has taken great pains to point out, and the computer umpire that's resolving your Spotting chances hasn't got a bias, it just uses the rules.

...why can't spotting be improved to more closely resemble reality?

It's technical. Try this thread: http://www.battlefront.com/community/showthread.php?t=112956 for some pretty detailed treatment of the technical challenges in figuring LOS. Then add the difficulty of "precisely" simulating foliage and shade and all the rest of the things that affect spotting, and you'll see that what we have is pretty good. Sure, it could be better: the spotting cycles could be shorter; the LOS map could be less granular, but these all put more and more load on the CPU, and it's an logarithmic progression, not a linear one.

It's resource based. There might be more efficient algorithms for some of these operations and calculations, but BFC has a coding staff of 2 these days (which is double what it used to be), and the actual feature list is incomplete, so improving that which we have which is core and already works pretty well most (i.e. 99.999%) of the time I imagine has a lower priority, if only because any changes to it are fundamental, and would need exhaustive testing by qualified volunteers.

[A Canadian Cat]

1. My three panzergrenadiers were in ambush behind bocage near a gap. <snip> Sure this could be a "lack of will to engage" problem instead of a spotting one (although they were keen on grenading the tank 20 seconds earlier), but putting the binos to the leader's head indicated to us that they couldn't see the tank.

So putting the binos up to look through is not an indication of anything other than what animation was running at that moment. If you select the infantry is the Sherman visible? This kind of encounter usually goes really badly for Shermans but hey not always.

<snip>My panther was near a gap in the bocage. The crew was opened up, armour arc sighted through the gap, and non-moving. Good bye Sherman, right? <snip> WTF?

Why The Face? I got nothing - but sometimes spotting does not go like you expect it in real life either. If you think there is a bug may I suggest reproducing it and running the situation a bunch of times to see what happens.

And please, no "fog of war" rebuttals. We get that.

Yeah, me to but honestly that is very often *the* answer to questions like this. Stuff happens and sometimes it sucks for you.

[Erwin]

ChappyCanuck: Regular players make posts like yours quite often. And we get shouted down regularly with the same excuses for the game engine as above. And that's understandable if we are not military professionals and don't know how things are in RL

However, one would have hoped that when military officers make a post like this, they DO know what RL is like and their opinions would get more respect.

[A Canadian Cat]

ChappyCanuck: Regular players make posts like yours quite often. And we get shouted down regularly with the same excuses for the game engine as above. And that's understandable if we are not military professionals and don't know how things are in RL

However, one would have hoped that when military officers make a post like this, they DO know what RL is like and their opinions would get more respect.

Really, really! So, you think that one post by an expert in real life should what, cause all of BFC staff to drop everything and scramble around looking for defects? I am sorry but what I wrote was certainly not meant as shouting down and @womble's post was probably even more helpful than mine. Both posts seem pretty respectful to me. I certainly meant no disrespect.

The fact is that nothing is going to get treated as a defect without clearly answering lots of questions and showing examples, saved games and tests. My suggestion is loose the chip and help the guy figure out what went side ways - or if it is just a case of bad stuff happens.

[Erwin]

But, it's not "one post", Ian. Many, many longtime players have made many similar posts over the years.

It's one thing when you have mere gamers making the observation. It's another when military professionals who presumably know what they are talking about re RL, make exactly the same observation.

[akd]

The spotting system is too complex and has too much inbuilt variability to draw conclusions based limited samples or individual circumstances. The system will necessarily result in outlier outcomes. These will seem more egregious when encountered because we don't keep a running tally of all the times spotting works out how we would expect, and because we tend to imbue the TacAI with an ability to anticipate the future based on the wealth of information available to us as the Player-God. The TacAI does not anticipate. It simply has chances to spot. The only 0% - 100% situation where a spotting "duel" outcome can be predicted with certainty are situations where there is not reciprocal LOS. These are extremely limited, e.g. a buttoned AFV that has no vision to its rear quarter.

