Roaming instead of hearing

[SlapHappy]

I have a particularly instructive save game file which illustrates the "eyes in the back of the head" capabilities of infantry in a test mission I cooked up. If anyone wants to see it, I could email it....it's about 280 KBs.

[Battlefront.com]

c3k,

Um, where was I? Oh yes, spotting behavior. Thanks for reminding me that there is some ability to spot other than where the individual is facing. However, isn't it true that the odds of seeing something outside of the individual's facing is GREATLY reduced? What I am interpreting this thread to mean is that an improvement in spotting outside of the facing would be an, er, improvement to the game.

I don't think it would be an improvement at all. 9-12 men do not all position themselves in a perfect circle and keep equal attention on all vectors at all times. That's just silly So obviously there is some natural tendency for the soldiers to look in one general direction vs. other directions. For example, the direction they are going to move to next, the direction they are shooting at, the direction enemy fire is coming from, not the direction they are taking cover behind, etc.

Sorry, the way we have it now is at least conceptually correct. Obviously the amount of attention our little pixel truppen spend taking in the scenery might be weighted too heavily towards some sectors and not others, but we think it's about right. Especially considering when units are in combat situations they tend to get tunnel vision. Picture perfect tactics go out the window after 2 days with little sleep, bouncing around in the back of a Bradley in 120 degree temps, and being in the 4th firefight of the morning on an empty stomach. Or at least so I'm told

Aacooper,

So, I suppose Battlefront is right not to have any national modifiers.

Of course we are While it certainly does make a game more interesting to have fictional attributes to create fantasy differences between units, we prefer to stick to a more rigorous view of history. We're not always right since there definitely could be some truth to the old wives tales, but overall I'm sure the game is far more realistic without national modifiers since most of the stuff people expect to see (in that regard) is bunk.

Steve

[c3k]

Steve,

A bit of a strawman, here:

c3k,

I don't think it would be an improvement at all. 9-12 men do not all position themselves in a perfect circle and keep equal attention on all vectors at all times. That's just silly So obviously there is some natural tendency for the soldiers to look in one general direction vs. other directions. For example, the direction they are going to move to next, the direction they are shooting at, the direction enemy fire is coming from, not the direction they are taking cover behind, etc.

There has been NO mention of positioning men in a perfect circle with equal atention on all vectors at all times. If that is how you perceive this debate, you must not be reading it closely enough.

The issue is how to improve in-game behavior so a group of 9-12 men are not totally ignorant of obvious actions occurring around them.

Is it a problem now? Not really. Could it be better? Yep.

So, this all comes down to quite a small tweak.

Ignoring your strawman, who I attack with flying monkeys and burning brooms, here is what I actually wrote:

Steve,

I certainly didn't intend that Civil War anecdote to be construed into ANYTHING related to CM; it was just an interesting tidbit about sound and its importance in warfare.

Back to CMSF: I think there is room for improvement in how a 9 man squad spots things around them. I'd have to run some tests (loaded statement, that!), but, it is my belief that right now all 9 men orient themselves towards the front. As they move, they all face forwards. Emphasis on "all" and "forwards". I know that there is some variability, but it is minor.

Given a possible center of field of view being 1 of 360 degrees, the pie wedge graphic on the base for each man is centered very closely to that of all other men. Especially true when moving. Not so true when a move ends in a building with no facing order. In that case an all-around position is taken up. I do not know if the pie wedge graphic defines that individual's field of view. However, I NEVER see the last man sweeping behind the team.

For the sake of argument, CMSF models all men slaved to a very similar field of view. "Roaming Eyes" would enhance the ability to spot enemy by having flankers and tail-end-charlies looking away from the primary axis of movement. With a FACE command, flankers should skew outwards, splitting the difference such that the the rightmost flank holder has the leftmost part of his field of view parallel to the FACE command. Vice versa for the left side flanker. Some numbers....

FACE aimed to the east, 090 degrees. Assume each man's field of view is 30 degrees. The left most man should center his view at an angle of 075 degrees. That way his scan encompasses 060 through 090. The far right guy should center on 105, scanning 090 to 120 degrees.

Without a FACE command, given a squad moving to the east, 090 degrees, point should face 090, scanning 075 to 105. Overlay scans from there. Last man checks rear.

These would not be fixed areas, but rather what they come back to after scanning around them. Letting their eyes roam...call it "roaming eyes".

"Head on a swivel, men, head on swivel..."

This ignores any kind of sound algorithm. What it does is builds on the strength of CMSF's spotting routine. It would allow small units a more accurate and timely view of their immediate battlespace.

Perhaps some tests are in order...

Thanks,

Ken

The biggest single improvement would be to enable the last man to scan to the rear of the team. Oh, that's up there in that quote. In an earlier post I stated that this "roaming eyes" behavior would drop when there is a definite FACE command, TARGET command, or if the unit is recieving enemy fire.