[womble]

Oh Erwin. Real life experience don't enter into it. Having experience of how the game's modelling of real life can be misinterpreted to make you think it isn't modelling real life very well is much more important in this case. And if you think "explanation" is the same as "excuse", you're mistaken. Take the recent post about the Sherman and the PzIV platoon. That highlights a bunch of misunderstandings and misperceptions about the game and what the player thinks they're seeing, and how those clash with the way the game really works, which, in that case, seems to be "plausibly". Without save games, there's no way to know if something is actually wrong or whether honest mistakes are being made in observation or interpretation. In the case above, I'm inclined to believe that there are both. One of the reasons people "like me" post in defense of BFC in situations "like this" is that we don't see these "inexplicable" events very often, if at all. Perhaps because we can discover explanations, whether that's because we have stumbled across similar situations in the past and had the factors affecting them explained, or because we're more inclined to dig deeper into the reasons for the apparent FUBAR, and discover that it's actually just a SNAFU, or because we're prepared to suspend disbelief just a little bit more than others to accommodate an imperfect engine in an imperfect world.

Oh, and if you think Steve's defense of the LOS map mechanism is an excuse, how long do you want to wait before your machine can handle real-time determination of LOS on the fly?

In this particular case I've offered potential explanations for the observed behaviour, similar to explanations that I've seen pan out as accurate in the past. To date, we haven't enough information to know that these explanations don't cover the situation.

We get more posts that end up with, "Ach. Didn't have Show Smoke on. That explains it," (or similar) than we do ones that highlight real issues with LOS/spotting (other than the obscured AS thing) that can't be explained by foliage or height or facing or somesuch.

[ChappyCanuck]

For the record, there is no foliage issue at all for my Panther example...open field, flat as a pancake, engagement range 20-30 meters....the crew knew he was coming by sound and were waiting with fingers on the trigger. Totally focused on a small gap in the bocage line....noisy, high silhouette Sherman drives right up in the gap and brews my Panther....and I still can't see him....I can see no justification for this fault.

[akd]

...open field, flat as a pancake, engagement range 20-30 meters....the crew knew he was coming by sound and were waiting with fingers on the trigger. Totally focused on a small gap in the bocage line....noisy, high silhouette Sherman drives right up in the gap and brews my Panther....and I still can't see him....I can see no justification for this fault.

Yeah, this is what I mean by attributing human predictive powers to the TacAI. The TacAI is not capable of doing any of the things you describe above. Your unit did not know the Sherman was coming and anticipate where the enemy would appear in the future. Your unit was not focused on anything more than the direction of current facing. It is a fault in the sense that the AI is not human, but it is probably not correctable.

The real question is if you reran this scenario many times, would you always get the same outcome?

[sburke]

For the record, there is no foliage issue at all for my Panther example...open field, flat as a pancake, engagement range 20-30 meters....the crew knew he was coming by sound and were waiting with fingers on the trigger. Totally focused on a small gap in the bocage line....noisy, high silhouette Sherman drives right up in the gap and brews my Panther....and I still can't see him....I can see no justification for this fault.

There may be no justification from a RL perspective, but there are possible issues from a computing/gaming perspective. CM does have a cycle for when spotting checks are made etc. and trying to increase those spotting attempts to reduce the few times this occurs would make the game pretty much unplayable except for the most high end machines (and questionable even then) Do you happen to have a save of the turn prior when you issued the commands as well as a save of the turn itself? I'd be willing to take a look if you want to PM me. I'd like both as sometimes as AKD mentioned you get an outlier and in those cases behavior can appear totally wonked. When I hit those I write it off to mud on the vision blocks or bird cr*p or whatever. I don't see it often enough to freak out over, but it can happen.

Despite Erwin's protestations, you aren't getting blown off. There are recognized limitations in the engine and Steve has gone into great detail as to what it costs in processing power to get any more detailed than they are. Unfortunately that is not likely going to change anytime soon. Whether there are some possible other items in play that would make this more plausible, I couldn't say. I have seen it often enough a player overlooked something, it happens to us all.

[ChappyCanuck]

AKD - I understand that. But I was in a somewhat perfect tank ambush position and lost without even firing a shot. And if this had just happened once, ok well maybe something happened, like a bee stung my gunner? lol But there seems to be a rash of very wonky things happening (mainly spotting issues) that have caused me to raise my eyebrows on many occasions since I have purchased this game. And I just don't recall CMx1 having the same spotting issues....yes I know I am not supposed to compare those two systems but one worked fine for us.

Womble - when I am paying hard earned dollars for a game system that is supposed to accurately portray WW2 combat, then yes I am definitely a glass half full guy. As I stated previously, I do enjoy the improvements, but there are key LOS & spotting issues that are driving me nuts (ie siting MGs, AT guns, etc)

No one wants a game system to work more than I do. I thrive on simulations such as this. But I should not be getting frustrated by good tactics going wrong because of some algorithm.