As you wrote,

Sorry, the way we have it now is at least conceptually correct. Obviously the amount of attention our little pixel truppen spend taking in the scenery might be weighted too heavily towards some sectors and not others, but we think it's about right. Especially considering when units are in combat situations they tend to get tunnel vision. Picture perfect tactics go out the window after 2 days with little sleep, bouncing around in the back of a Bradley in 120 degree temps, and being in the 4th firefight of the morning on an empty stomach. Or at least so I'm told

Steve

I agree: conceptually it is correct. I do think the weighting is off a bit, not a lot, just a bit. If I have 9 men, SOMEONE will be watching the rear and the flanks.

Regarding the "picture perfect tactics" going out the window: you describe men who are tired. Isn't that a toggle in the scenario editor? Wouldn't it be useful to have vigilance decline as a unit gets worn out? Embedded in that paragraph is the assumption that all units act at the same degraded level.

Maybe that's the difference: when I have what I think are fresh, fit troops, instead I have tired, hungry, fought out troops?

Thanks,

Ken

[Battlefront.com]

No strawman at all... I simply said the image of super-duper all around awareness is not something that happens in real life. If you think the weighting is off, do some tests and make a case that there's something wrong. Just saying "I think it's off" doesn't get us anywhere because a) you could be wrong and what is "off" and how do we adjust the variables to compensate?

Steve

[c3k]

Steve,

As to a), that is probable. And would thereby be problematic. That's why you MAKE computer games and I PLAY them!

I'll run some tests... Time is at a premium, but I'll get to it. (SlapHappy, how about some tests, as well?)

The condition I'll be testing is unit awareness of enemy vehicles/infantry around them. The variables will be the enemy units, the terrain the friendly is in, the orders, if any, the friendly is under (specifically TARGET LIGHT and FACE vice no order), enemy movement level, LOS to enemy unit (tall walls, vegetation, etc.)

Fair enough....

Meantime, I've got to extricate one of my Marine platoons from a nasty little ambush I let them wander into. (Ooops, sorry men. The sergeant was right; it DOES look like a bad place to go.)

Thanks,

Ken

[Battlefront.com]

c3k,

As to a), that is probable. And would thereby be problematic. That's why you MAKE computer games and I PLAY them!

Yup, but I'm not the one who thinks there's a problem, therefore I don't need to do any testing If I had to drop everything and start running tests every time someone thought "this might not be quite right" we'd never get a game out of Beta Testing.

As you've already shown, there are a number of variables at work. Which is why it's not all that useful to see someone say "situational awareness doesn't seem good enough". Chances are if we made blind adjustments we'd see the same statements once the changes were patched in. Therefore, we must isolate and quantify problems before changing anything. Otherwise we run a huge risk of breaking 10 things to fix 1, or perhaps breaking 10 things to fix nothing

Personally, if time is at a premium I'd skip this one. The game's been out for ages and generally this is not a source of complaint. When it is I bet you it's because of the limitations of how frequently spotting checks are performed. Certainly that was the main cause of similar issues in CMx1 and during CM:SF testing. Usually a second or two not spotting something due to being inbetween spotting cycles doesn't matter, but when it does it really does!

Steve

[SlapHappy]

Uhhh...I did and...uhhhh...Steve is right . Thanks for taking my back on this one, though, Ken. If anything, a squads ability to react to threats within their 360 degree LOS potential is perhaps more potent than it ought be. As I said, I have a save game that illustrates a particular game turn where three Syrians react to a lone soldier coming up on their five o'clock from outside of a building while they are facing opposite inside the building. But even this particular sample is not as convincing as some of the situations I ran within the entire test itself.

[SlapHappy]

Here we see that as soon as the first Blue trooper HUNTS around the corner, the Syrians in the building 50+ meters away facing in the opposite direction detect them via LOS:

http://farm3.static.flickr.com/2323/3529851910_63dd9f25fa_o.jpg

And in this frame, they have moved to the opposite side of the building and are exchanging fire with the BLUE squad:

http://farm4.static.flickr.com/3571/3529851958_5952f1e9ab_o.jpg

Can they see me?

http://farm4.static.flickr.com/3187/3529852066_b3f934e4c8_o.jpg

Yes they can!

http://farm4.static.flickr.com/3590/3529039465_8bd01664ac_o.jpg

[Battlefront.com]

Thanks SlapHappy! I saw your previous post but that, for some reason, c3k didn't find it to be adequate.