[ChappyCanuck]

sburke - thanks for your post. Again I am not a "techie" so I don't know what draws power and what does not, etc. I have since deleted the turn but if something similar happens again I will definitely send that to you.

We all want our forces on the screen to behave the way we intend (and yes I understand actions under fire to be chaotic), but maybe the system should be simplified somehow to behave the way we intend. If I want my HMG to set up, site in and fire out a window, in RL it will generally happen. But with the game as it is now, I am not confident that I can make that happen - sometimes I can, some times I can't...and I have no idea why I can't and when I can't. And that bothers me

[sburke]

sburke - thanks for your post. Again I am not a "techie" so I don't know what draws power and what does not, etc. I have since deleted the turn but if something similar happens again I will definitely send that to you.

We all want our forces on the screen to behave the way we intend (and yes I understand actions under fire to be chaotic), but maybe the system should be simplified somehow to behave the way we intend. If I want my HMG to set up, site in and fire out a window, in RL it will generally happen. But with the game as it is now, I am not confident that I can make that happen - sometimes I can, some times I can't...and I have no idea why I can't and when I can't. And that bothers me

NP thanks. I'd certainly be interested to look if you have any other occurences. Unfortunately simplifying in this case could only make the situation even worse. CMx2 is never going back to borg spotting (and frankly I don't know that these same things didn't occur there- folks tend to have some very rose tinted glasses when it comes to CMx1). As to Charles abilty to possibly do something else, no idea. The guy is already a genius as far as I am concerned and I am absolutely clueless as to what the possibilities are. The best I can do is look at it, say yeah that is wonked and pass it along.

[Stonecutter]

Hello, I am ChappyCanuck's buddy and am currently the opposing force in our Huzzar! battle While I am far from being an expert in game design or AI algorithms, I am an engineer and also have some experience related to the LOS issue as I was part of team that examined a military war simulation program similar to CMBN, albeit it was over 10 years ago. For our evaluation, we used the program and the associated maps to estimate the intervisibility (i.e. line of sight or LOS) at over 200 locations spread out over ~300 square kilometers in one of our training areas. We then chose 10 different LOS arcs from each location and proceeded to use a Coyote reconnaissance vehicle to measure real life LOS from each of those locations, using the appropriate arc (thank goodness for GPS and laser rangefinder!!). The findings validated the LOS algorithms in the simulation program with the following key observations:

1) Digital Terrain Elevation Data (DTED) - DTED sampling largely determines accuracy. Put more simply, if you sample the elevation of the ground every 25m to form your grid, it will be much more accurate than if you use a 100m grid. In game terms, ground based LOS accuracy can improved by reducing the size of each "hex" or "square". The offset to this is that the complexity increases inversely with the square size (e.g. simulating a 1km x 1km battlefield with 10m x 10m "hexes" = 10,000 hexes to check at each step during a turn while increasing this to 5m x 5m = 40,000 hexes) with a commensurate decrease in game speed. Faster processors can largely offset this but then more processing time is dedicated to determining merely if LOS does exist and not to the issue of whether or not the LOS is detected.

2) Rastor overlays - these represent the terrain features overlay (woods, buildings, hedges, etc.). This is where the realm of stochastic (statistics) decision making enters the game. Depending on how a terrain feature is modeled, this will impact the probability of being able to see through it and hence assessing whether or not LOS exists. Depending on the modelling algorithm and the "roll of the dice", this could even result in Unit A having LOS to Unit B while Unit B does not have LOS to Unit A - likely due to Unit A having "rolled" higher than Unit B. Ideally, there would be only one roll that applied to both units but this again depends on how LOS is determined.

3) Tac AI & Target Arcs - This is perhaps the area where I see the greatest potential for deviations from real life and an area that could benefit from some TLC. Based on numerous examples, the "survival instinct" for units in CMBN appears to be nearly eliminated when the Target Arc or Target Armour in Arc commands are used. In our example of the Sherman and Panther, I strongly suspect that the Sherman was just outside the Arc LOS even though the Sherman was within the general LOS of the Panther. Earlier in the game, one of ChappyCanucks' halftracks drove up to an opened up Sherman just a couple meters outside the covered armour arc. When I checked, the Sherman had LOS the entire time the halftrack approached but did nothing - it merely waited for the halftrack to stop ~10m away and then kill the Sherman Commander with MG fire!