Anyway, what I said in my previous posts still stands. First, units DO have awareness around them other than the direction they are facing. Second, there are a ton of variables which are at play as to how effective that awareness is. Third, unless someone can show a very specific case of "this is too much/too little" then I don't think we should monkey with stuff based on perception alone. Call me crazy if you like, but I don't think it is in anybody's best interests to go randomly changing variables without any sense of which ones or how much they should be changed. Not likely to go well

Steve

[c3k]

Very well!! :)

SlapHappy, thanks for posting.... I missed your earlier post. (Perhaps if it were more adequate, I would've read it! ) I must've been thinking about mermaids; slippery, wet, mermaids who wrestle... Where was I? Ah, yes.

Steve, thanks for, um, being Steve. I appreciate your openness, candor, and desire to improve what should be. And to leave well enough alone.

I won't run any tests. But, if I DID run a test, using a tall wall with a Syrian tank on one side and a US squad on the other, would the US squad detect the Syrian tank? What if it moved?

Thanks,

Ken

[SlapHappy]

My guess is no, but in my case, I will of course have to find out!

[c3k]

Alright, just tested.

I placed a US Stryker squad and its HQ inside a tall wall compound. The compound was about 100mx80m. I placed a Republican Guard Tank Company on the other side of the wall; T-72M1V (2001)'s. 10 of them.

Neither side knew about the other.

Next, I drove one tank in donuts around the compound. Range to HQ team was 30 meters. FAST command.

The US never knew about the tanks.

So, there is NO sound spotting; but we were told that, right?

This opens up a field of research; CMSF smoke and exhaust forensic research. The tank was unknown to the US player, but the dust and exhaust plumes were visible. The location, thickness, and color of both exhaust and dust and the location of the exhaust relative to the dust plume all give clues as to what vehicle produced them. Perhaps someone wants to run tests and create a downloadable pdf with graphics for each vehicle? That way every player can tell what they're facing based on the dust/exaust. (Cool game that allows this type of detail).

The REAL question is whether this behavior should be tweaked for Normandy. If I cannot see an enemy, but I HEAR them, would I try to take a peek? Sure...

In my compound test, perhaps a boost to look over the top might've been nice. (Of course, that would've led to an eliminated squad.)

In bocage, a peek over the top, through the vegetation would be cool to have.

In a city/town, running to the corner for quick glance would be cool to have.

Regardless, _I_, the player, knew there was some sort of enemy vehicle around the compound, so I was free to order my units to investigate further. In this case the dust gave me the clue which was missing from the lack of sound.

I would love to see sound clues included in the next game.

Thanks,

Ken

[Chops]

Ken,

I ran a similar test. I designed a small map with a tall brick wall running east-west across the full length of the map. On the north side of the wall was a dirt road running parallel to the wall, and on the south side of the wall, was a U.S. veteran, rested, 5-man Scout team with a +2 Leader facing to the east, stationary. I setup an AI plan for a Syrian T-72, and had it drive along the dirt road on the north side of the wall, including stopping adjacent to the Scout Team for several minutes. A Question Mark Contact Icon was never produced by the T-72. However, as you noted, my illustrious Scout Team did see the T-72's dust plume. In fact one of the Team members remarked "take a look at that". The Team could hear birds chirpping, but no T-72.

Seems that a dust plume should at least produce a Question Mark Icon.

[Battlefront.com]

Dust/Smoke is not "spottable" (nor is there any practical way to make it that way), so there's no way to tie it to the spotting information.

The test that c3k did is artificial, of course, but definitely shows what I said earlier about there being some circumstances where should absolutely should trigger spotting of some sort. But really... how frequently do you guys experience a situation where you have enemy vehicles that close without any spotting information and no clue that there are enemy vehicle somewhere around in that general area? So to me this is one of those situations which definitely isn't modeled correctly, but it doesn't appear to matter much because it doesn't have much practical impact on the game. There's tons of stuff like this in all the CM games if you look closely enough

Having said that, I really would like to get some sort of audio spotting for up close vehicles. Foot units... I'm still not convinced that would be a good idea.

Steve

[costard]

Command radius seems to have some sort of sound modeling involved - or should have: a unit in voice contact with its commander isn't necessarily in LOS. Also, sound modeling is important for night / stealth missions: an infantry unit approaching a stationary (but not hidden) vehicle at night should be given some indication of approximate distance and angle to target. A unit waiting in ambush ought have the opportunity to effect surprise with good noise discipline.

Effective surprise often relies on the use of stealth: I think BF is throwing away a chance at giving the game an aspect of realism that lends itself to a great immersive experience. That said, I'm sure its not easy to do. I thought CMx1 did a great job - to the extent of modeling a shift in the sound contact proportional to wind strength.

[LRC]

... how frequently do you guys experience a situation where you have enemy vehicles that close without any spotting information and no clue that there are enemy vehicle somewhere around in that general area?

Steve

In MOUT, short distance noice spotting is important (critical ?).

I had several (nasty) occurence of squads bumping into an armor "hiding behind a corner". This is quite unrealistic because in operations engine are always on.