I can accept when LOS doesn't happen when I expect it to based on the stochastic processes and modelling involved. While there is room for improvement here as noted above, there are tradeoffs in terms of processing power and complexity. However, the survival instinct is an area where I think the TacAI could definitely benefit from some tweaking (not twerking, you Miley fans out there!). At this point, my default solution is to only use the arc commands at long ranges where I am not too concerned about the enemy sneaking up on me or I want to ensure that my tanks and ATGs shoot up opposing vehicles rather than wasting ammo on lone scouts that are conducting recce by death.

[A Canadian Cat]

Welcome the forum. Huzzar! was one of the first scenarios I played against a live opponent and it still ranks towards the top of my most fun list.

3) Tac AI & Target Arcs - This is perhaps the area where I see the greatest potential for deviations from real life and an area that could benefit from some TLC. Based on numerous examples, the "survival instinct" for units in CMBN appears to be nearly eliminated when the Target Arc or Target Armour in Arc commands are used. In our example of the Sherman and Panther, I strongly suspect that the Sherman was just outside the Arc LOS even though the Sherman was within the general LOS of the Panther.

That is the kind of additional info that needs to be teased out of yucky situations like this. Target arcs can be like holding a knife by the wrong end - they cut you instead. I have more to say about target arcs and my experience later.

I can accept when LOS doesn't happen when I expect it to based on the stochastic processes and modelling involved. While there is room for improvement here as noted above, there are tradeoffs in terms of processing power and complexity.

There are quite a few long discussions about LOS issues. To learn about how the game works (sady using a game limitation as an example) check this thread out: http://www.battlefront.com/community/showthread.php?t=112956 specifically post #36 where Steve from BFC gets into the discussion. A lot of what he talks about will resonate with the work you described.

Also don't forget that there is more to spotting than just having LOS to the area in question. Someone has to actually see the thing that is in that location. It is a big world out there and we don't always look at the right spot at the right time to see everything. Or one person notices something important a few seconds before another.

However, the survival instinct is an area where I think the TacAI could definitely benefit from some tweaking (not twerking, you Miley fans out there!). At this point, my default solution is to only use the arc commands at long ranges where I am not too concerned about the enemy sneaking up on me or I want to ensure that my tanks and ATGs shoot up opposing vehicles rather than wasting ammo on lone scouts that are conducting recce by death.

The important thing to know about cover arcs is they are not instructing the unit to "only look here" they are instructing the unit to "only fire here". Our instinct tells us that they should give is a spotting advantage to only be scanning in a small area but target arc commands *do not do that*. They are firing arcs only. This is why I no longer use them except for a few special circumstances.

When I will use them:

• To keep scouts from shooting - short 20-40m circular arcs
  
• Keep AT assets from shooting at infantry - again circular only but covering the whole map or out to their effective range
  
• To have units look to their side - similarly 180 deg arcs covering the whole map to one side as a unit is moving making sure it is cleared at the end
  
• Have immobilized vehicles cover some part of the map they are not facing - again 180 deg extended to cover the whole map
  
• And occasionally to setup an ambush - but again only circular cover arcs will be used if I want to prevent firing until the enemy is close
  

The last thing you want is to have the enemy come to a spot just outside the arc.

On the subject of survival instinct you have a point. And it might be even more counter intuitive given that the more experienced a unit is the more likely they will obey your fire restrictions and the less likely they will deviate from your orders and save its own neck. For example I had a Panther once that had a Crack crew. I ordered them to fire on a building that had known enemy inside it. A few moments in to the next turn a Royal Tiger showed up to the Panther's left. Those brave tankers held true to the order I gave them - they new it was vital to take out that building ignoring their own safety for the greater good. Oh man that was soooo not what I really wanted, but it was what I told them to do. They died. Now I really love the Target briefly command. That way they could have got a few rounds on the building target and then been free react to anything that showed up unexpectedly. They might have survived.

[Pak40]

2. My panther, on one side of a bocage line, sound spotted an approaching Sherman on the other side of the bocage line, travelling down the road. My panther was near a gap in the bocage. The crew was opened up, armour arc sighted through the gap, and non-moving. Good bye Sherman, right? Well the high silhouetted, noisy Sherman that was moving across the gap, stopped (invisible to my waiting crew) engaged, and destroyed my panther. WTF?

I'm very curious about this. Please post the replay of this so we can view it. Either Youtube or let us download the replay file (even better)

[Doug Williams]

The important thing to know about cover arcs is they are not instructing the unit to "only look here" they are instructing the unit to "only fire here". Our instinct tells us that they should give is a spotting advantage to only be scanning in a small area but target arc commands *do not do that*. They are firing arcs only.

This is news to me. Bad news.

Thanks.

[Vanir Ausf B]

Your unit did not know the Sherman was coming and anticipate where the enemy would appear in the future. Your unit was not focused on anything more than the direction of current facing. It is a fault in the sense that the AI is not human, but it is probably not correctable.

The real question is if you reran this scenario many times, would you always get the same outcome?

This is true, however a sound contact would presumably make spotting of the Sherman more likely, in the same way that spotting information passed along via C2 does.

[womble]

This is true, however a sound contact would presumably make spotting of the Sherman more likely, in the same way that spotting information passed along via C2 does.

AIUI, that's true, certainly for while the unit is "where the sound contact is". Whether there's a radius of effect and you'd be more likely to spot the tank once it's moved away from the sound contact, I'm not so sure, and we can be fairly certain the TacAI can't decide "there's three '? armour' down that road; probably means a tank's coming into sight about... Now!" like we can.

[akd]

This is true, however a sound contact would presumably make spotting of the Sherman more likely, in the same way that spotting information passed along via C2 does.

At or near the location of the contact, AFAIK. Multiple contacts from one unit however do not provide a vector that the TacAI is able to recognize and react to.

[Vanir Ausf B]

Two things to keep in mind about spotting in CMx2 is that 1) Units do not spot continuously, and 2) There is a very high degree of randomness.

My understanding of the spotting system is that if the LOS map says units in action spot X may have LOS to units in action spot Y then spotting checks are done for every individual soldier or vehicle crew member in both action spots (since LOS must always be reciprocal). This is unlike CMx1 games where units were treated as monolithic for spotting purposes, both in terms of spotting checks and positioning, making it much less CPU intensive.

The interval between spotting checks seems to vary depending on circumstance, but you can definitely run into corner cases where an enemy unit will move into LOS of a friendly unit and who spots who first comes down to which unit is up for a spotting check next. This is particularly noticeable when the range is very short. Increasing the frequency of spotting checks is the obvious solution, hardware requirements permitting.

In my opinion there is a bit too much randomness in spotting. Admittedly, I base this on spotting tests I have done at a much greater range than the examples cited here. However, the odds of every member of an infantry team failing a spot check on a tank at 8 meters should be at or very near 0. Of course, we don't know from the description how long the the Sherman was in LOS after the smoke cleared. If it was just a few seconds then they may not have lived long enough to get a spot check.

[Vanir Ausf B]

I wonder how feasible it would be to spread out over time the spotting checks of individual team/crew members. For example, instead of doing a spot check of all 5 members of a tank crew every 5 seconds (I'm picking a convenient number here), do 1 crew member every second. I am, or course, assuming the game doesn't already do this.

[lofty12]

I had an almost identical experience as the op with the same scenario which is uncanny.

I had a stug (unbuttoned, no cover arc) sitting stationary on a road pointed straight at a gap in some bocage, distance would be 10-15 metres. During the turn I tracked a Sherman move over the field towards to the gap with audio.

It moved into the gap, rotated its turret and took out my stug. The only time it became visible was when it fired by which point it was right in the middle of the gap. It must have taken a good 5 seconds to move dead centre in the gap so im baffled to know how it seen me long enough to take a shot and yet it was invisible to my stug.

It was the most ridiculous and immersive killing moment ive had playing cm. I just cant think of any logical reason how this could've happened in real life.

I know its only a game but this really did show up how poor the los mechanics can be in certain situations.

I still have the pbem turn, I would be interested to hear what a dev or anyone else would make of it after having a look.

[sburke]

I had an almost identical experience as the op with the same scenario which is uncanny.

I had a stug (unbuttoned, no cover arc) sitting stationary on a road pointed straight at a gap in some bocage, distance would be 10-15 metres. During the turn I tracked a Sherman move over the field towards to the gap with audio.

It moved into the gap, rotated its turret and took out my stug. The only time it became visible was when it fired by which point it was right in the middle of the gap. It must have taken a good 5 seconds to move dead centre in the gap so im baffled to know how it seen me long enough to take a shot and yet it was invisible to my stug.

It was the most ridiculous and immersive killing moment ive had playing cm. I just cant think of any logical reason how this could've happened in real life.

I know its only a game but this really did show up how poor the los mechanics can be in certain situations.

I still have the pbem turn, I would be interested to hear what a dev or anyone else would make of it after having a look.

go head and PM me and I'll take a look. Just to be sure, what version are you running - hopefully the latest. (I deleted my older installs )